D1.1 Agriculture Pilot Definition · D1.1 – Agriculture Pilot Definition Dissemination level PU -Public Type of Document Report ... H2020 Contract No. 732064 Final – v1.0, 30/6/2017

This document is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 732064. It is the property of the DataBio consortium and shall not be distributed or reproduced without the formal approval of the DataBio Management Committee.

Project Acronym: DataBio

Grant Agreement number: 732064 (H2020-ICT-2016-1 – Innovation Action)

Project Full Title: Data-Driven Bioeconomy

Project Coordinator: INTRASOFT International

DELIVERABLE

D1.1 – Agriculture Pilot Definition

Dissemination level PU -Public

Type of Document Report

Contractual date of delivery M06 – 30/6/2017

Deliverable Leader LESPRO

Status - version, date Final – v1.0, 30/6/2017

WP / Task responsible WP1

Keywords: Agriculture, pilot, big data, modelling, user analysis,

user requirements, stakeholders

D1.1 – Agriculture Pilot Definition H2020 Contract No. 732064 Final – v1.0, 30/6/2017

Dissemination level: PU -Public Page 2

Executive Summary The objective of WP1 Agriculture pilot is to demonstrate how the Big data technologies will

be integrated into the pilots, in order to validate the Big data technologies on practical cases

from agriculture and how it can fulfil the end user communities’ expectations. The Big

technologies will be tested in three areas: arable farming, horticulture and Subsidies an

insurance, where every area will be tested in in pilots with different topics and running in

different countries.

Task 1.1 Co-innovative preparations deal with user understanding specifying the needs of

users and different stakeholders and its main objective is to analyse a set of functional and

non-functional requirements specified from the analysis of the pilot cases. Opportunities for

different solution technologies were reviewed with stakeholders and users are used as an

input and a set of scenarios are described within the bio-economy domain related to the

agriculture sector. Functional requirements are defined and used as input for the application

specification, development and piloting. User and stakeholder study to specify the (most

beneficial) areas of interest from different point-of-views and resulting to detailed scenario

building of the application scenarios from which use cases are defined. This subtask feeds

from user and stakeholder study as input.

The results are the pilot cases definitions including requirements specifications and

evaluation plans.

The organizations that were planned to participate in this task, and their respective planned

work effort in person-months, are Lespro (2), Intrasoft (5), VTT (3), IBM (2), Softeam (2),

Limetri (2), CREA (2), Fraunhofer (6), Vito (6), Tragsa (6), NP (2), Federu (6), CSEM (2), Rikola

(4), Novam (6), EXUS (6), CERTH (6), CITOLIVA (3), GAIA (9), ZETOR (8), CAC (6)

The deliverable D1.1 Agriculture Pilot Definition specifies the pilot case definitions,

requirement specifications, as well as implementation and evaluation plans.



Deliverable Leader: Karel Charvát (LESPRO)

Contributors:

Karel Charvát jr (LESPRO), Šárka Horáková (LESPRO), Savvas

Rogotis (NP), Antonella Cattuci (e-GEOS), Per Gunnar Auran

(SINTEF Fishery), Athanasios Poulakidas (INTRASOFT), Ephrem

Habyarimana (CREA), Pilot leaders

Reviewers: Fabiana Fournier (IBM), Tomas Mildorf (UWB), Caj Södergård

(VTT)

Approved by: Athanasios Poulakidas (INTRASOFT)

Document History

Version Date Contributor(s) Description

0.1 20/04/2017 Initial draft

0.2 18/06/2017 Content transferred to new template

0.3 20/06/2017 Pilot descriptions inserted, ArchiMate

diagrams added

0.4 29/06/2017 Final completing all chapters and

formatting

1.0 30/06/2017 Compliance to submission format and

minor changes.



Table of Contents EXECUTIVE SUMMARY ..................................................................................................................................... 2

TABLE OF CONTENTS ........................................................................................................................................ 4

TABLE OF FIGURES ........................................................................................................................................... 8

LIST OF TABLES ................................................................................................................................................ 9

DEFINITIONS, ACRONYMS AND ABBREVIATIONS ........................................................................................... 11

INTRODUCTION .................................................................................................................................... 13

1.1 PROJECT SUMMARY ..................................................................................................................................... 13 1.2 DOCUMENT SCOPE ...................................................................................................................................... 16 1.3 DOCUMENT STRUCTURE ............................................................................................................................... 16

SUMMARY ............................................................................................................................................ 17

2.1 OVERVIEW ................................................................................................................................................. 17 2.2 PILOT INTRODUCTIONS ................................................................................................................................. 17 2.3 OVERVIEW OF PILOT CASES ............................................................................................................................ 18 2.4 AGRICULTURE DATASETS UTILIZED IN PILOTS ...................................................................................................... 22 2.5 REPRESENTATION OF PILOT CASES ................................................................................................................... 22 2.6 PILOT MODELLING FRAMEWORK ..................................................................................................................... 22

PILOT 1 [A1.1] PRECISION AGRICULTURE IN OLIVES, FRUITS, GRAPES ................................................... 27

3.1 PILOT OVERVIEW ......................................................................................................................................... 27 3.1.1 Pilot introduction .......................................................................................................................... 27 3.1.2 Pilot overview................................................................................................................................ 27

3.2 PILOT CASE DEFINITION ................................................................................................................................. 29 3.2.1 Stakeholder and user stories ......................................................................................................... 32 3.2.2 Motivation and strategy ............................................................................................................... 32

3.3 PILOT MODELLING WITH ARCHIMATE .............................................................................................................. 33 3.3.1 Agriculture pilot A1.1 Motivation view ......................................................................................... 33 3.3.2 Agriculture pilot A1.1 Strategy view ............................................................................................. 33

3.4 PILOT EVALUATION PLAN .............................................................................................................................. 34 3.4.1 High level goals and KPI's ............................................................................................................. 34 3.4.2 Initial roadmap ............................................................................................................................. 35

3.5 BIG DATA ASSETS ......................................................................................................................................... 35

PILOT 2 [A1.2] PRECISION AGRICULTURE IN VEGETABLE SEED CROPS ................................................... 37







4.5 BIG DATA ASSETS ......................................................................................................................................... 43

PILOT 3 [A1.3] PRECISION AGRICULTURE IN VEGETABLES_2 (POTATOES) ............................................. 45





5.5 BIG DATA ASSETS ......................................................................................................................................... 52

PILOT 4 [A2.1] BIG DATA MANAGEMENT IN GREENHOUSE ECO-SYSTEM .............................................. 54





6.5 BIG DATA ASSETS ......................................................................................................................................... 64

PILOT 5 [B1.1] CEREALS AND BIOMASS CROP ....................................................................................... 65



7.3 PILOT MODELLING WITH ARCHIMATE .............................................................................................................. 69 7.3.1 Agriculture pilot B1.1 motivation view ......................................................................................... 69 7.3.2 Agriculture pilot B1.1 strategy view .............................................................................................. 70


7.5 BIG DATA ASSETS ......................................................................................................................................... 73

PILOT 6 [B1.2] CEREALS AND BIOMASS CROP_2 ................................................................................... 74


8.2 PILOT CASE DEFINITION ................................................................................................................................. 76



8.2.1 Stakeholder and user stories ......................................................................................................... 78 8.2.2 Motivation and strategy ............................................................................................................... 79

8.3 PILOT MODELLING WITH ARCHIMATE .............................................................................................................. 80 8.3.1 Agriculture pilot B1.2 Motivation view ......................................................................................... 80 8.3.2 Agriculture pilot B1.2 Strategy view ............................................................................................. 81


8.5 BIG DATA ASSETS ......................................................................................................................................... 83

PILOT 7 [B1.3] CEREAL AND BIOMASS CROPS_3.................................................................................... 84



9.3 PILOT MODELLING WITH ARCHIMATE .............................................................................................................. 92 9.3.1 Agriculture pilot B1.3 Motivation view ......................................................................................... 92 9.3.2 Agriculture pilot B1.3 Strategy view ............................................................................................. 93


9.5 BIG DATA ASSETS ......................................................................................................................................... 95

PILOT 8 [B1.4] CEREALS AND BIOMASS CROPS_4 .................................................................................. 97

10.1 PILOT OVERVIEW .................................................................................................................................... 97 10.1.1 Pilot introduction...................................................................................................................... 97 10.1.2 Pilot overview ........................................................................................................................... 97

10.2 PILOT CASE DEFINITION ............................................................................................................................ 98 10.2.1 Stakeholder and user stories .................................................................................................. 100 10.2.2 Motivation and strategy ........................................................................................................ 101

10.3 PILOT MODELLING WITH ARCHIMATE ....................................................................................................... 102 10.3.1 Agriculture pilot B1.4 Motivation view .................................................................................. 102 10.3.2 Agriculture pilot B1.4 Strategy view ....................................................................................... 103

10.4 PILOT EVALUATION PLAN ....................................................................................................................... 103 10.4.1 High level goals and KPI's ....................................................................................................... 103 10.4.2 Initial roadmap ....................................................................................................................... 103

10.5 BIG DATA ASSETS .................................................................................................................................. 104

PILOT 9 [B2.1] MACHINERY MANAGEMENT ........................................................................................ 105

11.1 PILOT OVERVIEW .................................................................................................................................. 105 11.1.1 Pilot introduction.................................................................................................................... 105 11.1.2 Pilot overview ......................................................................................................................... 105

11.2 PILOT CASE DEFINITION .......................................................................................................................... 107 11.2.1 Stakeholder and user stories .................................................................................................. 110 11.2.2 Motivation and strategy ........................................................................................................ 110

11.3 PILOT MODELLING WITH ARCHIMATE ....................................................................................................... 111 11.3.1 Agriculture pilot B2.1 Motivation view ................................................................................. 111 11.3.2 Agriculture Pilot B2.1 Strategy view ....................................................................................... 112

11.4 PILOT EVALUATION PLAN ....................................................................................................................... 112 11.4.1 High level goals and KPI's ....................................................................................................... 112



11.4.2 Initial roadmap ....................................................................................................................... 113 11.5 BIG DATA ASSETS .................................................................................................................................. 113

PILOT 10 [C1.1] INSURANCE (GREECE)................................................................................................. 114



12.3 PILOT MODELLING WITH ARCHIMATE ....................................................................................................... 119 12.3.1 Agriculture pilot C1.1 Motivation view .................................................................................. 119 12.3.2 Agriculture C1.1 Strategy view ............................................................................................... 120


12.5 BIG DATA ASSETS .................................................................................................................................. 122

PILOT 11 [C1.2] FARM WEATHER INSURANCE ASSESSMENT ............................................................... 123



13.3 PILOT MODELLING WITH ARCHIMATE ....................................................................................................... 128 13.3.1 Agriculture pilot C1.2 Motivation view .................................................................................. 128 13.3.2 Agriculture pilot C1.2 Strategy view ....................................................................................... 130


13.5 BIG DATA ASSETS .................................................................................................................................. 132

PILOT 12 [C2.1] CAP SUPPORT ............................................................................................................ 133





14.5 BIG DATA ASSETS .................................................................................................................................. 143

PILOT 13 [C.2.2] CAP SUPPORT (GREECE) ............................................................................................ 144

15.1.1 Pilot introduction.................................................................................................................... 144 15.1.2 Pilot overview ......................................................................................................................... 144

15.2 PILOT CASE DEFINITION .......................................................................................................................... 146



15.2.1 Stakeholder and user stories .................................................................................................. 148 15.2.2 Motivation and strategy ........................................................................................................ 149



15.5 BIG DATA ASSETS .................................................................................................................................. 153

CONCLUSION ...................................................................................................................................... 154

REFERENCES ....................................................................................................................................... 155

Table of Figures FIGURE 1: ARCHIMATE 3.0 MODELLING FRAMEWORK. ..................................................................................................... 23 FIGURE 2: RELATIONSHIPS OF THE MOTIVATION ELEMENTS ................................................................................................ 26 FIGURE 3: RELATIONSHIPS OF THE STRATEGY ELEMENTS .................................................................................................... 26 FIGURE 4: AGRICULTURE PILOT A1.1 MOTIVATION VIEW .................................................................................................. 33 FIGURE 5: AGRICULTURE PILOT A1.1 STRATEGY VIEW ....................................................................................................... 34 FIGURE 6: AGRICULTURE PILOT A1.1 INITIAL ROADMAP .................................................................................................... 35 FIGURE 7: AGRICULTURE PILOT A1.1 BDVA REFERENCE MODEL ......................................................................................... 36 FIGURE 8: AGRICULTURE PILOT A1.2 MOTIVATION VIEW .................................................................................................. 42 FIGURE 9: AGRICULTURE PILOT A1.2 STRATEGY VIEW ....................................................................................................... 42 FIGURE 10: AGRICULTURE PILOT A1.2 INITIAL ROADMAP .................................................................................................. 43 FIGURE 11: AGRICULTURE PILOT A1.2 BDVA REFERENCE MODEL ....................................................................................... 44 FIGURE 12: AGRICULTURE PILOT A1.3 MOTIVATION VIEW ................................................................................................ 50 FIGURE 13: AGRICULTURE PILOT A1.3 STRATEGY VIEW ..................................................................................................... 51 FIGURE 14: AGRICULTURE PILOT A1.3 INITIAL ROADMAP .................................................................................................. 52 FIGURE 15:AGRICULTURE PILOT A1.3 BDVA REFERENCE MODEL ....................................................................................... 53 FIGURE 16: AGRICULTURE PILOT A2.1 MOTIVATION VIEW ................................................................................................ 61 FIGURE 17: AGRICULTURE PILOT A2.1 STRATEGY VIEW ..................................................................................................... 62 FIGURE 18: AGRICULTURE PILOT A2.1 INITIAL ROADMAP .................................................................................................. 63 FIGURE 19: AGRICULTURE PILOT A2.1 BDVA REFERENCE MODEL ....................................................................................... 64 FIGURE 20: AGRICULTURE PILOT B1.1 TRAGSA MOTIVATION VIEW .................................................................................. 70 FIGURE 21: AGRICULTURE PILOT B1.1 STRATEGY VIEW ..................................................................................................... 71 FIGURE 22: AGRICULTURE PILOT B1.1 INITIAL ROADMAP .................................................................................................. 72 FIGURE 23: AGRICULTURE PILOT B1.1 BDVA REFERENCE MODEL ....................................................................................... 73 FIGURE 24: AGRICULTURE PILOT B1.2 MOTIVATION VIEW ................................................................................................ 80 FIGURE 25: AGRICULTURE PILOT B1.2 STRATEGY VIEW ..................................................................................................... 81 FIGURE 26: AGRICULTURE PILOT B1.2 INITIAL ROADMAP .................................................................................................. 82 FIGURE 27: AGRICULTURE PILOT B1.2 BDVA REFERENCE MODEL ....................................................................................... 83 FIGURE 28: AGRICULTURE PILOT B1.3 MOTIVATION VIEW ................................................................................................ 92 FIGURE 29: AGRICULTURE PILOT B1.3 STRATEGY VIEW ..................................................................................................... 93 FIGURE 30: AGRICULTURE PILOT B1.3 INITIAL ROADMAP .................................................................................................. 94 FIGURE 31: AGRICULTURE PILOT B1.3 BDVA REFERENCE MODEL FOR IOT ........................................................................... 95 FIGURE 32: AGRICULTURE PILOT B1.3 BDVA REFERENCE MODEL FOR SATELLITE DATA ........................................................... 96 FIGURE 33: AGRICULTURE PILOT B1.4 MOTIVATION VIEW .............................................................................................. 102 FIGURE 34: AGRICULTURE PILOT B1.4 STRATEGY VIEW ................................................................................................... 103 FIGURE 35: AGRICULTURE PILOT B1.4 INITIAL ROADMAP ................................................................................................ 104



FIGURE 36: AGRICULTURE PILOT B1.4 BDVA REFERENCE MODEL ..................................................................................... 104 FIGURE 37: ZETOR TRACTORS .................................................................................................................................... 106 FIGURE 38: AGRICULTURE PILOT B2.1 MOTIVATION VIEW .............................................................................................. 111 FIGURE 39: AGRICULTURE PILOT B2.1 STRATEGY VIEW ................................................................................................... 112 FIGURE 40: AGRICULTURE PILOT B2.1 INITIAL ROADMAP ................................................................................................ 113 FIGURE 41: AGRICULTURE PILOT B2.1 STRATEGY VIEW ................................................................................................... 113 FIGURE 42: AGRICULTURE PILOT C1.1 MOTIVATION VIEW .............................................................................................. 119 FIGURE 43: AGRICULTURE PILOT C1.1 STRATEGY VIEW ................................................................................................... 120 FIGURE 44: AGRICULTURE PILOT C1.1 INITIAL ROADMAP ................................................................................................ 121 FIGURE 45: AGRICULTURE PILOT C1.1 BDVA REFERENCE MODEL ..................................................................................... 122 FIGURE 46: AGRICULTURE PILOT C1.2 MOTIVATION VIEW .............................................................................................. 129 FIGURE 47: AGRICULTURE PILOT C1.2 STRATEGY VIEW ................................................................................................... 130 FIGURE 48: AGRICULTURE PILOT C1.2 INITIAL ROADMAP ................................................................................................ 131 FIGURE 49: AGRICULTURE PILOT C1.2 BDVA REFERENCE MODEL ..................................................................................... 132 FIGURE 50: AGRICULTURE PILOT C2.1 MOTIVATION VIEW .............................................................................................. 140 FIGURE 51: AGRICULTURE PILOT C2.1 STRATEGY VIEW ................................................................................................... 141 FIGURE 52: AGRICULTURE PILOT C2.1 INITIAL ROADMAP ................................................................................................ 142 FIGURE 53: AGRICULTURE PILOT C2.1 BVDA REFERENCE MODEL ..................................................................................... 143 FIGURE 54: AGRICULTURE PILOT C2.2 MOTIVATION VIEW .............................................................................................. 150 FIGURE 55: AGRICULTURE PILOT C2.2 STRATEGY VIEW ................................................................................................... 151 FIGURE 56: AGRICULTURE PILOT C2.2 INITIAL ROADMAP ................................................................................................ 152 FIGURE 57: AGRICULTURE PILOT C2.2 BDVA REFERENCE MODEL ..................................................................................... 153

List of Tables TABLE 1: THE DATABIO CONSORTIUM PARTNERS ............................................................................................................. 13 TABLE 2: OVERVIEW OF AGRICULTURE PILOT CASES .......................................................................................................... 18 TABLE 3: ARCHIMATE MOTIVATION AND STRATEGY VIEWS................................................................................................ 23 TABLE 4: ELEMENTS USED IN THE ARCHIMATE MOTIVATION AND STRATEGY VIEWS ................................................................ 24 TABLE 5: AGRICULTURE PILOT A1.1 OVERVIEW OF PILOT ACTIVITIES .................................................................................... 27 TABLE 6: SUMMARY OF PILOT A1.1 (ISO JTC1 WG9 USE CASE TEMPLATE) ......................................................................... 29 TABLE 7: AGRICULTURE PILOT A1.1 STAKEHOLDERS AND USER STORIES................................................................................ 32 TABLE 8: SUMMARY OF PILOT A1.2 (ISO JTC1 WG9 USE CASE TEMPLATE) ......................................................................... 38 TABLE 9: AGRICULTURE PILOT A1.2 STAKEHOLDERS AND USER STORIES ................................................................................ 41 TABLE 10: SUMMARY OF PILOT A1.3 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 46 TABLE 11: AGRICULTURE PILOT A1.3 STAKEHOLDERS AND USER STORIES.............................................................................. 49 TABLE 12: SUMMARY OF PILOT A2.1 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 56 TABLE 13: AGRICULTURE PILOT A2.1 STAKEHOLDERS AND USER STORIES .............................................................................. 59 TABLE 14: SUMMARY OF PILOT B1.1 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 65 TABLE 15: AGRICULTURE PILOT B1.1 STAKEHOLDERS AND USER STORIES .............................................................................. 68 TABLE 16: AGRICULTURE PILOT B1.2 OVERVIEW OF PILOT ACTIVITIES .................................................................................. 74 TABLE 17: SUMMARY OF PILOT B1.2 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 76 TABLE 18: AGRICULTURE PILOT B1.2 STAKEHOLDERS AND USER STORIES .............................................................................. 78 TABLE 19: AGRICULTURE PILOT B1.3 OVERVIEW OF PILOT ACTIVITIES .................................................................................. 85 TABLE 20: SUMMARY OF PILOT B1.3 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 87 TABLE 21: AGRICULTURE PILOT B1.3 STAKEHOLDERS AND USER STORIES .............................................................................. 90 TABLE 22: SUMMARY OF PILOT B1.4 (ISO JTC1 WG9 USE CASE TEMPLATE) ....................................................................... 98 TABLE 23: AGRICULTURE PILOT B1.4 STAKEHOLDERS AND USER STORIES ............................................................................ 100 TABLE 24: SUMMARY OF PILOT B2.1 (ISO JTC1 WG9 USE CASE TEMPLATE) ..................................................................... 107 TABLE 25: AGRICULTURE PILOT B2.1 STAKEHOLDERS AND USER STORIES ............................................................................ 110



TABLE 26: AGRICULTURE PILOT C1.1 OVERVIEW OF PILOT ACTIVITIES ................................................................................ 114 TABLE 27: SUMMARY OF PILOT C1.1 (ISO JTC1 WG9 USE CASE TEMPLATE)...................................................................... 116 TABLE 28: AGRICULTURE PILOT C1.1 STAKEHOLDERS AND USER STORIES ............................................................................ 118 TABLE 29: SUMMARY OF PILOT C1.2 (ISO JTC1 WG9 USE CASE TEMPLATE)...................................................................... 124 TABLE 30: AGRICULTURE PILOT C1.2 STAKEHOLDERS AND USER STORIES ............................................................................ 127 TABLE 31: SUMMARY OF PILOT C2.1 (ISO JTC1 WG9 USE CASE TEMPLATE)...................................................................... 134 TABLE 32: AGRICULTURE PILOT C2.1 OVERVIEW OF PILOT ACTIVITIES ................................................................................ 138 TABLE 33: AGRICULTURE PILOT C2.1 STAKEHOLDERS AND USER STORIES ............................................................................ 139 TABLE 34: AGRICULTURE PILOT C2.2 OVERVIEW OF PILOT ACTIVITIES ................................................................................ 145 TABLE 35: SUMMARY OF PILOT C2.2 (ISO JTC1 WG9 USE CASE TEMPLATE)...................................................................... 146 TABLE 36: AGRICULTURE PILOT C2.2 STAKEHOLDERS AND USER STORIES ............................................................................ 148



Definitions, Acronyms and Abbreviations Acronym/

Abbreviation Title

BDVA Big Data Value Association

BDT Big Data Technology

CAP Common Agricultural Policy

CEN European Committee for Standardization

EO Earth Observation

ESA European Space Agency

EAGF European Agricultural Guarantee Fund

EU European Union

FAO Food and Agriculture Organisation of the United Nations

fAPAR fraction of Absorbed Photosynthetically Active Radiation

FAS Farm Advisory System

GAEC Good Agricultural and Environmental Conditions

GEOSS Group on Earth Observations

GPRS General Packet Radio Service

GS Genomic Selection

HPC High Performance Computing

IACS Integrated Administration and Control System

ICT Information and Communication Technologies

IoT Internet of Things

ISO International organization for Standardisation

KPI Key Performance Indicator

LPIS Land Parcel Identification System

NDVI Normalized Difference Vegetation Index

NGS Next-Generation Sequencing

NUTS Nomenclature of Territorial Units for Statistic

PC Personal Computer

PF Precision Farming

PU Public

RPAS Remotely Piloted Aircraft System

RTK Real Time Kinematic

SMEs Small and medium-sized enterprises

TRL Technology Readiness Level

UAV Unmanned Aerial Vehicle

UI User Interface

UVA, UVB (UV) ultraviolet rays, (A) long wave, (B) short wave

VRA Variable Rate Application

WP Work Package



Term Definition

Big Data A term of data sets that are so large or complex that traditional data

processing application software is inadequate to dealing with them

In situ Latin phrase translated “on site” or “on position”- it means “locally” or “in

place” to describe an event where it takes place

NDVI A simple graphical indicator that can be used to analyse remote sensing

measurements

WP (Work

Package)

A building block of the work breakdown structure that allows the project

management to define the steps necessary for completion of the work



Introduction 1.1 Project Summary The data intensive target sector on which the

DataBio project focuses is the Data-Driven

Bioeconomy. DataBio focuses on utilizing Big

Data to contribute to the production of the

best possible raw materials from agriculture,

forestry and fishery (aquaculture) for the

bioeconomy industry, as well as their further

processing into food, energy and

biomaterials, while taking into account various accountability and sustainability issues.

DataBio will deploy state-of-the-art big data technologies and existing partners’ infrastructure

and solutions, linked together through the DataBio Platform. These will aggregate Big Data

from the three identified sectors (agriculture, forestry and fishery), intelligently process them

and allow the three sectors to selectively utilize numerous platform components, according

to their requirements. The execution will be through continuous cooperation of end user and

technology provider companies, bioeconomy and technology research institutes, and

stakeholders from the big data value PPP programme.

DataBio is driven by the development, use and evaluation of a large number of pilots in the

three identified sectors, where associated partners and additional stakeholders are also

involved. The selected pilot concepts will be transformed to pilot implementations utilizing

co-innovative methods and tools. The pilots select and utilize the best suitable market-ready

or almost market-ready ICT, Big Data and Earth Observation methods, technologies, tools and

services to be integrated to the common DataBio Platform.

Based on the pilot results and the new DataBio Platform, new solutions and new business

opportunities are expected to emerge. DataBio will organize a series of trainings and

hackathons to support its uptake and to enable developers outside the consortium to design

and develop new tools, services and applications based on and for the DataBio Platform.

The DataBio consortium is listed in Table 1. For more information about the project see [REF-

01].

Table 1: The DataBio consortium partners

Number Name Short name Country

1 (CO) INTRASOFT INTERNATIONAL SA INTRASOFT Belgium

2 LESPROJEKT SLUZBY SRO LESPRO Czech Republic

3 ZAPADOCESKA UNIVERZITA V PLZNI UWB Czech Republic



4

FRAUNHOFER GESELLSCHAFT ZUR FOERDERUNG DER

ANGEWANDTEN FORSCHUNG E.V. Fraunhofer Germany

5 ATOS SPAIN SA ATOS Spain

6 STIFTELSEN SINTEF SINTEF ICT Norway

7 SPACEBEL SA SPACEBEL Belgium

8

VLAAMSE INSTELLING VOOR TECHNOLOGISCH

ONDERZOEK N.V. VITO Belgium

9

INSTYTUT CHEMII BIOORGANICZNEJ POLSKIEJ

AKADEMII NAUK PSNC Poland

10 CIAOTECH Srl CiaoT Italy

11 EMPRESA DE TRANSFORMACION AGRARIA SA TRAGSA Spain

12 INSTITUT FUR ANGEWANDTE INFORMATIK (INFAI) EV INFAI Germany

13 NEUROPUBLIC AE PLIROFORIKIS & EPIKOINONION NP Greece

14

Ústav pro hospodářskou úpravu lesů Brandýs nad

Labem UHUL FMI Czech Republic

15 INNOVATION ENGINEERING SRL InnoE Italy

16 Teknologian tutkimuskeskus VTT Oy VTT Finland

17 SINTEF FISKERI OG HAVBRUK AS

SINTEF

Fishery Norway

18 SUOMEN METSAKESKUS-FINLANDS SKOGSCENTRAL METSAK Finland

19 IBM ISRAEL - SCIENCE AND TECHNOLOGY LTD IBM Israel

20 MHG SYSTEMS OY - MHGS MHGS Finland

21 NB ADVIES BV NB Advies Netherlands

22

CONSIGLIO PER LA RICERCA IN AGRICOLTURA E

L'ANALISI DELL'ECONOMIA AGRARIA CREA Italy

23 FUNDACION AZTI - AZTI FUNDAZIOA AZTI Spain

24 KINGS BAY AS KingsBay Norway

25 EROS AS Eros Norway

26 ERVIK & SAEVIK AS ESAS Norway

27 LIEGRUPPEN FISKERI AS LiegFi Norway

28 E-GEOS SPA e-geos Italy



29 DANMARKS TEKNISKE UNIVERSITET DTU Denmark

30 FEDERUNACOMA SRL UNIPERSONALE Federu Italy

31

CSEM CENTRE SUISSE D'ELECTRONIQUE ET DE

MICROTECHNIQUE SA - RECHERCHE ET

DEVELOPPEMENT CSEM Switzerland

32 UNIVERSITAET ST. GALLEN UStG Switzerland

33 NORGES SILDESALGSLAG SA Sildes Norway

34 EXUS SOFTWARE LTD EXUS

United

Kingdom

35 CYBERNETICA AS CYBER Estonia

36

GAIA EPICHEIREIN ANONYMI ETAIREIA PSIFIAKON

YPIRESION GAIA Greece

37 SOFTEAM Softeam France

38

FUNDACION CITOLIVA, CENTRO DE INNOVACION Y

TECNOLOGIA DEL OLIVAR Y DEL ACEITE CITOLIVA Spain

39 TERRASIGNA SRL TerraS Romania

40

ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS

ANAPTYXIS CERTH Greece

41

METEOROLOGICAL AND ENVIRONMENTAL EARTH

OBSERVATION SRL MEEO Italy

42 ECHEBASTAR FLEET SOCIEDAD LIMITADA ECHEBF Spain

43 NOVAMONT SPA Novam Italy

44 SENOP OY Senop Finland

45

UNIVERSIDAD DEL PAIS VASCO/ EUSKAL HERRIKO

UNIBERTSITATEA EHU/UPV Spain

46

OPEN GEOSPATIAL CONSORTIUM (EUROPE) LIMITED

LBG OGCE

United

Kingdom

47 ZETOR TRACTORS AS ZETOR Czech Republic

48

COOPERATIVA AGRICOLA CESENATE SOCIETA

COOPERATIVA AGRICOLA CAC Italy



1.2 Document Scope Deliverable D1.1 – Agriculture Pilot Definition (due M06) specifies the pilot case descriptions,

requirement specifications, and implementation and evaluation plans. The document

describes 13 pilots and it will serve as basis for implementation of agriculture pilots, which

will be described in Agriculture Pilots intermediate report - Pilot results and feedback from

users in Month 24.

1.3 Document Structure

This document is comprised of the following chapters:

Chapter 1 presents an introduction to the project and the document.

Chapter 2 gives a general overview of the Agriculture Pilots t and summarises key points of

the pilot cases.

Chapters 3 to 17 describe the individual pilot cases.



Summary 2.1 Overview The agriculture sector is of strategic importance for the European society and economy. Due

to its complexity, agri-food operators have to manage many different and heterogeneous

sources of information. Agriculture is facing many economic challenges in terms of

productivity or cost-effectiveness, as well as an increasing labour shortage partly due to

depopulation of rural areas. Current systems still have significant drawbacks in areas such as

flexibility, efficiency, robustness, sustainability, high operator cost and capital investment.

Furthermore, reliable detection, accurate identification and proper quantification of

pathogens and other factors, affecting plant health, common agriculture policy, insurance,

are critical to be kept under control so as to reduce economy expenditures, trade disruptions

and even human health risks. Agriculture requires collection, storage, sharing and analysis of

large quantities of spatially and non-spatially referenced data. These data flows currently

hinder the adoption of precision agriculture as the multitude of data models, formats,

interfaces and reference systems in use result in incompatibilities. In order to plan and make

economically and environmentally sound decisions a combination and management of

information is needed.

2.2 Pilot introductions Big data technology (BDT) is a new technological paradigm that is driving the entire economy,

including low-tech industries such as agriculture where it is implemented under the banner

of precision farming (PF) [REF-03]. BDT in agriculture builds on geo-coded maps of agricultural

fields and the real-time monitoring of activities on the farm in order to increase the efficiency

of resource use, reduce the uncertainty of management decisions [REF-04]. Under PF, yield is

increased due particularly to the precise selection and application of exact types and doses of

agricultural inputs (crop varieties, fertilizers, pesticides, herbicides, irrigation water) for

optimum crop growth and development.

In terms of technology readiness level (TRL), the agriculture pilots are mostly positioned at

the sixth and seventh TRL. Improved technologies such as new elite varieties were developed,

big data such as weather, soil, crop (phenotypic data), and other environmental data are

routinely collected and meta-analysed, and technological and managerial services are already

offered to farmers in a few nations for a number of crops, although not in a scale that would

enable the application of big data analytics. There also exist experiences with farm telemetry

or utilization of satellite data (Earth Observation) in some countries. In addition, the required

skills are available in the organizations participating in the pilots, and the organizations are

ready to change their internal and external business processes, which is a key factor for

adopting the new technology.

The European farming system represents a mixture of small and big farms [REF-05]. In order

for WP1 pilots to account for both small and bigger farms, agriculture data serving as an input

into the big data analytics system will be gathered on a finer and a larger scale. The finer scale



is tailored to both farm sizes but with a particular focus on bigger farms with more financial

resources. Finer scale data include (1) data collected manually on soil, plants and other

agriculturally relevant factors, and through surveys and interviews; (2) historic big agriculture

and meteorological datasets; and (3) field-bound wireless sensor networks. Larger scale data

will be mainly derived from earth observation (EO) and include agriculturally relevant

information collected using remote sensing technologies and earth surveying techniques, and

from data coming from agriculture machinery. EO and finer scale information will be used

through big data analytics (WP4) to monitor and assess the status of, and changes in, the

agriculture pilots implemented in this project all across the European Union. Big data analytics

components and tools will then provide pilot managers with highly localized descriptive

(better and more advanced way of analysing an operation), prescriptive (timely

recommendations for operation improvement i.e., seed, fertilizer and other agricultural

inputs application rates, soil analysis, and localized weather and disease/pest reports, based

on real-time and historical data), and predictive plans (use current and historical data sets to

forecast future localized events and returns).

2.3 Overview of pilot cases The agriculture pilot cases are divided into three main topics as shown in the table below. For

all the pilots, co-innovative requirements (Task 1.1) were defined within the first six months

(M1-M6) of the project. Pilots activities under real production environment conditions will be

run over two to three cropping seasons (M6-M34) depending upon the plant species of

interest. (Tasks 1.2, 1.3, 1.4)

Table 2: Overview of agriculture pilot cases

Task (topic) Subtask Pilot group Pilot

T1.2 (A) Precision

Horticulture including

vine and olives

T1.2.1 A1: Precision agriculture

in olives, fruits, grapes

and vegetables

A1.1: Precision

agriculture in olives,

fruits, grapes

A1.2: Precision

agriculture in vegetable

seed crops

A1.3: Precision

agriculture in vegetables

-2 (Potatoes)



T1.2.2 A2: Big Data management

in greenhouse eco-

systems

A2.1: Big Data

management in

greenhouse eco-

systems

T1.3 (B) Arable

Precision Farming

T1.3.1 B1: Cereals and biomass

crops

B1.1: Cereals and

biomass crops

B1.2: Cereals and

biomass crops 2

B1.3: Cereals and

biomass crops 3

B1.4: Cereals and

biomass crops 4

T1.3.2 B2: Machinery

management

B2.1: Machinery

management

T1.4 (C) Subsidies and

insurance

T1.4.1 C1: Insurance C1.1: Insurance (Greece)

C1.2: Farm Weather

Insurance Assessment

T1.4.2 C2: CAP support C2.1: CAP Support

C2.2: CAP Support

(Greece)

The topics are defined as follows:

A. Precision Horticulture including vine and olives led by NP: In our days, farmers face a

series of challenges in their business. Resistant crop diseases and climate change

affects their crop production. At the same time, as the global demand for commodities

increases, farmers are forced to maximize their production. Following the rules of the

modern agro-food market, farmers and cooperatives that wish to export their

products abroad, need to follow smart agriculture practices.



B. Arable Precision Farming led by Vito: The overall objective is to implement big data

technology tools for precision and resilient farming of the food crop species of interest

including durum wheat, corn, grapes, etc. Focus of this pilot will be not only on

production aspects, but also on protection of water and soil as well as on energy

saving.

C. CAP support and insurance lead institution led by e-GEOS: The focus will be on using

Earth Observation data for the purpose of insurance and EU Common Agricultural

Policy.

Each topic includes two pilot groups:

Pilot group A1: Precision agriculture in olives, fruits, grapes and vegetables (NEUROPUBLIC,

VITO, GAIA, InfAI and CAC)

The following services will be offered:

• Remote plant disease diagnosis and assessment based on the processing of Satellite images;

• weather condition alert system which will result in the decision taking of specific actions; (e.g., crop protection);

• provision of automated irrigation systems based in precision irrigation enabling in this way an efficient water resource management system;

• support of efficient soil fertilization and spray practices consistent with the specific needs of the farm and the protection of the environment;

• advisory services regarding crop diversification will be also provided to the farmers directing them in more productive and resilient cultivations.

It will be focused on combined use of soil data, weather data, map data, satellite (LR, HR, VHR,

SAR), farm logs, UAV, farm profile data, and data collected by mobile audio-visual devices.

Pilot group A2: Big Data management in greenhouse eco-systems (CERTH, CREA)

The overall objective of the proposed pilot is to provide knowledge, know‐how & tools related

to the information flow, management and data analytics in greenhouse horticulture. To this

purpose, genomics, metabolomics and phenomics data will be combined. During this project,

it will be used already produced genomic data which will be integrated with new ones in order

to assess the genetic potential of new tomato varieties and their performance in greenhouses.

The aim is to integrate metabolomics and genomics data to obtain a complete identity of the

varieties for breeding applications. Liquid chromatography - mass spectrometry (LC-MS), Gas

chromatography - mass spectrometry (GS-MS), High-performance liquid chromatography

(HPLC) will be used to collect the metabolomics data. Market potential and industry interests:

Tomato is among the top cultivated crops in greenhouses, with billions of euros turnover

worldwide. Tomato is considered one of the most nutritive solanum vegetables due to its high

content in sugars, vitamins and antioxidants and its consumption is steadily increasing. The

pilot is expected to leverage the productivity and the quality of tomato.



Pilot group B1: Cereals and biomass crops (Vito, Lesprojekt, NEUROPUBLIC, Federunacoma,

CREA, NOVAMONT, ZETOR, GAIA, CERTH, NB Advies, CiaoT, ASTER, InfAI, Lesprojekt,

Federunacoma, e-GEOS, PSNC, TRAGSA)

This pilot aims to provide information for precision agriculture, mainly based on time series

of high resolution (Sentinel-2 type) satellite images, complemented with UAV images, metro

and field (sensor) data. The information can be used as input for farm management

(operational decisions, tactical decisions). Information layers may include: - Vegetation

indices (NDVI, fAPAR, …) and derived anomaly maps. Anomaly maps can be used to set

priorities for field visits (local/regional level). Pilots on durum wheat will be conducted in

different environments in Italy in collaboration with Horta Srl, (private company), CNR-Ibimet

(public research institute) and local Producer organization and cooperatives in Italy using in

addition to the tools listed above. Pilots on precision irrigation in corn will be conducted in a

NEUROPUBLIC pilot site in Kalampaki area, in Drama Greece. The pilots will run in partnership

with end users GAIA EPICHEIREIN and the local Agricultural Association, representing the local

corn producers. Biomass crops (CREA, VITO, CERTH, NB Advies, CiaoT, ASTER, InfAI,

Lesprojekt, Federunacoma, e-GEOS, PSNC, TRAGSA). Biomass crops including biomass

sorghum, fiber hemp and milk thistle can be used for several purposes including, respectively,

biofuel, fiber, and biochemicals, with a high macroeconomic impact. The pilots on these crops

will be run in collaboration between CREA and private companies (end-users) Cooperativa

Agricola Cesenate (seed company), Novamont (Bio-based company), and Centro Ricerche

Produzioni Animali, and another 15 agricultural firms distributed across the Italian territory.

Pilot group B2: Machinery management (Lesprojekt, Federunacoma, ZETOR)

From technical point of view the monitoring system involves tracking of the vehicles’ position

using GPS combined with acquisition of information from on-board terminal (CAN-BUS) and

their online or offline transfer to GIS environment. Such systems collect large amounts of

data. The monitoring system will be done in large, medium-sized and small farms based on

the level of information processing and their interaction with other farm data, three use cases

will be handled.

Pilot group C1: Insurance (e-GEOS, VITO, NEUROPUBLIC, NB Advies, CSEM)

The objective of this pilot is the provision and assessment on a test area of services for

agriculture insurance market, based on the usage of Copernicus satellite data series also

integrated with meteorological data, and other ground available data.

Pilot group C2: CAP support (e-GEOS, CSEM, NEUROPUBLIC, GAIA)

The objective of the pilot is the provision of products and services, based on specialized highly

automated processors processing big data, in support to the CAP and relying on multi-

temporal series of free and open EO data, with focus on Copernicus Sentinel 2 data. Products

and services will be tuned in order to fulfil requirements from the 2015-20 EU CAP policy, and

will be general information layers and indicators on EU territory with different level of



aggregation and detail up to farm level. The proposed pilot project has been tailored on the

specific needs of three end users, one operating at National level (Romania Agriculture

Ministry), one operating at Regional level (AVEPA Paying Agency) in one of the most

important agricultural regions in Italy, and one operating in Greece.

2.4 Agriculture datasets utilized in pilots The datasets used by the agriculture pilots can be coarsely divided into four distinct

categories. In situ measurements are data obtained by sensors in the field, Machinery

Measurements are coming from sensors in agriculture machinery. Remote measurements are

measurements which may cover a greater geographical area, such as measurements from

satellites. VGI data and data collected by farmers. The biggest data sets will come from Earth

Observation and Machine monitoring. The current experience from Czech Republic

demonstrate that machinery monitoring in Czech Republic is yearly able to generate more

than 20 TB of data and the needs of satellite data is approximately 5 TB per year. The data

from unmanned aerial vehicles (UAV) will be much larger.

2.5 Representation of pilot cases Each pilot is described in following structure:

● PILOT OVERVIEW o Pilot introduction

o Pilot overview

● PILOT CASE DEFINITION o Stakeholder and user stories

o Motivation and strategy

● PILOT MODELLING WITH ARCHIMATE o Motivation view

o Strategy view

● PILOT EVALUATION PLAN o High level goals and KPI's

o Initial roadmap

● BIG DATA ASSETS

2.6 Pilot modelling framework The pilot cases are modelled using the ArchiMate 3.0 modelling framework. Figure 1

summarizes the overall ArchiMate 3.0 framework. The figure also depicts the input provided

by the domain WPs (WP1, WP2, WP3 and their pilots) and that provided by the technology

WPs (WP4, WP5), which will be correlated in the next stages of modelling process.



Figure 1: ArchiMate 3.0 modelling framework.

The modelling presented in this deliverable focuses on the “Motivation” and “Strategy” views.

The “Motivation” view models the reasons that guide the design of the architecture. The

“Strategy” view adds how the course of action is realized. Table 3 provides an extended

description of the two views. After the completion of this deliverable, the plan is to extend

the modelling with other views, while investigating the correlations with the technology WP

input.

Table 3: ArchiMate Motivation and Strategy views.

View name Description

Motivation

view

Motivation elements are used to model the motivations, or reasons, that guide the

design or change of an Enterprise Architecture. It is essential to understand the

factors, often referred to as drivers, which influence other motivation elements.

They can originate from either inside or outside the enterprise. Internal drivers, also

called concerns, are associated with stakeholders, which can be some individual

human being or some group of human beings, such as a project team, enterprise, or

society. Examples of such internal drivers are customer satisfaction, compliance to

legislation, or profitability.



Strategy

view

The immediate decision support system is built on top of a data collection and

distribution system. The data collection and distribution system is used to collect

sensor data from the on-board systems and makes them available in a single system.

The data distribution system ensures that the decision support system only interface

with a single system, instead of multiple sensors. The decision support system

presents the data from the data distribution system and collect them in an internal

storage system for presentation of current performance vs. historic performance.

The main elements used in the above views are explained in Table 4. Their relationships are

shown in Figure 2and Figure 3. For further information see [REF-02].

Table 4: Elements used in the ArchiMate Motivation and Strategy views

Element Definition Notation

Stakeholder The role of an individual, team,

or organization (or classes

thereof) that represents their

interests in the outcome of the

architecture.

Driver An external or internal condition

that motivates an organization

to define its goals and

implement the changes

necessary to achieve them.

Assessment The result of an analysis of the

state of affairs of the enterprise

with respect to some driver.

Goal A high-level statement of intent,

direction, or desired end state

for an organization and its

stakeholders.

Outcome An end result that has been

achieved.



Principle A qualitative statement of intent

that should be met by the

architecture.

Requirement A statement of need that must

be met by the architecture.

Constraint A factor that prevents or

obstructs the realization of

goals.

Meaning The knowledge or expertise

present in, or the interpretation

given to, a core element in a

particular context.

Value The relative worth, utility, or

importance of a core element or

an outcome.

Resource An asset owned or controlled by

an individual or organization.

Capability An ability that an active

structure element, such as an

organization, person, or system,

possesses.

Course of

action

An approach or plan for

configuring some capabilities

and resources of the enterprise,

undertaken to achieve a goal.



Figure 2: Relationships of the Motivation elements

Figure 3: Relationships of the Strategy elements



Pilot 1 [A1.1] Precision agriculture in olives,

fruits, grapes 3.1 Pilot overview

3.1.1 Pilot introduction

The world population is expected to reach 9 billion by 2050 and feeding that population will

require a 70 percent increase in food production (FAO 2009) [REF-06]. At the same time,

farmers are facing a series of challenges in their businesses that affect their farm production,

such as crop pests and diseases with increased resistance along with drastic changes due to

the effects of the climate change. These factors lead to rising food prices that have pushed

over 40 million people into poverty since 2010, a fact that highlights the need for more

effective interventions in agriculture (World Bank 2011) [REF-07]. In this context, agri-food

researchers are working on approaches that aim at maximizing agricultural production and

reducing yield risk. The benefits of the ICT-based revolution have already significantly

improved agricultural productivity; however, there is a demonstrable need for a new

revolution that will contribute to “smart” farming and help addressing all the aforementioned

problems (World Bank 2011) [REF-07].

There is a need for services that are powered by scientific knowledge, driven by facts and

offer inexpensive yet valuable advice to farmers. In this context, smart farming is expected to

reduce production costs, increase production (quantitatively) and improve its quality, protect

the environment and minimize farmers’ risks.

3.1.2 Pilot overview

The main focus of this pilot is to offer smart farming services dedicated for olives, fruits and

grapes, based on a set of complementary monitoring technologies. Smart farming services

comprise irrigation, fertilization and pest/disease management advice provided through

flexible mechanisms and UIs (web, mobile, tablet compatible). The pilot will target towards

promoting the adoption of technological advances (IoT, Big Data analytics, EO data) and

collaborating with certified professionals to optimize farm management procedures. NP and

GAIA Epicheirein will support the activities for the execution of the full life-cycle of the pilot.

The following table provides an overview of the pilot activities.

Table 5: Agriculture pilot A1.1 Overview of pilot activities

Pilot Site A Pilot Site B Pilot Site C

Location Chalkidiki, Greece Stimagka, Greece Veria, Greece

Area Size 600ha 3,000ha 10,000ha

Targeted Crops Olive Trees Grapes Peaches



End-Users Single farmer,

Agronomists

Farming

organization,

Agronomists

Farming

cooperative,

Agronomists

The underlying reason for the selection of these particular crop types is the significant

economic impact that they share in the Greek farming landscape. Olive tree cultivation

accounts for nearly 2 billion euros in annual net income, while peach and grape cultivations

reach close to 460 million and 390 million annual net income respectively.

Method

This pilot is targeting towards providing a set of smart farming services to the farmer utilizing

available precision agriculture techniques. The services will be provided as advices, which

need many prerequisites and primary material in order to be accurate. Data is the raw

material and there are three different means of collecting data, which will be exploited within

the pilot activities. Data directly from the field, collected from a network of telemetric IoT

stations called GAIAtrons; remotely with image sensors on in-orbit platforms; and by

monitoring the application of inputs and outputs in the farm (e.g. in-situ measurements, farm

logs, farm profile). Every data source has unique characteristics with relevant impact on the

very content of this data. Field sensing provides real-time accurate direct measures of many

physical parameters of the soil (soil temperature, humidity), atmosphere microclimate of the

field crop and plant (ambient temperature, humidity, barometric pressure, solar radiation,

leaf wetness, rainfall volume, wind speed and direction) with temporal continuity. Remote

sensing provides indirect measures of some physical properties of plants and soil with spatial

continuity in medium to large spatial scale. Combining this information can provide a good

knowledge of the most important physical parameters of soil, microclimate, plants and water

(which are all the environmental resources, which govern farming) in both spatial and

temporal dimensions. Monitoring the application of inputs and outputs on the farm is a data

element that is necessary to assess the correctness of the given advice and use it as feedback

to improve the system over time. This pilot will combine advanced data handling techniques

(i.e. assimilation, fusion and spatio-temporal interpolation) to transform the collected data

into actionable advice. In order for this advice to reflect the actual situation at a given field,

we will deploy scientific models and we will seek to incorporate the human experience of the

farmer or certified advisors.

Relevance to and availability of Big Data and Big Data infrastructure

NP has already started collecting field-sensing data through its network of telemetric IoT

stations, called GAIAtrons. GAIAtrons offer configurable data collection and transmission

rates. Since 01/03/2016 over 1M samples have been collected and stored to NP’s cloud

infrastructure that refer to atmospheric and soil measurements from various agricultural

areas of Greece. Moreover, within the same cloud infrastructure (GAIA cloud), remote sensing

data from the new Sentinel 2 optical products are being extracted and stored since the

beginning of 2016. This comprises both raw and processed (corrected products, extracted



indices) data represented in raster formats that are being handled and distributed using

optimal big data management methodologies. Finally, through flexible work calendars, NP

has collected more than 120000 records related to work plans of the farmers that can be used

in the context of the pilot activities.

Benefit of pilot

The pilot is expected to have a direct impact on farm profitability in three (3) major crop types

of Greece, from an economic perspective. This will ensure that the proposed solutions can be

replicated to other crop types and market segments in the near future. The holistic approach

that is being proposed will significantly improve the capacity of the responsible partners in

providing smart farming advisory services. In addition, it would lead to improvements in a)

NP’s GAIA cloud’s stability, availability, security, interoperability and overall maturity, b) NP’s

GAIABus DataSmart functionality in terms of real-time analytics, data stream and decision

support processes, multi-temporal object-based monitoring, cloud-based services that

integrate earth observation with image processing, machine learning and spatial modelling,

c) advancing the current system by fusing telemetry IoT stations’ data with remote sensing

data and incorporating advanced visualization and event-based capabilities.

3.2 Pilot case definition

Table 6: Summary of pilot A1.1 (ISO JTC1 WG9 use case template)

Use case title Precision agriculture in olives, fruits, grapes

Vertical (area) Agriculture

Author/company/email NP, GAIA Epicheirein

Actors/stakeholders and their roles and

responsibilities

● Single Farmer/Farming Organization or Cooperative, responsible for performing farming activities

● Agronomists, involved in providing relevant and up-to-date advices to the farmers

Goals Provide smart farming advisory services (focusing on irrigation, fertilization and pest/disease management), based on a set of complementary monitoring technologies, in order to increase farm profitability and promote sustainable farming practises.

Use case description Refer to the pilot case definition section and diagrams in the pilot modelling sections.

Current

solutions

Compute(System) System is based on IoT data, farm logs, work calendars and in-situ measurements. Expert knowledge is provided through static scientific



models that offer insight about optimal farm management.

Storage All available data are stored in a cloud infrastructure.

Networking Web-based UIs and dashboards available for monitoring farm activities.

Software Real-time analytics, data stream processes and decision support system

Big data

characteristics

Data source (distributed/centralized)

Centralized (within GAIA Cloud): Field sensing data from GAIAtrons, Remote sensing (Earth observation) data, Farm data

Volume (size) ● ~5.5 TB/year for remote sensing data, including raw data and extracted biophysical and vegetation indices for the pilot areas

● several GBs/year field sensing data collected by the deployed GAIAtrons (related to the number of GAIAtrons to be used within the pilot activities)

● Hundreds of thousands of records related to farm activities/profiles/measurements

Velocity

(e.g. real time)

Configurable data transmission for field sensing (a new set of measurements is being sent every 10 minutes in present configuration). Every 10 days new EO products available. Within 2018 EO products will be available every 5 days.

Variety

(multiple datasets, mashup)

Field Sensing: Soil temperature, humidity (multi-depth), ambient temperature, humidity, barometric pressure, solar radiation, leaf wetness,



rainfall volume, wind speed and direction

Remote Sensing: 13 spectral bands

Variability (rate of change)

Same as above, rate of change depends very much on data source/type.

Big data science (collection, curation,

analysis,

action)

Veracity (Robustness Issues, semantics)

Need for a system that can constantly provide relevant and up-to-date advices to its end-users

Visualization Spatio-temporal information visualization for improving farm management and facilitating the decision-making process

Data quality (syntax) The quality of field sensing data is controlled by several filtering, outlier detection and stream processing mechanisms. The integrity of remote sensing data quality is being assessed by a hash check upon product download.

Data types Remote sensing data provided in raster format (.jp2). Field sensing data provided as time series unstructured data with configurable frequency

Data analytics Descriptive and prescriptive analytics for the provision of irrigation, fertilization and pest management advices.

Big data specific challenges (Gaps)

There is a need for smarter fusion of the heterogeneous data types that are being collected towards providing accurate insights. To this end, it is important to explore mechanisms that could combine raster and vector data at parcel level (polygon) and station level (point).

Big data specific challenges in bio-

economy

In order to facilitate the adoption of the big data technologies by the farmers, imposed barriers in data visualization should be encountered (e.g. give more emphasis to vector data, improvement of the aggregation mechanism (drill down, zoom in, roll up, zoom out)).



Security and privacy

technical considerations

A system intended to collect data from field sensors, installed in remote locations, is definitely going to face network connectivity challenges. In order to provide up-to-date and relevant advices, the system should be able to exhibit high availability and accuracy in its sensor readings and transmission mechanisms. Moreover, field sensing data should be securely transmitted to the cloud infrastructure and protected against various types of attacks that might set the system at risk.

Highlight issues for generalizing this Use

case (e.g. for ref. architecture)

EO data management mechanisms can be exploited for other use cases where EO data might provide valuable insights.

3.2.1 Stakeholder and user stories

Table 7: Agriculture pilot A1.1 Stakeholders and user stories

Stakeholders User story Motivation

Farmer As a farmer I want to reduce costs and improve farm productivity

Increase my profits following sustainable agriculture practices

Agronomists As an agronomist I want to have a comparative advantage in a highly competitive market and to offer the best possible services to my clients

Increase my profits by providing better advices based on evidences, well-established arguments and scientific knowledge.

3.2.2 Motivation and strategy

The main motivation for this pilot is:

• to raise the awareness of the farmers, agronomists, agricultural advisors, farmer cooperatives and organizations (e.g. group of producers) on how new technological tools could optimize farm profitability and offer a significant advantage on a highly competitive sector.

• to promote sustainable farming practises over a better control and management of the resources (water, fertilizers, etc.).

• to increase the technological capacity of the involved partners through a set of pilot activities that involves management of big data for high value crops.

The pilot motivation and strategy is summarized using ArchiMate diagrams in the next

section, while goals and KPIs are addressed in the successive evaluation plan.



3.3 Pilot modelling with ArchiMate The current section presents the "Agriculture A.1.1 modelling with ArchiMate" view point

described using the ArchiMate standard.

3.3.1 Agriculture pilot A1.1 Motivation view

This section provides the "Agriculture A1.1 Motivation view" view defined in the "Agriculture

A.1.1 modelling with ArchiMate" view point.

Figure 4: Agriculture pilot A1.1 Motivation view

Farmers want cost reduction and improved productivity in order to increase their profits

following sustainable agriculture practices.

3.3.2 Agriculture pilot A1.1 Strategy view

This section provides the "Agriculture A1.1 Strategy view" view defined in the "Agriculture




Figure 5: Agriculture pilot A1.1 Strategy view

The main focus of this pilot is to offer smart farming services dedicated for olives, fruits and

grapes, based on a set of complementary monitoring technologies. Smart farming services

comprise irrigation, fertilization and pest/disease management advice provided through

flexible mechanisms and UIs (web, mobile, tablet compatible). The pilot will target towards

promoting the adoption of technological tools (IoT, Big Data analytics, EO data) and

collaborating with certified professionals to boost/optimize farm productivity.

3.4 Pilot Evaluation Plan

3.4.1 High level goals and KPI's

Two relevant KPIs have been identified so far, namely:

• %Reduction potential in operational costs for performing the same farming activities (through better management of resources) following the advisory irrigation, fertilization, pest/disease management services vs what would be the operational costs following standard farming practices based on historical data: Quantify %reduction potential in operational costs for all three crop types (in fresh water/fertilizer usage, sprays following the aforementioned advisory services).



• %Increase in farm yield following the advisory irrigation, fertilization, pest/ disease management services vs what would be the yield following standard farming practices based on historical data: Quantify %increase in farm yield for all three crop types.

3.4.2 Initial roadmap

A coarse roadmap with important milestones for the pilot is included below. It has been

adapted to the two scheduled iterations of the DataBio platform and depends on these

internal project deliveries from work package 4 (WP4).

Figure 6: Agriculture pilot A1.1 initial roadmap

3.5 Big data assets The diagram below summarizes Big Data technology components used in this pilot using the

extended BDVA reference model. Where applicable, specific partner components have been

indicated in the list using the component ids (DataBio project specific) that are likely to be

used, or evaluated for use, by this pilot.



Figure 7: Agriculture pilot A1.1 BDVA reference model



Pilot 2 [A1.2] Precision agriculture in

vegetable seed crops 4.1 Pilot overview


Eastern Italy is by tradition one of the areas in the world where seed production is at its best.

Seed Companies from all over the world produce on contract with local growers’ vegetables,

sugar beets, alfa-alfa and many other species.

One of the key factor for the achievement of seeds of good quality depends on the choice of

the right time of harvesting: if too early the vigour of the seed harvested will be affected; if

too late the mature seeds are going to drop to the ground and the best part of harvest get

lost.

The pilot will concentrate its main focus in monitoring the maturity of seed crops of different

species with satellite imagery. There will be an on-land observation of the crop development

which will be matched with satellite images in order to check the possibility to establish a

correspondence between images and the maturity stage of each crop.

In first growing season, the crop monitored will be sugar beet for seed production, with the

aim to expand the observation to other seed crops.


Location: 5 farms, Region Emilia Romagna, for the total acreage of 14,79 hectares in the first

year.

To be expanded to other crops in the same Region and in Region Marche.

Method

This pilot will use satellite imagery (Sentinel-2) and telemetry IoT for crop monitoring and

yield/seed maturity estimation. The pilots will be run by C.A.C. in collaboration with VITO.

The crop involved in first year is sugar beet; according to the results achieved the model may

be expanded to other seed crops, namely cabbage and onion. VITO will use satellite data to

monitor the crops and will develop yield/seed maturity models. Telemetry IoT technology

will be implemented by C.A.C. on 5 farms located in Emilia Romagna and Marche.

Specifically, as part of pilot innovative solution, an online platform will be used to provide

satellite imagery, weather and soil data and yield/seed maturity predictions. VITO, in

collaboration with a number of Belgian partners, has developed a web application

“WatchITgrow®” for potato monitoring and yield prediction in Belgium. The existing

WatchITgrow® application will “filled” with satellite, weather and soil data for the Italian pilot

sites. To be able to provide maturity estimates developments are needed and it is necessary

to collect field data. The data will be collected by C.A.C.



The farmer and pilot owners can use the satellite imagery (biomass index, 10m resolution) to

monitor and benchmark the maturity curve of seed crops till harvesting in correlation with

weather and microclimatic conditions recorded on site through dedicated meteorological

units.

A weather station will be installed in the vicinity of each field with sensors for air moisture

and temperature, soil temperature, rainfall – remote monitored.

Telemetry IoT stations will transmit data to the cloud infrastructure in the process of crop

monitoring, biotic and abiotic stress diagnostic, alert and operational recommendations.

Benefit of pilot

The solution that will be developed will be for the benefit of the co-operative which is

organising the production with its associated growers.

Each crop gets in maturity stage according to the cycle of the variety, microclimate, land

conditions, water supply etc.

The aim is to monitor the stage of maturity of each crop using satellite imagery (and possibly

telemetry IoT). This information can help fieldsmen to organise efficiently their time in

assisting the growers.

The fieldsman and the farmers who are participating in the pilot will have access to satellite

images, weather and soil data and information on seed maturity via an online platform. The

farmers will provide crop data about their fields for system learning.



Use case title Precision agriculture in seed crops


Author/company/email Stefano Balestri / C.A.C. / balestriacseeds.it

Isabelle Piccard / VITO


responsibilities

Fieldsmen, Growers and their co-operatives

Goals To produce a modelling in order to predict the maturity of seed crops in order to organize harvest in the most efficient way and get mature, high quality seeds




Current

solutions

Compute(System) On spot decisions made on the empiric experience of fieldsmen

Storage Local system + Company information system

Networking “Crop report” application on web, chat groups on Wats’app

Software Mobile application

Big data characteristics


Availability of Sentinel-2 data (derived vegetation indices).

Scientific modelling – built phenology model.

Visualization – Processed data and model results are published in an intuitive way.

Volume (size) Hundreds of terabytes per year when all sources of data are considered.

Velocity

(e.g. real time)

Satellite data: Sentinel-2A+B images are acquired with a time step of 5 days. The images are pre-processed and distributed by ESA within 24 hours after acquisition. Further processing by VITO starts as soon as the images are available from ESA. Generally, the final information products become available for the end-users between 24 and 48 hours after image acquisition.

Telemetry IoT data: Time step for data collection is customizable, 1-60 minutes; big data: air temperature, air moisture, rainfall, soil temperature.

Phenotypic data are collected each cropping season.

Variety


Great variety. (1) Satellite: imagery, multispectral data, indices (soil, water, vegetation, biophysical), (2) Telemetry IOT: air temperature, air moisture, rainfall, soil temperature. (3) analytics and phenotypic data.






analysis,

action)


Need to have tools to produce and process ground-truth data for satellite data calibration.

Visualization Visualization of crop monitoring output at least bi-weekly during the cropping season, indices and predictions; real-time monitoring output, alerts, and recommendations.

Data quality (syntax) Data validity filtering w.r.t. completeness. Data fusion and modelling of heterogeneous data (EO data, telemetry IoT data, field data)

Data types Imagery, graphics, vector, numbers, analytical results, measurements, metadata, geolocations, spectra, time series.

Data analytics Predictive analytics for the development of data-driven yield models; predictive feedback (monitoring), real-time streaming data analytics to alert and provide operational recommendations using cloud-based crop management analytics including web portal cloud solution.


There is a need for: (1) improving analytic and modelling systems that provide reliable and robust statistical estimated using large size of heterogeneous data; (2) reduced uncertainty of management decisions.

Big data specific challenges in bioeconomy

Delivering content and services to various computing platforms from Windows desktops to Android and iOS mobile devices



Farm owner and geolocalization are highly sensitive, should be anonymized



Real-time streaming data analytics and predictive analytics using machine learning for crop monitoring and developing yield models based on big data are universal solutions with domain agnostic applications.

More information (URLs)

www.databio.eu

<other URLs to be added later if relevant>

Note:

http://www.databio.eu/




The end users are farmers, farming cooperatives, seed industry and research institutions.

VITO will use satellite data to monitor the crops and will develop maturity models. Telemetry

IoT technology will be implemented by C.A.C. to provide real-time analytics solutions.

Combining crop monitoring and real-time analytics will provide localized predictive plans to

improve farming operations and quality of harvest. Stakeholders and user stories are

summarized in the table below.

Table 9: Agriculture pilot A1.2 stakeholders and user stories


Farmers, farming

cooperatives

want to monitor the maturity of

their seed crops throughout the

season

to evaluate the right time

for harvesting, optimize

field operations and get

high quality of their

products.

Fieldsmen Remote monitoring of maturity of

seed crops

Organise the harvesting

operations more

effectively, save time and

money.


The main motivations for this pilot are:

• Identifying the potential for using satellite data and machine learning for monitoring crop development and maturity and the development of prediction models

• Evaluating the comparative importance between the use of proximal wireless sensor network data and satellite data






This section presents the "Agriculture A1.2 Motivation view" view defined in the "Agriculture






This section presents the "Agriculture A1.2 Strategy view" view defined in the "Agriculture







Three relevant KPIs have been identified so far:

• Model accuracy: maturity/yield prediction accuracy showing an acceptable error rate when tested on data that it was not trained on.

• Revenue potential with alternative cropping strategy vs. what happened: Quantify increased revenue potential on historical data, .e.g. what was the accumulated value, vs what could have been achieved using conventional cropping techniques.

• System usage: Number of users of DataBio technologies - yield models and proximal wireless sensors in seed crops. This is technology transfer and takes time to establish; for this project, a baseline will be measured first, then followed-up by monitoring usage after system deployment.




internal project deliveries from work package 4 and 5 (WP4, WP5).











Pilot 3 [A1.3] Precision agriculture in

vegetables_2 (Potatoes) 5.1 Pilot overview


Potato has been the major crop in this region for many years. Due to the reform in CAP

(Europe’s Common Agricultural Policy) the market is changing and farmers are urged to

increase their yields, but in a sustainable way. This means they need to be more conscious

the energy and other resources they use in producing their crops. AVEBE is a cooperative for

the potato growing farmers and is supporting their growers in an innovation program called

Towards 20-15-10, a program which started in 2012. The objectives of this program are to

realize in 2020 an average of 15 tons of starch per ha with a variable cost price of €10 per 100

kg starch. To monitor these objectives farmers are sharing data about their yields and farming

practices in study groups.


The goal of this pilot is to provide the potato farmers information during the growing season

about the potential and actual yield predictions and the actions they can take to mitigate the

foreseen yield losses. The pilot will supply the farmers with benchmark data about their crops

compared to the region, previous growing seasons etc. The data provided could also be the

basis for timely and more location specific treatment.

Method

The basis for the yield predictions will be the combination of data from different sources in a

self-learning system.

Using historical yield data and historical earth observation data, machine learning will be

applied to model the potato growth and calculate yield prediction for the current year, based

on recent earth observation data. The more yield (historical) yield data is supplied, the better

the predictions will be.

Specifically, as part of pilot solution, an online platform will be used to provide satellite

imagery, weather data and yield predictions. The farmer can use the satellite imagery

(biomass index, 10m resolution) to monitor and benchmark their field productivity potential

relative to production levels achieved in the region. After one year, yield prediction can be

implemented using the test data from year one, and other historical data (when available).


VITO has archives of earth observation (EO) data based on several satellite platforms. The

recently released Sentinel 2 (A and B) will be the most detailed sources, which we hope to be

able to use.



The study groups have collected detailed field data for some years which will useful for

calibrating the system.

Benefit of pilot

This pilot will focus on the family farms in the Veenkoloniën region in the Netherlands, which

are members of the AVEBE cooperative. The main research partner will be VITO.

Especially these farms will benefit from this pilot, but as a spinoff the farms will be able to

grow their crops in a more sustainable way, which will be beneficial to all, farmers and the

people in the whole region.

5.2 Pilot case definition Current situation: Farmers monitor their crops just by their own observations and samples,

which is time consuming. Furthermore, the disadvantages are that is hard to create a good

overview based on just some observations and samples. Deviations in growth within the field

are hard to observe.

Pilot solution:

The solution that will be developed will consist of an online platform providing satellite

imagery with a 10m resolution (Sentinel 2A + B, possibly complemented with coarser

resolution data to overcome cloud problems), weather data, soil data and yield predictions.

The system will be based on the combination of data from different sources (a.o. Satellite

imagery) in a self-learning system, providing better predictions when more yield (historical)

yield data is supplied.

This first version will be the basis for the pilot; it will be “filled” with specific data, like varieties

of potatoes are grown, soils and weather conditions. To be able to provide yield forecasts

developments are needed. The coming year’s data we will collect data about the potato crops

that are grown at the pilot site in order to develop a model for yield prediction. In previous

years several study groups of farmers have collected data about their potato crops.

Potentially these historical yield data sources can also be used for ‘learning’ the model.

Each farmer who is participating in the pilot will provide crop data about his fields as input for

system learning. The farmer can use the satellite imagery (biomass index) to monitor his field

and benchmark his fields with other potato fields in the area. Benchmarking options will be

provided as well as multi-year comparison.

After one year yield prediction can be implemented using the test data from year 1, and other

historical data (when available).


Use case title Precision agriculture in Potatoes; Benchmarking and yield prediction




Author/company/email Nicole Bartelds / NB Advies / [email protected]


responsibilities

Farmer, responsible for growing potatoes in a sustainable way.

Cooperative / processing industry, responsible for processing potatoes while realizing the best price for the farmers

Goals Improve farming practices by providing benchmark information to the farmers.


Current

solutions

Compute(System) Non-existing system today

Storage Local system of study groups

Networking Not available online

Software Some standalone data management



Combination of both types:

Centralized: EO data by VITO, weather data (Sentinel 2A + B, possibly complemented with coarser resolution data

Distributed: field characteristics (sample data yield data, potato varieties, planting data etc.) from different study groups

Volume (size) Terabytes per year: Primarily for EO data

Velocity

(e.g. real time)

Almost real-time (24h after acquisition) processed EO data from Sentinel 2 (potentially every 5 days), yearly crop/field data

Variety


Medium variety of data sources EO: imagery, multispectral data, indices; weather: temperature, rain; crop/field characteristics


EO data potentially every 5 days


analysis,

action)


EO data in itself is not a reliable source for yield prediction. The system will need adequate calibration based on field data.

Visualization An online system should provide both a map view and tabular view of the



benchmark data. There should be a time slider to follow data trends during the growing season.

Data quality (syntax) EO data should be cleaned from cloud and other atmospheric disturbances

Data types Image data / services for EO data, WFS webservices from LPIS for field boundaries, API for weather data, local data sources for crop/field characteristics and yield data

Data analytics Predictive analytics using machine learning; real-time streaming data analytics to alert and provide operational recommendations


Big Data needs to deliver added-value integrated services. The farmers will need actionable information. Creating an early warning service for yield or quality deviation a farmer can act upon will be challenging.


economy

The Earth observation data will be challenging in sense of dealing with the Volume of data. Whereas the farm data will be challenging in Variety of data coming from different sources, in different formats, using different semantics.



Privacy

● Farmer is in control about the level of data sharing ● Published farm data should be aggregated to avoid that

data is traceable to one specific farm ● .. more privacy requirements from the LTO privacy

baseline on sharing farm data Security

● Access for registered users only ● Registration reserved for involved stakeholders



Predictive analytics using machine learning based on big data is a general problem that extends to all the three sectors in DataBio and beyond.


www.databio.eu

https://watchitgrow.be/en

Note: <additional comments>



https://watchitgrow.be/en



Table 11: Agriculture pilot A1.3 Stakeholders and user stories

Who (type of user) I want to (can you perform

some task)

Why (achieve some goals)

farmers want to monitor the growth

their potato crop during the

season

to evaluate possible actions

like fertilizing and crop health

measures

farmers want to evaluate differences

in plant growth between and

within fields

to evaluate possible actions

like fertilizing and crop health

measures for specific areas

farmers want to evaluate weather

data, planting date and yield

for one or several years

to explain yield differences

and evaluate possible tactical

actions for future seasons

cooperative want a platform for growers to support their growers in

achieving better yields

cooperative want better yield predictions to improve their processing

capacity, planning and sales

cooperative want sufficiently good yield

predictions based on satellite

imagery

to reduce costs for field

sampling


The main motivations for this pilot is:

• to identify the potential of using of satellite data and machine learning to benchmark and optimize the yield and quality of the potato crops through the development of a monitoring and yield prediction model based on weather and EO data

• to identify the potential of yield prediction to improve capacity planning and sales forecasts.








This section presents the "Agriculture A1.3 Motivation view" view defined in the "Agriculture



Yield prediction of crops is committed to agronomists who go to reference fields to collect

samples. The activity implies recording of information on the status of crops to the

cooperatives office. Furthermore, agronomists assist the growers in deciding the timing of

each field operation, such as sowing/transplanting time, cultivation, spraying, irrigation,

harvesting.

The fields are scattered on a wide area and agronomists consume time to visit each field which

are in distances of several kilometres. There are therefore limitations to the number of

growers which each agronomist can co-ordinate.


This section presents the "Agriculture A1.3 Strategy view" view defined in the "Agriculture





Cost-effective, high-resolution solutions will be implemented in this pilot to substitute for the

conventional farm monitoring methods. The new technologies are expected to improve

insight about the impact of farming practices in order to sustainably improve productivity and

profits.



Three relevant KPIs that has been identified so far:

• Prediction quality: Evaluate the correctness of the model by testing on historic data that the system was not trained on.

• Revenue potential with potential yield vs. what happened: Quantify the value of the potential yield and the actual realization on historical data, and motivate the price of the service which would be profitable for the farmer.

• Improved ratio of realized yield to potential: Visualize the trend of yield optimisation over time.




internal project deliveries from work package 4 and 5 (WP4, WP5).










Figure 15:Agriculture pilot A1.3 BDVA reference model



Pilot 4 [A2.1] Big data management in

greenhouse eco-system 6.1 Pilot overview


The overall objective of the proposed pilot is to implement Genomic Selection (GS) Models in

support for greenhouse horticulture value chain, with particular focus on tomato. Genomics,

metabolomics and phenomics data will be integrated. Genomic selection is a new paradigm

in population genetic improvement that uses a larger number of genome-wide distributed

molecular markers to predict individual breeding values. The GS has demonstrated superior

performance in comparison with the methods used in breeding for quantitatively inherited

characters, i.e., phenotypic selection and quantitative trait loci approaches (Bernardo and Yu,

2007; Heffner et al., 2011; Lorenzana and Bernardo, 2009). The superiority of the GS strategy

is mostly associated with higher accuracy in predicting the individual’s genetic merit and the

shortening of a breeding cycle due to intercrosses driven by genetic predictions, which results

in higher genetic gain per unit of cost and time. These GS attributes are expected to have

wide-range implications in this pilot as the cost of cultivar development is going to be

reduced. Therefore, farmers can grow a better tomato variety sooner due to rapid variety

development and release, making more income. GS in greenhouse tomato is expected to

exert significant impact on the market potential and industry interests in this crop. Indeed,

tomato is among the top cultivated crops in greenhouses, with billions of euros turnover

worldwide. Tomato is considered one of the most nutritive solanum vegetables due to its high

content in sugars, vitamins and antioxidants and its consumption is steadily increasing.


The pilot will be run by a close partnership between CREA and CERTH, and will build upon

ongoing greenhouse horticulture breeding works in the Thessali Region, Greece, where

tomato materials are grown throughout the year in two greenhouses (2ha) and 2 walking

growth chambers. CREA and CERTH will share complete complementary tasks with the former

handling genomic predictions and selection, while the latter will be responsible for

phenomics, metabolomics, genomics and environmental datasets acquisition. The end users

of this pilot include farmers and farming cooperatives who currently grow crops following

standard farming practices and selection based on phenotypes, which is time and resource

consuming, with low resolution and efficiency. The end users therefore want cost-effective,

high-resolution solutions capable of expediting breeding activities in order to simplify

breeding scheme, shorten the time to cultivar development; selecting for genetic merit

estimated through genomic modelling in order to sustainably improve productivity and

profits. Within the DataBio framework, the services that are expected to be provided include

mainly farmer-customized estimates and selection for individual (plant, genotype) genetic

merit for a trait of interest, or several traits of interest for the farmer aggregated in Index.

Method



The pilot focuses mainly on implementing models based on a combination of genomic and

observational/environmental data to be applied in horticultural crop breeding in greenhouses

ecosystems in the Thessalia region, Greece. CERTH developed and applied high throughput

phenotyping and whole-genome genotyping systems to sustain modern high throughput

horticultural cultivar development. Current and historic information of breeding interest will

be gathered and used. On the other hand, CREA developed and applied high-resolution and

high throughput breeding approach based mainly on genomic models that demonstrated

superior predicting ability in several crop breeding systems including cereals and horticultural

crops. The two technologies at CREA and CERTH will be tandemly used in this pilot.

Specifically, parametric and nonparametric models will be implemented to anticipate the

possibility for their alternative use depending particularly upon the type of information

available. The potential of these models will be evaluated based on the predictive ability for

individual genetic merit. The implemented algorithms will differ in the assumptions of the

distribution of marker effects, and can therefore offer the possibility to account for different

models of genetic variation. Some models are expected to be suited to infinitesimal model

assumptions, others are expected to be best suited to finite loci model, whereas other models

extend Fisher’s infinitesimal model of genetic variation to accommodate non-additive genetic

effects. The different priors within the genomic models combined with the likelihood function

lead to joint posterior densities from which is possible to draw values from the fully

conditional densities using appropriate sampler. Phenotypic and whole-genome molecular

data will be fed into the models after the quality was checked and filters applied as

appropriate. One key breeding problem modelled includes predicting the performance of new

and unphenotyped lines/genotypes.


conventional breeding methods. The new technologies are expected to expedite breeding

activities, simplify breeding scheme, and shorten the time to cultivar development; selecting

for genetic and phenotypic merit estimated through genomic modelling in order to

sustainably improve productivity and profits. Specifically, metabolomics data (LC/MS/MS,

GS/MS, HPLC) will be collected. Other phenotypic data will include: (1) environmental indoor

air temperature, air relative humidity, solar radiation, crop leaf temperature (remotely and in

contact), soil/substrate water content; (2) environmental outdoor wind speed and direction,

evaporation, rain, UVA, UVB; (3) farm in-situ measurements (soil nutritional status testing),

farm logs (work calendar, technical practices at farm level, irrigation information), and farm

profile (static farm information, such as size, crop type, etc.). Genomic data will be derived

through Next Generation Sequencing protocols to generate whole or partial genome

sequences, transcriptome, and genotypic datasets. High quality phenotypic and genomic

information will be integrated in the process of genomic modelling. Models will be tested,

and those with higher predicting ability identified and implemented on a breeding scenario-

by-scenario basis. Contrary to conventional phenotype-based breeding, varietal selection in

this pilot will be executed based only on molecular marker information.




This pilot will build upon ongoing greenhouse horticulture breeding works in Thessalia

(Greece) and on peer-reviewed genomic prediction algorithms already in use at CREA. On the

other hand, large amounts of data were gathered at CERTH including metabolomics data

collected using LC/MS/MS, GS/MS, HPLC tools, whole genome, transcriptomes, and genotypic

datasets generated using Next Generation Sequencing protocols. Environmental indoor

collected data including Air temperature, Air relative humidity, Solar radiation, Crop leaf

temperature (remotely and in contact), Soil/substrate water content, while outdoor data

include wind speed and direction, Evaporation, Rain, UVA, UVB. Farm Data include in-situ

measurements (soil nutritional status testing), farm logs (work calendar, technical practices

at farm level, irrigation information,), and farm profile (Static farm information, such as size,

crop type, etc.). CERTH will provide NGS Sequencing machines (Illumina MiSeq and NextSeq),

HPLC, GC-MS, LC-MS, and access to the HPC system to perform primary analysis of NGS and

Metabolic data.

Benefit of pilot

The genomic data and relevant analytics will be directly useful to the involved Greek farmers

and farming cooperatives located in the Region of Thessali. The pilot is expected to directly

improve horticulture productivity and profits due not only to increased yields, but also to the

reduction of farming costs and risks, and the early variety release. Higher yields are expected

in virtue of high predicting ability of the genomic algorithms deployed to select for superior

genotypes. The breeding costs will be drastically cut in virtue of the adoption of simplified

breeding schemes, reduction of phenotypic evaluations, and shortening the breeding cycle.

Early variety release is an intrinsic GS property, particularly in virtue of genetic merit-driven

intercrosses. Therefore, farmers can grow a better variety sooner due to rapid variety

development and release, making more income. In addition, the results of this pilot will help

undertake further extension services endeavours and investigations to improve the whole

genome data collection and the predicting ability of the GS algorithms. The implantation of

accurate models under appropriate breeding scenarios, and the use of high quality NGS

sequencing machines and HPC system was anticipated to ensure that the proposed solutions

can be dependable and replicable to other crop types and market segments. Scientific papers

along with large amounts of data will be produced, preserved and made FAIR (Findable,

Accessible, Interoperable, Reusable) to further science and knowledge for the wellbeing of

man.



Use case title Genomic Selection in greenhouse horticulture




Author/company/email 1. Ephrem Habyarimana / CREA / [email protected]

2. Anagnostis Argiriou / CERTH / [email protected]


responsibilities

CERTH/CREA/Farmers, farming cooperatives, responsible for technological, farming and operational decisions.

Agroindustry/consumer, responsible for the manufacturing quality, sustainability, the business push, and market pull.

Goals Using environmental and whole genome biochemical data, and genomic prediction algorithms to predict horticultural species genetic merit upon which superior ideotype are identified and selected. Refer to the evaluation section for specific goals and KPIs.


Current

solutions

Compute(System) HPC Cluster, Workstations, PCs, Laptops (Xeon, CORE i7, etc.), standards OSs. Phenotypic data are processed and analysed. Sensor data and whole-genome biochemical information are not harnessed yet.

Storage RAID storage (approx. 100 TB) in the premises of CERTH/INAB

Networking Social networks: LinkedIn, Facebook, Twitter

Software Multiple individual algorithm systems, not integrated processing and display.


Data source 1. Biochemical: LC/MS/MS, GS/MS, HPLC to collect metabolomics data;

2. Genomic: Next Generation Sequencing protocols to generate genomic, transcriptomic, genotypic dataset

3. Environmental indoor: Air temperature, Air relative humidity, Solar radiation, Crop leaf temperature (remotely and in contact), Soil/substrate water content

4. Environmental outdoor: Wind speed and direction, Evaporation, Rain, UVA, UVB

5. Farm Data: In-Situ measurements: Soil nutritional status testing;

mailto:[email protected]




Farm logs (work calendar, technical practices at farm level, irrigation information,); Farm profile (Static farm information, such as size, crop type, etc.).

Volume (size) 10 to 100 Gb depending on the compression level.

Velocity

(e.g. real time)

Variety


Great variety: genomic, phenomic, metabolomics, environmental public data and analytics data.




analysis,

action)


Visualization Visualization of crop monitoring output and GS predictions.

Data quality (syntax) Data validity filtering w.r.t. completeness and dependability.

Data types Experimental: whole genome genotypic data, metabolomic and phenomic (lab) data. Mainly ASCII files (fastq and text files).

Observational: phenomics (field), sensor data, environmental (Environmental indoor: air temperature, air relative humidity, solar radiation, soil/substrate water content).

Data analytics Predictive analytics genetic merit and generation of selection indices, transcriptomics and metabolomics analytics, genomic assembly and annotation algorithms.


There is a need for: (1) cost-effective, high-resolution solutions capable of expediting breeding activities in order to simplify breeding scheme, shorten the time to cultivar development; selecting for genetic merit estimated through genomic modelling in order to sustainably improve productivity and profits; (2) farmer-customized GS for a trait of interest, or several traits of interest for the farmer aggregated in Index; (3) closing the gap



between agricultural business planning and the responsible and sustainable maximization of the profit deriving mainly from increased crop productivity and efficiency of resource use, reduced uncertainty of management decisions, accounting for environmental standards and regulations.


Delivering content and services to various computing platforms from

Windows desktops to Android and iOS mobile devices



Greenhouse owner and geolocalization are highly sensitive, should be anonymized



Genomic prediction analytics based on environmental data, phenomics, and whole-genome biochemical information are universal solutions with domain agnostic applications.


www.databio.eu


Note:


The end users are farmers, farming cooperatives, agroindustry and research and

technological institutions. CERTH implemented methodologies and know how to produce and

process useful: (1) experimental data: whole genome genotypic data, metabolomic and

phenomic (lab) data; (2) observational data: phenomics (field), sensor data, environmental

indoor and outdoor. On the other hand, CREA developed GS algorithms that demonstrated

high predicting ability in different crops including Solanaceae species, the family which

tomato belongs to.

Table 13: Agriculture pilot A2.1 stakeholders and user stories


Farmers, farming

cooperatives

want cost-effective, high-

resolution breeding solutions.

To select the best possible and

affordable varieties without

confounding effects of the

environment, which guarantees

consistent and dependable

yields.




Farmers, farming

cooperatives

want a simplified and less time-

consuming breeding scheme

to efficiently shorten the time to

bring the new varieties to

market and maximize the profits

over the longer variety’s life

cycle.

Farmers, farming

cooperatives

want to efficiently and accurately

breed for varieties with high

nutritional (health-promoting

natural biochemicals) and trading

values

to responsibly target, promote

and/or extend market segments

that are ready to pay for

additional quality of the

produce.


The main motivations for this pilot are

• to predict the performance of unphenotyped tomato materials in grasshouse agroecosystems using molecular data information.

• to empirically demonstrate the benefit of using genomic data in terms of increasing genetic gain by unit cost and time




described using the ArchiMate standard. It lists the views and nomenclatures composing the

view point.


This section provides the "Agriculture A2.1 Motivation view" view defined in the "Agriculture





Crops are grown following standard farming practices and selection based on phenotypes,

which is time and resource consuming, with low resolution and efficiency. On the other hand,

frequent and regular biochemical analyses to sustain conventional breeding add greatly to

the cost of breeding activities with consequent high costs for the developed cultivars.


This section provides the "Agriculture A2.1 Strategy view" view defined in the "Agriculture






conventional breeding methods. The new technologies are expected to expedite breeding

activities, simplify breeding scheme, and shorten the time to cultivar development; selecting

for genetic and phenotypic merit estimated through genomic modelling in order to

sustainably improve productivity and profits.




• Model accuracy: performance prediction accuracy showing an acceptable error rate including when the testing and training sets are genetically distant.

• Revenue potential with alternative cropping strategy vs. what happened: Quantify increased revenue potential GS versus phenotypic selection.

• GS take-up: number of vegetable growers adopting GS approach. This is technology transfer and takes time to establish; for this project, a baseline will be measured first, then followed-up by monitoring usage after system deployment.












indicated in the list using the component name.




Pilot 5 [B1.1] Cereals and biomass crop 7.1 Pilot overview


The pilot aims to develop an accurate "irrigation maps" and "vigor maps" (combining EO data

and sensors data) which allows mapping different areas in Spain - beginning in Castilla -and

set up an informative and management system for early warning of inhomogeneity.

Those new services will be dedicated to the analytical and accurate finding of heterogeneities

in crops related to irregular irrigation, mechanical problems affecting irrigation systems,

incorrect distribution of fertilizers or any other sources of inhomogeneity that could explain

crops growing differences. This Service will be a powerful preventive tool for general farmers

and land owners in order to avoid production losses.


This service aims to provide information for precision agriculture, mainly based on time series

of high resolution (Sentinel-2 type) satellite images, complemented with sensor data and, in

some specific cases, with RPAS data. The information can be used as input for farm

management (operational decisions, tactical decisions). Information layers may include:

Vegetation indexes (NDVI, Normalized green red difference index) and derived anomaly

maps.

This service will offer cost saving for farmers communities due to a better quality

management in agricultural zones, especially focused on irrigated crops. Monitoring and

managing irrigation policies and agricultural practices will offer meaningful water and energy

saving. Besides this, fertilizers control and monitoring can produce, eventually, a prominent

economic saving per year and hectare. This better management of hydric and energetic

resources is also related to Green-house effect gases reduction, directly linked to better

environmental conditions in agriculture.

Table 14: Summary of pilot B1.1 (ISO JTC1 WG9 use case template)

Use case title B1.1 TRAGSA

Vertical (area) Cabreros del Río, Castile, Spain

Author/company/email

Sofía Iglesias/TRAGSA Group/[email protected]

Jesús Estrada/TRAGSA Group/[email protected]

Actors/stakeholders and their

roles and responsibilities

TRAGSA

Goals Develop an accurate “Irrigation maps” and “Vigor maps”

mailto:Group/[email protected]



Use case description

Combining EO data and sensor data which allow mapping different areas in Spain and setting up an informative and management system for irrigation and early warning of heterogeneity or malfunction of irrigation systems and devices.

Current

solutions

Compute(System)

Dedicated server

2 processors Intel(R) Xeon(R) CPU E5606 @ 2.13GHz, 8

cores with 96 GB ECC of Memory

Storage Dedicated server

Hard Disks: 2 disks - 2 TB (RAID). Total: 2 TB

Networking 250 MBps

Software OS: Debian 8.8

Apache web server 2.4.10

Tomcat: 7.0.56 and 8.0.14

R: 3.3.3.

PostGres: 9.4

MySQL: 5.5.55

Python: 2.7.9

Virtuoso: 07.20.3212


Data source (distributed/cen

tralized)


Centralized: Remote sensing data as Sentinel 2B data provided by ESA at Sentinel Data Hub (https://cophub.copernicus.eu/)

Orthophotos (Spanish Coverage). RGB and NIR bands provided by Spanish National Geographic Institute at http://centrodedescargas.cnig.es/CentroDescargas/catalogo.do

Surveys and field data (Confidential)

Likely, GEOSS open sources available at GEOSS portal (http://www.geoportal.org) will be used as testing and validation data

Distributed/local:

LPIS system is provided in a distributed way by NUTS2-Level Administration. Castile LPIS system is accessible at

http://ftp.itacyl.es/cartografia/05_SIGPAC/2017_ETRS8

9/Parcelario_SIGPAC_CyL_Provincias/

https://cophub.copernicus.eu/

http://centrodedescargas.cnig.es/CentroDescargas/catalogo.do

http://centrodedescargas.cnig.es/CentroDescargas/catalogo.do

http://www.geoportal.org/

http://ftp.itacyl.es/cartografia/05_SIGPAC/2017_ETRS89/Parcelario_SIGPAC_CyL_Provincias/

http://ftp.itacyl.es/cartografia/05_SIGPAC/2017_ETRS89/Parcelario_SIGPAC_CyL_Provincias/



TRAGSA Drones produce JPEG and LAS files using thermal and multispectral sensors. More information is available in the DataBio deliverable D6.2 – Data Management Plan

Volume (size) Remote sensing data as Sentinel 2B have an average size of TB per year. LPIS Spanish system has a size of hundreds of Gigabytes, likewise the Spanish Orthophoto project (PNOA).

RPAS data has a size of Tb per year.

Field data information has a size of Mb per year.

Velocity

(e.g. real time)

Sentinel 2B has the highest updating rate within Pilot Information sources (5 days). All external sources have a yearly updating ratio.

Variety


The formats to be used will be imagery and terrain models.


Agricultural information, typically, depends on seasons. Highest variability rate is few days (2-3)

Big data science

(collection, curation,

analysis,

action)

Veracity (Robustness

Issues, semantics)

All data sources are official and trusted ones: European Space Agency (ESA) and Spanish Public Administration.

Visualization Standard imagery visualization services. Spanish Public Administration usually provides WMS services for information visualization.

Data quality (syntax)

Despite of the data providers are supposed to produce good quality information, all datasets are processed by TRAGSA to produce improved images. Specifically, orthophotos will be transformed by an orthorectification method developed under WP5.

Data types JPEG or JPG2000 for images, .LAS for terrain models. Text document for surveys and field data.

Data analytics Anomaly maps will be used to answer questions about

distribution of plant protection products or correct and

consistent growing of the crops.

Big data specific

challenges (Gaps)

TRAGSA Group is currently able to process, manage and submit results about a high number of parcels. Processes will be scaled to provincial, regional and national levels.

Big data specific

TRAGSA Group is currently able to process, manage and submit results about a high number of parcels. Processes will be scaled to provincial,



challenges in bio-economy

regional and national levels. In regard to Farmers Communities, the aim is to provide similar services to more than ten new different Communities.



No datasets of this pilot are considered sensitive or related to personal data.

Highlight issues for generalizing

this Use case (e.g. for ref.

architecture)

Main technical problems are related to data storage and resulting maps transmission.

More information

(URLs)

ESA: https://sentinel.esa.int/web/sentinel/sentinel-data-access

PNOA: http://www.ign.es/web/ign/portal/obs-portal-pnoa

GEOSS: https://www.earthobservations.org/index.php

Note: No additional comments.

7.2 Pilot case definition Location: Cabreros del Río, Leon, Spain. Potentially more Farmers Communities after the

second year of the project.

Supported by: “Ribera del Porma” Farmers Community

Area size: “Ribera del Porma” Farmers Community: 24.270 ha


Table 15: Agriculture pilot B1.1 stakeholders and user stories

Who (type of user) I want to (can you

perform some task)


Farmers, irrigations communities

(end users)

To know current

status of crops

vigor in specific

parcels.

To visualized updated (and

preventive maps) to avoid future

risks.

Farmers, farming cooperatives

(end users)

To know accurate

temporal evolution

of water demand.

To visualized updated (and

preventive maps) to concentrate

efforts and save field visits.

https://sentinel.esa.int/web/sentinel/sentinel-data-access

http://www.ign.es/web/ign/portal/obs-portal-pnoa

https://www.earthobservations.org/index.php



Cost Savings.

Farmers, farming cooperatives

(end users)

To better define the

places for RPAS

visits, in the case

they were

necessary.

To concentrate efforts and save

costs and resources.

GIS Technician and IT teams

(TRAGSA, as manager of Irrigation

Community, or SMEs and other

third parties)

To produce

updated irrigation

and vigor maps.

To provide automated, high-quality,

highly-updated and cheap products

for:

- manage the water supply

- check the available

information of the parcels

- manage the crops


Therefore, in this pilot will be developed the following:

• Agricultural Heterogeneity Analysis for analytical and accurate finding of heterogeneities in crops related to irregular irrigation, mechanical problems affecting irrigation systems, incorrect distribution of fertilizers or any other sources of inhomogeneity that could explain crops growing differences.

• Irrigation services improved using the information provided by the previous point

These services will be a powerful preventive tool for general farmers and landowners in order

to avoid production losses.

7.3 Pilot modelling with ArchiMate

7.3.1 Agriculture pilot B1.1 motivation view

This service will offer Public Administrations, Farmers Communities, Rural owners and

Farmers a remote plant disease diagnosis and assessment based on the processing of Satellite

images. Besides this, this service will monitor irrigation systems performance in order to show

accurate crop strength maps. Efficient soil fertilization procedures based on precise measures

will be a direct consequence of Satellite Data use. These products will be provided through

web services.

Below the "Agro Pilot B1.1 Motivation view" view defined in the "Agro Pilot 1.3.1 View Points"

view point is presented.



Figure 20: Agriculture Pilot B1.1 TRAGSA Motivation view

7.3.2 Agriculture pilot B1.1 strategy view

Irrigated agriculture is responsible for compensating the agricultural trade balance in

countries with agro-climatic conditions that are unfavourable to agriculture. Water shortages

make it necessary to have powerful systems that help provide the plant with the minimum

water needed to ensure high productivity crops. Therefore, the use of advanced technologies

(remote sensing) is very important in order to maximize the benefits of irrigated agriculture

at the lowest cost. On the other hand, the use of these technologies also provides accurate

information to anticipate health problems in crops and to prevent them in advance, which

allows preserving most of the crop in optimal sanitary conditions.

Below the "Agro Pilot B1.1 TRAGSA Strategy view" view defined in the "Agro Pilot 1.3.1 View

Points" view point is presented.



Figure 21: Agriculture pilot B1.1 Strategy view




• Surface processed, from the original 24.270Ha.

• Field visits saved and their economic impact.

• Economic improvements: water, energy and other resources savings due to better information on crops status.

In the current development state of the pilot is not possible to define accurate quantitative

figures for the previous KPIs. In the following and intermediate reports, these KPIs will be

improved from the current explanatory state to a more rigorous definition.




Figure 22: Agriculture pilot B1.1 initial roadmap



7.5 Big data assets

Figure 23: Agriculture pilot B1.1 BDVA reference model



Pilot 6 [B1.2] Cereals and biomass crop_2 8.1 Pilot overview


Farming 4.0 [REF-08] holds the key for meeting the rising demands for increased yield at all

farm levels. The agricultural sector faces unprecedented challenges to cope with recent

trends (precision farming, automation, consolidation, professionalism, labour shortage, etc.)

that shape a continuously evolving farming environment. It is evident, that “producing more

with less” is more than a will, but rather a “must-have” requirement for future farmers, with

the decision-making process playing a vital role in farm profitability. Based on a review of 234

studies that were published from 1988 to 2005, precision agriculture was determined to be

the reason for a profit increase in 68% of the cases. From an economic point of view, farm

profitability comes as a result of following different management strategies. In a market that

sometimes is struggling to remain afloat, farmers are aiming to use technological

advancements for cost reduction primarily, without any significant impact on their

production. Thereby, a major expected benefit from precision agriculture derives from the

optimization of inputs and of farm management (leading to cost reduction) with farm size

being a critical parameter for farm profitability.


The main focus of this pilot is to offer smart farming services dedicated for arable crops, based

on a set of complementary monitoring technologies. Smart farming services are offered as

irrigation advices through flexible mechanisms and UIs (web, mobile, tablet compatible). The

pilot will target towards promoting the adoption of technological tools (IoT, Big Data analytics,

EO data) and collaborating with certified professionals to optimize farm management

procedures. NP and GAIA Epicheirein will support the activities for the execution of the full

life-cycle of the pilot.

Table 16: Agriculture pilot B1.2 overview of pilot activities

Pilot Site

Location Elassona, Greece

Area Size 2500ha

Targeted Crops Maize

End-Users Farming Cooperative, Agronomists



Method

This pilot is targeting towards providing a smart farming services to the farmer utilizing

available precision agriculture techniques. The services will be provided as advices which need

many prerequisites and primary material in order to be accurate. Data is the raw material and

there are three (3) different means of collecting data, which will be exploited within the pilot

activities. Data directly from the field, collected from a network of telemetric IoT stations

called GAIAtrons; remotely with image sensors on in-orbit platforms; and by monitoring the

application of inputs and outputs in the farm (e.g. in-situ measurements, farm logs, farm

profile). Every data source has unique characteristics with relevant impact on the very content

of this data. Field sensing provides real-time accurate direct measures of many physical

parameters of the soil, atmosphere microclimate of the field crop and plant with temporal

continuity. Remote sensing provides indirect measures of some physical properties of plants

and soil with spatial continuity in medium to large spatial scale. Combining this information

can provide a good knowledge of the most important physical parameters of soil,

microclimate, plants and water (which are all the environmental resources which govern

farming) in both spatial and temporal dimensions. Monitoring the application of inputs and

outputs on the farm is a data element that is necessary to assess the correctness of the given

advice and use it as feedback to improve the system over time. This pilot will combine

advanced data handling techniques (i.e. assimilation, fusion and spatio-temporal

interpolation) to transform the collected data into actionable irrigation advice. In order for

this advice to better reflect the actual situation at a given field, we will seek to incorporate

the human experience of the farmer or certified advisors.










optimal big data management methodologies. Finally, through flexible work calendars, NP

has collected more than 120000 records related to work plans of the farmers that can be used

in the context of the pilot activities.

Benefit of pilot

The pilot is expected to have a direct impact on farm resource management in a crop type

with significant economic value, from an economic perspective in Greece. This will provide a

proof of concept that the proposed solutions can be replicated to other crop types and market

segments in the near future. The holistic approach that is being proposed will significantly

improve the capacity of the responsible partners in providing smart farming advisory services.



In addition, it would lead to improvements in a) NP’s GAIA cloud’s stability, availability,

security, interoperability and overall maturity, b) NP’s GAIABus DataSmart functionality in

terms of real-time analytics, data stream and decision support processes, multi-temporal

object-based monitoring, cloud-based services that integrate earth observation with image

processing, machine learning and spatial modelling, c) advancing the current system by fusing

telemetry IoT stations’ data with remote sensing data and incorporating advanced

visualization capabilities.



Use case title Cereals and biomass crops




responsibilities

● Farming Cooperative, responsible for performing farming activities

● Agronomists, involved in providing relevant and up-to-date advices to the farmers

Goals Provide smart farming advisory services (focusing on irrigation), based on a set of complementary monitoring technologies, in order to increase farm profitability and promote sustainable farming practises.


Current

solutions

Compute(System) System is based on IoT data, farm logs, work calendars and in-situ measurements. Expert knowledge is provided through static scientific models that offer insight about optimal farm management.

Storage All available data are stored in a cloud infrastructure.

Networking Web-based UIs and dashboards available for monitoring farm activities.

Software Real-time analytics, data stream processes and decision support system





Centralized: Field sensing data from GAIAtrons, Remote sensing (Earth observation) data, Farm data

Volume (size) ● ~1 TB/year for remote sensing data, including raw data and extracted biophysical and vegetation indices for the pilot areas


● Hundreds of thousands of records related to farm activities/profiles/measurements

Velocity

(e.g. real time)


Variety


Field Sensing: Soil temperature, humidity (multi-depth), ambient temperature, humidity, barometric pressure, solar radiation, leaf wetness, rainfall volume, wind speed and direction





analysis,

action)


Need for a system that can constantly provide relevant and up-to-date advices to its end-users

Visualization Spatio-temporal information visualization for improving farm management and facilitating the decision-making process

Data quality (syntax) The quality of field sensing data is being controlled by several filtering, outlier detection and stream processing mechanisms. The integrity of remote sensing data quality is being



assessed by a hash check upon product download.


Data analytics Descriptive and prescriptive analytics for the provision of irrigation advices.


There is a need for smarter fusion of the heterogenous data types that are being collected towards providing accurate insights. To this end, it is important to explore mechanisms that could combine raster and vector data at parcel level (polygon) and station level (point).


In order to facilitate the adoption of the big data technologies by the farmers, imposed barriers in data visualization should be encountered (e.g. give more emphasis to vector data, improvement of the aggregation mechanism (drill down, zoom in, roll up, zoom out)).






EO data management mechanisms can be exploited for other use cases where EO data might provide valuable insights




Farmer As a farmer I want to follow an irrigation plan based on my crop needs and farm characteristics.

Increase my profits following sustainable agriculture practices

Agronomists As an agronomist I want to have a comparative advantage in a highly

Increase my profits by providing better advices based on evidences,



competitive market and to offer the best possible services to my clients

well-established arguments and scientific knowledge.



• to raise the awareness of the farmers, agronomists, agricultural advisors, farmer cooperatives and organizations (e.g. group of producers) on how new technological tools could optimize farm profitability and offer a significant advantage on a highly competitive sector.

• to promote sustainable farming practises over a better control and management of the resources (water).







8.3.1 Agriculture pilot B1.2 Motivation view

This section provides the "Agro Pilot B1.2 Motivation view" view defined in the "Agro Pilot

1.3.1 View Points" view point.

Figure 24: Agriculture pilot B1.2 Motivation view

Motivation elements are used to model the motivations, or reasons, that guide the design or

change of an Enterprise Architecture. It is essential to understand the factors, often referred

to as drivers, which influence other motivation elements. They can originate from either

inside or outside the enterprise. Internal drivers, also called concerns, are associated with

stakeholders, which can be some individual human being or some group of human beings,

such as a project team, enterprise, or society. Examples of such internal drivers are customer

satisfaction, compliance to legislation, or profitability.



8.3.2 Agriculture pilot B1.2 Strategy view

This section provides the "Agro Pilot B1.2 Strategy view" view defined in the "Agro Pilot 1.3.1

View Points" view point.


The immediate decision support system is built on top of a data collection and distribution

system. The data collection and distribution system is used to collect sensor data from the

on-board systems and makes them available in a single system. The data distribution system

ensures that the decision support system only interface with a single system, instead of

multiple sensors. The decision support system presents the data from the data distribution

system and collects them in an internal storage system for presentation of current

performance vs. historic performance.





One relevant KPI has been identified so far, namely:

• %Reduction potential in operational costs for performing the same farming activities following the advisory irrigation services vs what would be the operational costs following standard farming practices based on historical data.

• %Decrease in the environmental footprint, through better management of resources.















Pilot 7 [B1.3] Cereal and biomass crops_3 9.1 Pilot overview


This pilot aims mainly at implementing remote sensing and proximal sensor network big data

technologies for biomass crop monitoring, predictions, and management in order to

sustainably increase farming productivity and quality, while at the same time, minimizing

farming and environment associated risks. Biomass crops of interest include biomass

sorghum, fiber hemp and cardoon which can be used for several purposes including,

respectively, biofuel, fiber, and biochemicals, with a high macroeconomic impact. Producing

biofuel from plant biomass is an attractive alternative to fossil sources, not only because the

latter are non-renewable and environmentally harmful, but also because of the pressing issue

for nations to get independent of foreign energy sources. Several countries worldwide have

initiated programs to convert biomasses into biofuels and, in European countries, dedicated

biomass crops are being increasingly developed. The pilot is therefore well aligned with the

lines of action and the objectives and action plans of the bioenergy supply chains of Italy

(https://www.politicheagricole.it/) and the European Union. In 2011, the European

Commission included the biomasses for agroindustrial use among the six highly innovative

markets to be promoted in the near future; this concept was repeated by the Seventh

Framework Program for Research and Technological Development and by Horizon 2020

program.


The pilot was designed to offer precision farming services consisting in different crop

monitoring and management technologies specifically geared to biomass crops with

particular interest in sorghum, fiber hemp and cardoon. Offered smart farming services

include (1) Biomass crop (sorghum/hemp/cardoon) monitoring using remote sensor and real-

time streaming sensor network big data, (2) crop growth and yield modelling, (3) early

warning: yield and quality deviation, threatening events (biotic/abiotic stress, threshold

alerts) for crop growth and development, (4) visualization: processed data and model results

are published in an intuitive way and viewable on the computer, smartphone or tablet, (5)

GPRS connectivity, (6) cloud based crop management infrastructure, (7) web portal cloud

solution. The pilot secured adhesion of fifteen farmers and/or farming cooperatives with

more than 120 ha. The pilot will run intensive extension services to promote the adoption of

the new technologies (remote sensing, IoT). CREA will work on sorghum and fiber hemp, while

Novamont will work on cardoon. VITO will support remote sensing technologies, while CREA

will support proximal streaming sensor network data technologies. The following table

provides an overview of the pilot activities.



Table 19: Agriculture pilot B1.3 Overview of pilot activities

CREA sorghum pilot CREA fiber hemp

pilot

NOVAMONT

cardoon pilot

Location 24 sites in Emilia

Romagna, Italy

3 sites in Emilia

Romagna and

Veneto, Italy

4 sites in North and

South-Western

Sardinia, Italy

Area Size 120ha 6ha 65ha

End-Users CREA, farmers,

farming

cooperatives

CREA, single farmer NOVAMONT

Current Situation Crops are monitored based on visual observations, which is time

consuming. Furthermore, visual observations are not dependable

as they cannot allow accurate prediction of yields or the

identification of within field phenotypic variations.

Biomass sorghum, fiber hemp and cardoon were included in the pilots as motivated under

the above introduction section.

Method

This pilot will use satellite imagery (Sentinel-2) and/or telemetry IoT to monitor biomass crops

and predict yields. The pilots will be run by CREA and Novamont in collaboration with VITO.

Main crops will include sorghum and hemp for CREA, and Cardoon for Novamont. VITO will

use satellite data to monitor the crops and will develop yield models. Telemetry IoT

technology will be implemented by CREA on 5 sorghum piloting sites in Anzola experimental

station (Bologna, Italy).

Specifically, as part of pilot innovative solution, an online platform will be used to provide

satellite imagery, weather and soil data and yield predictions. VITO, in collaboration with a

number of Belgian partners, has developed a web application “WatchITgrow®” for potato

monitoring and yield prediction in Belgium. The existing WatchITgrow® application will

“filled” with satellite, weather and soil data for the Italian pilot sites. To be able to provide

yield forecasts for sorghum, hemp and cardoon developments are needed. The field data that

will be collected by CREA and Novamont will be input for setting up yield and phenology

models using machine learning techniques and will support continuous system learning.

The farmer and pilot owners can use the satellite imagery (biomass index, 10m resolution) to

monitor and benchmark their field productivity potential relative to production levels

achieved in the region. After one year, yield prediction can be implemented using the test

data from year one, and other historical data (when available).



Telemetry IoT stations will transmit data to the cloud infrastructure in the process of crop

monitoring, biotic and abiotic stress diagnostic, alert and operational recommendations using

cloud-based crop management analytics including web portal cloud solution.

The telemetry IoT is designed to provide smart farming services. These include but are not

limited to remote monitoring capability (crop, soil and environmental properties), viewing (on

web, PC and mobile devices) and evaluating real-time data instantaneously for better decision

to improve yield and quality, conserve resources, and increase profits, cloud-based data

analysis to deliver summarized data to pilot manager (user), notify users via text and/or email

when critical thresholds have been breached. In the framework of this pilot, the telemetry

IoT will serve as a means to assess the usefulness of shifting from field-bound sensors data

towards cost-effective remote sensing (satellite data) solution.


Historic phenotypic data were recorded over more than thirty years in some crops

(Habyarimana et al., 2016). Telemetry IoT hardware and software were anticipated, with

GPRS connectivity, wireless sensor network (WSN), cloud based crop management

infrastructure, and web portal cloud solution. The WSN has a customizable time step for data

collection (1-60 minutes) and can collect big data: air temperature, air moisture, solar

radiation, leaf wetness, rainfall, wind speed and direction, soil moisture, soil temperature, soil

EC/salinity, PAR, and barometric pressure.

ESA’s high-frequency Sentinel satellites deliver a constant stream of information to identify

changes in crops and soil. The satellites provide cost-effective information on crop growth

and development every 5 days at field scale (up to 10m detail), which allows farmers,

agronomists, processing companies, to make better, more informed decisions.

Benefit of pilot

The availability of an online monitoring system whereby the farmer or agronomist has a global

view on the whole production area from space will facilitate the organization of field visits,

farm and business operations.

The pilot is expected to directly improve farm productivity and profits due not only to

increased yields, but also to the reduction of farming costs and risks. Productivity

improvements will mainly derive from: (1) timely efficient farming operations and responding

to biotic and abiotic stresses and IoT alerts, (2) rationalization of agricultural inputs, and (3)

accurate prediction models.

The results of this pilot will help undertake further extension services endeavours and

investigations to improve remote sensing and telemetry IoT technologies. The large sample

of piloting sites was anticipated to ensure that the proposed solutions can be replicated to

other crop types and market segments in the near future. Scientific papers along with large

amounts of data will produced and made FAIR (Findable, Accessible, Interoperable, Reusable)

to further science and knowledge of man.





Use case title Biomass crops monitoring and performance predictions


Author/company/email 1. Ephrem Habyarimana / CREA / [email protected]

2. Sara Guerrini / Novamont / [email protected]

3. Isabelle Piccard / VITO / [email protected]


responsibilities

CREA/Farmers Cooperatives, responsible for technological, farming and operational decisions.

Novamont/Farmers, responsible for technological, farming and operational decisions.

Agroindustry/consumer, responsible for the manufacturing quality, sustainability, the business push, and market pull.

Goals Using satellite imagery and/or telemetry IoT to monitor biomass crops and predict yields. Refer to the evaluation section for specific goals and KPIs.


Current

solutions

Compute(System) PCs, Laptops (CORE i7), standards OSs. Multi-year/environment data are processed and meta-analyzed. Vegetation indices, satellite and streaming telemetry IoT big data are not harnessed yet.

Storage Disk drives and SSDs + backup

Networking Social networks: LinkedIn, Facebook, Twitter

Software Multiple individual algorithm systems, not integrated processing and display.


Data source CREA will run 24 piloting sites (>120 ha) in Emilia Romagna, while Novamont will run 4 piloting sites (65 ha) in North and South-Western Sardinia, Italy.

An online platform will be used to provide satellite imagery with a 10m resolution (Sentinel 2A + B, possibly complemented with coarser






resolution data and proximal sensor data to overcome cloud problems and provide ground-truth data), weather data, soil data and yield predictions.

Telemetry IoT will be designed to provide smart farming services. These include but are not limited to remote monitoring capability (crop, soil and environmental properties), viewing (on web, PC and mobile devices) and evaluating real-time data instantaneously for better decision to improve yield and quality, conserve resources, and increase profits, cloud-based data analysis to deliver summarized data to pilot manager (user), notify users via text and/or email when critical thresholds have been breached.

Volume (size) Hundreds of terabytes per year when all sources of data are considered.

Velocity

(e.g. real time)

Satellite data: Sentinel-2A+B images are acquired with a time step of 5 days. The images are pre-processed and distributed by ESA within 24 hours after acquisition. Further processing by VITO starts as soon as the images are available from ESA. Generally, the final information products become available for the end-users between 24 and 48 hours after image acquisition.

Telemetry IoT data: Time step for data collection is customizable, 1-60 minutes; big data: air temperature, air moisture, solar radiation, leaf wetness, rainfall, wind speed and direction, soil moisture, soil temperature, soil EC/salinity, PAR, barometric pressure.

Phenotypic data are collected each cropping season.

Variety


Great variety. (1) Satellite: imagery, multispectral data, indices (soil, water, vegetation, biophysical), (2) Telemetry IOT: air temperature, air moisture, solar radiation, leaf



wetness, rainfall, wind speed and direction, soil moisture, soil temperature, soil EC/salinity, PAR, barometric pressure. (3) analytics and phenotypic data.




analysis,

action)


Need to have tools to produce and process ground-truth data for satellite data calibration.

Visualization Visualization of crop monitoring output at least bi-weekly during the cropping season, indices and predictions; real-time monitoring output, alerts, and recommendations.

Data quality (syntax) Data validity filtering w.r.t. completeness. Data fusion and modelling of heterogeneous data (EO data, telemetry IoT data, field data)

Data types Imagery, graphics, vector, numbers, analytical results, measurements, metadata, geolocations, spectra, time series.

Data analytics Predictive analytics for the development of data-driven yield models; predictive feedback (monitoring), real-time streaming data analytics to alert and provide operational recommendations using cloud-based crop management analytics including web portal cloud solution.


There is a need for: (1) improving analytic and modelling systems that provide reliable and robust statistical estimated using large size of heterogeneous data; (2) closing the gap between agricultural business planning and the responsible and sustainable maximization of the profit deriving mainly from increased crop productivity and efficiency of resource use, reduced uncertainty of management decisions, accounting for environmental standards and regulations.


Delivering content and services to various computing platforms from

Windows desktops to Android and iOS mobile devices





Farm owner and geolocalization are highly sensitive, should be anonymized



Real-time streaming data analytics and predictive analytics using machine learning for crop monitoring and developing yield models based on big data are universal solutions with domain agnostic applications.


www.databio.eu


Note:


The end users are farmers, farming cooperatives, agroindustry and research institutions. VITO

will use satellite data to monitor the crops and will develop yield models. Telemetry IoT

technology will be implemented by CREA to provide real-time analytics solutions. Combining

crop monitoring, yield models and real-time analytics will provide localized descriptive,

prescriptive and predictive plans to improve farming operations and productivity.

Stakeholders and user stories are summarized in the table below.



Farmers, farming

cooperatives,

research

institutions

want to monitor the growth their

biomass crops throughout the

season

to evaluate the right time for

possible actions like fertilizing,

irrigation, crop protection, and

harvesting, optimize field

operations and save money and

time.

Farmers, farming

cooperatives,

research

institutions

want to evaluate differences in

plant growth between and within

fields

to evaluate possible actions like

fertilizing, irrigation, and crop

protection measures for specific

areas

Farmers, farming

cooperatives,

research

institutions

want to evaluate weather data,

planting date and yield for one or

several years

to explain yield differences and

evaluate possible tactical actions

for future seasons




Farmers, farming

cooperatives

want better yield predictions to improve their processing

capacity, harvest and sales (stock

market) planning

Farmers, farming

cooperatives,

research

institutions

want empirical comparison

between remote and proximal

sensor-based solutions.

to identify the cost-effective

decision support tool.


The main motivations for this pilot are:

● identifying the potential for using satellite data and machine learning for biomass crop

monitoring and the development of yield models.

● Evaluating the comparative importance between the use of proximal wireless sensor

network data and satellite data in biomass farm telemetry







This section presents the "Agro Pilot B1.3 Motivation view" view defined in the "Agro Pilot






This section presents the "Agro Pilot B1.3 Strategy view" view defined in the "Agro Pilot 1.3.1





Three relevant KPIs have been identified so far, namely:

• %Increase in farm productivity following the advisory irrigation, fertilization, pest/ disease management services vs what would be the revenue following standard farming practices based on historical data: Quantify %increase in farm productivity for all three crop types.

• %Decrease in operational costs for performing the same farming activities (through better management of resources) following the advisory irrigation, fertilization, pest/disease management services vs what would be the revenue following standard



farming practices based on historical data: Quantify %Decrease in operational costs for all three crop types.

• %Decrease in fresh water and fertilizer usage following the aforementioned advisory services




internal project deliveries from work packages 4 and 5 (WP4, WP5).




9.5 Big data assets The diagrams below summarize Big Data technology components used in this pilot using the


indicated in the list using the component name.

Figure 31: Agriculture pilot B1.3 BDVA reference model for IoT



Figure 32: Agriculture pilot B1.3 BDVA reference model for Satellite data



Pilot 8 [B1.4] Cereals and biomass crops_4 10.1 Pilot overview


The pilot aims to develop a platform for mapping of crop vigor status by using EO data

(Landsat, Sentinel) as the support tool for variable rate application (VRA) of fertilizers and

crop protection. This includes identification of crop status, mapping of spatial variability and

delineation of management zones.

The Czech Republic has a specific land use defined by the highest average holding area in EU

(over 130 ha per farm). The national statistical report of agriculture sector ("Green Report

2015", Ministry of agriculture, Czech Republic) shows that there are 637 farms with acreage

of managed land over 1000 ha which together cultivate 50.3% of agricultural land in Czech

Republic. Also, there is known higher average size of fields. Statistical evaluation of the size

of land parcels in LPIS shows that 60% of arable land is located within the fields with the area

over 20 hectares. Higher diversity of the relief and pedoclimatic conditions in combination

with the size of land blocks occur in visible heterogeneity of land. This leads to an increased

interest in the precision farming practices and technologies for site-specific crop

management, where remote sensing (EO) plays a crucial role.

The pilot farm Rostenice a.s. with 8.300 ha of arable land represents a bigger enterprise

established by aggregating several farms in past 20 years. Main production is focused on the

cereals (winter wheat, spring barley, grain maize), oilseed rape and silage maize for biogas

power station. Crop cultivation is under standard practices, partly conservation practices is

treated on the sloped fields threatened by soil erosion. Over 1600 ha is mapped since 2006

by high density soil sampling (1 sample per 3 ha) as the input information for variable

application of base fertilizers (P, K, Mg, Ca). Farm machines are equipped by RTK guidance

with 2-4 cm accuracy. Farm agronomists don’t use any strategy for VRA of nitrogen fertilizers

and crop protection because of lack of reliable solutions in CZ.


The main focus of the pilot will be on the monitoring of cereal fields by high resolution satellite

imaging data (Landsat 8, Sentinel 2) and delineation of management zones within the fields

for variable rate application of fertilizers. The main innovation is to offer a solution in form of

web GIS portal for farmers, where users could monitor their fields from EO data based on the

specified time period, select cloudless scenes and use them for further analysis. This analysis

includes unsupervised classification for defined number of classes as identification of main

zones and generating prescription maps for variable rate application of fertilizers or crop

protection products based on the mean doses defined by farmers in web GIS interface.





Use case title Cereal crop monitoring (Pilot B1.4_CZ_LESPRO)

Location Rostenice (Vyskov, Czech Republic)

Area Size 8,300 ha

Target Crops Cereals (winter wheat, spring barley, grain maize)

Goals Develop a platform for mapping of crop vigor status by using EO data (Landsat, Sentinel)


Development of Web Map Service / Portal for visualization, analysis and processing of EO data over farm fields as the support tool for creating prescription maps for VRA (mainly application of N fertilizers or crop protection).

Current

solutions

Compute(System) Dedicated server

Intel® Xeon® E5-2420 v2, 2.20GHz, 6 cores, 64GB ECC RAM

Storage HDD: 2x 4TB + 2x 8TB (12 TB total)

Networking -

Software -



Centralized:

Landsat data repository (https://espa.cr.usgs.gov)

Sentinel 2A/B data source (https://scihub.copernicus.eu/)

Google Earth Engine platform for fast viewing EO data:

(https://earthengine.google.com/)

Field boundaries from Czech LPIS database as shp or xml (http://eagri.cz/public/app/eagriapp/lpisdata/)

https://espa.cr.usgs.gov/

https://scihub.copernicus.eu/

https://earthengine.google.com/

http://eagri.cz/public/app/eagriapp/lpisdata/



Ortophotos, topography maps, cadastral maps – as WMS service (http://geoportal.cuzk.cz/)

Farm data:

Crop rotation, crop treatments records, yield maps, soil maps

Volume (size) EO data – hundreds of GB per year for farm area (one Sentinel 2 tile, two Landsat tiles)

LPIS – 0.5 Gb for Czech Rep. as actual data, for pilot area up to 200 MB including historical data (from 2004)

Farm data – yield maps up to 1GB per year (point + raster data). Treatments records as tables.

Velocity

(e.g. real time)

Sentinel 2 approx. 4 days revisit time

Landsat 8 16 days revisit time (resp. 8 days)

LPIS changes – monthly to once per year

Farm data -

Variety


EO data is acquired as raster format (geotiff), LPIS as xml/shp vector data.

Farm data as shp, geotiff, dbf and other


EO data 2-3 days during vegetation season (March-July), once a week in the rest of the year.

Farm data are updated irregularly.

Big data science (collection,

curation,

analysis,

action)


Centralized data are trusted: EO data - European Space Agency (ESA) and United States Geological Survey (USGS). LPIS data Ministry of Agriculture of Czech Republic.

Visualization Standard visualization service for raster and vector data. WMS visualization defined by providers (Czech Office for Surveying, Mapping and Cadastre)

Data quality (syntax) Data quality depends on the providers

Data types Geotiff, shp, xml, dbf and other

Data analytics

http://geoportal.cuzk.cz/



Big data specific challenges

(Gaps)

Combining EO data with farm geospatial records.


Identification of crop growing parameters and its classification by multitemporal analysis of high resolution (spatial and temporal) EO data. Transferability of results and processing workflow across various field and crop management condition.



No datasets of this pilot are considered sensitive or related to personal data.

Highlight issues for generalizing

this Use case (e.g. for ref.

architecture)

Main technical problems are related to automatization of EO data processing and storage.

More information

(URLs)

Landsat ESPA https://espa.cr.usgs.gov

ESA https://scihub.copernicus.eu/

Google Earth Engine https://earthengine.google.com/

Czech LPIS (public) http://eagri.cz/public/app/lpisext/lpis/verejny2/plpis/




Who (type of user) I want to (can you perform some

task)


Farm Enterprise,

Agronomist, Farm

Advisor, Service Dealer

want to monitor the growth of

cereal crops during the season


fertilizing and crop health

measures

Farm Enterprise,

Agronomist, Farm


want to evaluate differences in

plant growth between important

crop growth stages


fertilizing and crop health

measures for specific areas

https://espa.cr.usgs.gov/

https://scihub.copernicus.eu/

https://earthengine.google.com/

http://eagri.cz/public/app/lpisext/lpis/verejny2/plpis/



Farm Enterprise,

Agronomist, Farm


want to analyse crop biomass and

yield variability within each field

for one or several years.

to explain yield differences and

evaluate possible tactical

actions for future seasons

Farm Enterprise,

Agronomist, Farm


want to classify management

zones and to estimate crop

treatment intensity

to prepare prescription maps

for fertilizing and crop

protection


Agronomists want effective management of agrochemicals (fertilizers, pesticides) in form of

site specific crop management treatments and improved farm productivity while following

sustainable agriculture practices.

Our solution is to develop web map portal (service) for visualization, analysis and processing

of EO data for selected farm area:

● Visualization of crop status by vegetation indices within selected time period

● Identification of spatial variability of crops within each farm field and alerting service

for farmers (in case that crop variability is higher than usually for that site)

● Zoning of fields by using machine learning algorithms for selected time-period and

creating of prescription maps (incl. export into shp or isoxml) by considering restricted

zones for application.





This section presents the "Agro Pilot B1.4 Motivation view" view defined in the "Agro Pilot






This section presents the "Agro Pilot B1.4 Strategy view" view defined in the "Agro Pilot 1.3.1





Three relevant KPI were identified for this pilot:

• Area of processed EO data

• Accuracy of management zones delineation by field survey and yield maps

• Increase of fertilizers use efficiency and farm productivity




internal project deliveries from work packages 4 and 5 (WP4, WP5).




10.5 Big data assets




Pilot 9 [B2.1] Machinery management 11.1 Pilot overview


This pilot is focused mainly on collecting telemetry data from machinery and analysing them

in relation with other farm data. The main challenge is access to data and data integration,

when farmer uses tractors and equipment from various manufacturers with different

telematics solutions and different data ownership/sharing policy. Czech tractor manufacturer

Zetor that is now developing and implementing first telematics system for its tractors will

cooperate on creating interoperable solution for access to tractor data, FederUnacoma will

support the pilot with its huge experience in electronic communication standards and in AEF

project teams working on M2M synchronization, wireless infield communication and FMIS

systems. Lesprojekt will ensure integration with other relevant farm data and related analysis.


In many cases farms or agriculture service organizations owns tractors of more than one

brand/family. Although the communication protocols used in control units of farm machinery

and data collection are subject of standardization, the telematics solutions including data

ownership/usage policy are usually specific to each tractor brand/family and the level.

Furthermore, attention shall be paid to ISO and CEN standards regulating data sharing in

agriculture basing on the input coming from industry organizations like CEMA and AEF.

Although this is not issue and can be even desirable for purposes of tractor producer’s

customer care responsible for solving technical problems on tractor, for farmers it can be hard

or impossible to connect the data coming from tractor with other farm data relevant for

agronomical / economical evaluation of machinery usage. Despite the fact tractor have

telematics solution, farmer sometimes need to use third party device and software to obtain

data for field specific analysis

Method

Zetor company is currently developing and testing modular telematics solution which is

supposed to be part of all Zetor tractors. The solution will provide several levels of

functionality ranging from basic telematics for customer care and basic location information

for customer to field specific economic analysis and precision agriculture.



Figure 37: Zetor tractors

The highest level of modular solution will offer connection to other data relevant for farm

management like field boundaries obtained Land Parcel Information system (LPIS), elevation

model and possibly yield potential maps derived from EO data. This connection will enable

evaluation of economic efficiency and other analysis on the level of individual fields or even

parts of the field.

These analyses will become part of Zetor solution and/or they will be enabled by defining data

exchange policy and opening subset of telematics data to third party farm management

information systems and analytical services.

Lesprojekt is member of Wirelessinfo association and participates on development of system

for monitoring and analysis of farm machinery data. The data are obtained using third party

monitoring units installed to tractors but the system is ready to accept data from build-in

telematics system if tractor manufacturer opens the system to external services. The analysis

is focused mainly on evaluation of economic efficiency of machinery usage and it enables

analysis on the level of individual tractors, fields or parts of the field (Management Zones).

Lesprojekt will cooperate with Zetor on the development and testing of telematics solution.

FederUnacoma association which is expert in standardization in agricultural industry will

provide support and advisory services for this pilot.


This pilot has two main connections to big data.

1. Management zones reflecting in field variability, which are going to be connected with machinery data which are based on EO data covering vegetation period through several years (up to 8 years). Tools for automatic selection of suitable scenes and EO data processing chain are needed.

2. Data coming from machinery can reach hundreds of MB per one tractor per year. Total volume is multiplied by number of tractors.





Use case title Machinery management


Author/company/email Karel Charvát jr./ Lesprojekt /[email protected]

Jaroslav Šmejkal/ Zetor/ [email protected]

Alessio Bolognesi/ FederUnacoma /

<[email protected]>

Actors/stakeholders and

their roles and

responsibilities

Farmers, Advisory services, Tractor manufacturers

Goals ● Support development and implementation of Zetor’s

telematics solution

● Support authorized users’ access to telematics data using

tractor manufacturer’s build-in telematics solution or

third-party monitoring tools (depends on tractor

manufacturer data ownership/usage policy.)

● Support integration of tractor and agricultural machinery

data with other relevant farm data and interfaces for data

import into FMIS

● Extracting comparable information from data coming from

various tractors of different manufacturers

● Evaluation of economic efficiency of tractor/machinery

usage and crop profitability.

Use case description Refer to the pilot case definition section and diagrams in the pilot

modelling sections.

Current

solutions

Compute(System) Individual systems of different tractor

manufacturers/families. Various FMIS

(farm management information

systems)

Storage Various telematics system and FMIS

systems

Networking Various



Software Various

Big data

characteristics

Data source

(distributed/centralized)

EO: centralized

Farm data: distributed

Tractor data: distributed

Volume (size) Tractor data: megabytes per each

tractor per week

Field boundaries: megabytes per

farm, gigabytes per country

Other farm data: Up to gigabyte per

year

EO data – hundreds of GB per year for farm area (Czech Republic)

Management zones based on EO data:

megabytes per farm per year

Velocity

(e.g. real time)

Tractor data: Real time

Field boundaries: typically one per

year

Other farm data: irregularly, based on

data type

EO data: Landsat8 every 16 days,

Sentinel 2 every 4 days

Variety

(multiple datasets,

mashup)

Multiple datasets

Variability (rate of

change)

High

Big data science

(collection, curation,

analysis,

Veracity (Robustness

Issues, semantics)

Data coming from various tractor

types or various monitoring units (e.g.

fuel consumption) can have different

interpretation. Sometimes

recalculation is necessary to make



action) data from different tractors

comparable

Visualization Web based maps (e.g. HSLayers) using

various layers: background map, field

polygons, management zones,

tractors trajectory.

Graphs shoving results of various

analysis related to tractor utilization

or operation cost.

Data quality (syntax) Various datasets have various quality.

Quality of tractor data depends on

tractor type, build in telematics

solution or third-party monitoring

unit.

Quality of field boundaries data varies

between countries and sometimes

between farms.

EO data quality is changing over time.

Data types Imagery,

Machinery data in various formats,

Field data and management zones

data in vector formats + related

attribute data

Data analytics Descriptive analytics focused on

tractor utilization, economic

efficiency, and in-field variability

related to management zones.

Predictive analytics: Detection and

prediction failures in Zetor tractors.

Big data specific

challenges (Gaps)

EO data selection, processing, management zones delineation and

integration with tractors monitoring require manual inputs. There

is a need for reduction of manual inputs.



Big data specific

challenges in

bioeconomy

Using big data technologies to improve energy efficiency, protect

soil and environment, reduce using chemicals



Telematics data, visualisations, and analysis must be accessible

only to authorized person

Highlight issues for

generalizing this Use

case (e.g. for ref.

architecture)

EO data management might be at could have many similarities

with other pilots

More information (URLs) www.databio.eu




Who (type of

user)

I want to (can you perform some task) Why (achieve some goals)

Zetor company Implement telematics solution for

their tractors which offers access to

tractor data to authorized users

(farmers, customer care) and

interoperability with other farm

related data, FMIS etc.

Offer added value to customers.

Farmer/advisors Track my tractors which might be

made by different manufacturers and

analyse their utilization in relation

with other farm data or id-field

variability data based on EO.

Have an overview of the location

and movement of tractors.

Evaluate economic efficiency of

machinery usage and crops

profitability, including in.

Optimize cultivation technology


● The main motivation for farmers/advisors is using single environment to analyse and




display data coming from various tractor in relation with other farm and external data.

This analysis can support decisions affecting economic efficiency.

● The main motivation for Zetor is implementation of telematics solution, which satisfy

the user need and helps to improve customer care.



This section presents the "Agro B2.1 Motivation view" view defined in the "Agro Pilot 1.3.1





11.3.2 Agriculture Pilot B2.1 Strategy view

This section provides the "Agro Pilot B2.1 Strategy view" view defined in the "Agro Pilot 1.3.1





These relevant KPIs have been identified so far:

• Numbers of tractors and agricultural machinery using DataBio solutions.

• Number of various tractor brand/models tested.

• Amount of collected data.




Figure 40: Agriculture Pilot B2.1 initial roadmap





Pilot 10 [C1.1] Insurance (Greece) 12.1 Pilot overview


The agri-food sector is constantly exposed to major risks threatening its viability. Production

risks are among the biggest concerns of the agribusiness value chain as they relate to the

uncertainty about the production levels that the farmers could reach following standard

farming practices. The agricultural sector is extremely vulnerable to physical hazards (e.g.,

floods, hail) and biological threats (e.g., pests, diseases). Thereby, insurance in the agri-food

sector deals with the increasing demand for agricultural insurance products and is expected

to play a vital role in the forthcoming years as a tool for risk management. However, due to

its multi-parametric nature, agricultural insurance is considered a special category in the

insurance product portfolio. Difficulties in obtaining enough and valuable data for damage

assessment, the complex biological processes that are incorporated in the crops growth

stages, and the vast variability of production according to geographical criteria, creates an

environment of great uncertainty that requires new techniques and expert knowledge.


The main focus of the proposed pilot is to evaluate a set of tools and services dedicated for

the agriculture insurance market that aims to eliminate the need for on-the-spot checks for

damage assessment. The pilot will concentrate on fusing heterogeneous data (EO data, field

data) for the assessment of damages at field level. NP will support the activities for the

execution of the full life-cycle of the pilot. Moreover, a major Greek insurance company,

Interamerican, will be actively engaged in the pilot activities, bringing critical insights and its

long standing expertise into fine-tuning and shaping the technological tools to be offered to

the agriculture insurance market.

Table 26: Agriculture pilot C1.1 overview of pilot activities

Location North Greece

Area Size 12000ha

Targeted Crops 7 crop types (wheat, stone fruits, etc.)

End-Users Insurance value chain stakeholders

Method

The overall objective of the pilot is to validate a holistic framework comprising of EO-based

data fused with field measurements from IoT stations and offered through flexible UIs and

information visualization mechanisms. The system is designed appropriately to support key



business processes and the need of the insurance market value chain stakeholders. The

methodology of the pilot activities involves the integration of high power computing, machine

learning methods, geospatial data analytics with data coming from EO platforms and will

allow for integration with IoT based data streams and services. The convergence of the

aforementioned technologies in a single dedicated framework is expected to deal effectively

with insurance market demands which require a smooth transition from traditional insurance

policies (expensive, require human experts for damage assessment) to more flexible index-

based insurances. Index products are principally oriented in investigating yield loss due to

extreme events throughout the cultivating period, e.g. rainfall deficit/excess or high/low

temperature. Most of the developed indices are crop-specific divided into fractions of crop

life cycle and calibrated using historical yield statistics [REF-09]. This way, index-based

insurance provides transparency and reduces bureaucracy since it is based on objective

predefined thresholds. Furthermore, it has low operational costs requiring minimal human

intervention. On the top of that, this new type of insurance can eliminate field loss

assessment, adverse selection and moral hazards since the whole process is fully automated,

meaning that the point where the pay-out starts (trigger) and the point where the maximum

pay-out is reached (exit) are based on a prespecified fixed model per crop. This pilot can

drastically contribute towards the aforementioned directions and the generation of index

insurance products via the IoT stations network, which can provide historical and current

weather data, enriched with yield data information extracted from the work calendar and

stored in the NP’s cloud infrastructure. On the other hand, satellites can provide crucial

information to support insurance by overcoming challenges related to lack of stations in all

insured areas or unavailability of long-time series data. This way, EO data can be exploited for

damage assessment (against weather risks) and damage frequency over years as well as to

create value propositions for the key stakeholders of the insurance industry (Interamerican).


Within its cloud infrastructure (GAIA cloud), NP has already started collecting remote sensing

data from the new Sentinel 2 optical products which are being extracted and stored since the

start of 2016. This comprises both raw and processed data (atmospherically corrected

products, extracted vegetation indices) represented in raster formats that are being handled

and distributed using optimal big data management methodologies. Moreover, NP collects

field-sensing data through its network of telemetric IoT stations, called GAIAtrons. GAIAtrons

offer configurable data collection and transmission rates. Since 01/03/2016 over 1M samples

have been collected and stored to NP’s cloud infrastructure that refer to atmospheric and soil

measurements from various agricultural areas of Greece.

Benefit of pilot

The pilot is expected to have a direct impact on key business processes of the insurance

industry. A set of benefits include a) reduce bureaucracy, transparency, b) eliminate the need

for on-the-spot checks for damage assessment, c) eliminate adverse selection, d) eliminate

moral hazard, e) reduce operational and transaction costs, f) offer rapid pay-out. In addition,

it would lead to improvements in a) NP’s GAIA cloud’s stability, availability, security,



interoperability and overall maturity, b) NP’s GAIABus DataSmart functionality in terms of

real-time analytics, data stream and decision support processes, multi-temporal object-based

monitoring, cloud-based services that integrate earth observation with image processing,

machine learning and spatial modelling. Moreover, it will allow new technological trends,

such as deep networks, to be exploited in order to facilitate and improve long-established

insurance procedures.


Table 27: Summary of pilot C1.1 (ISO JTC1 WG9 use case template)

Use case title Insurance


Author/company/email NP


responsibilities

Interamerican – Insurance company,

Single Farmers – insuring their crops against risks of various types.

Goals Provide damage assessment information through an automated holistic framework. Better control insurance claims and shape the insurance products (data abundancy, law of large numbers)


Current

solutions

Compute(System) Insurance companies offering products to farmers do not rely on additional information (satellite data, sensor data, etc.) for claim assessment. Instead, they most often offer rely on on-the-spot checks that require significant resources for the calculation of damages against weather (e.g. rainfall, temperature) and biological (e.g. pests, diseases, contamination) perils.

Storage -

Networking -

Software -



Centralized (Within GAIA Cloud): Field sensing data from GAIAtrons, Remote sensing (Earth observation) data, anonymized IASC data

Volume (size) ~7.5 TB/year for remote sensing data, including raw data and extracted



biophysical and vegetation indices for the pilot areas, ~5GB/year for field sensing data for 27 deployed GAIAtrons

Velocity

(e.g. real time)


Variety







analysis,

action)


Need for a system that can constantly provide support to key business processes of the insurance market

Visualization Spatio-temporal information for visually assessing the risk/damage level

Data quality (syntax) The quality of field sensing data is being controlled by several filtering, outlier detection and stream processing mechanisms. The integrity of remote sensing data quality is being assessed by a hash check upon product download.


Data analytics Diagnostic and descriptive


There is a need for VHR data for validation and optimization of methodologies. There is a need for smarter fusion of the heterogeneous data types that are being collected towards providing accurate insights. To this end, it is important to explore



mechanisms that could combine raster and vector data at parcel level (polygon) and station level (point).


It has been identified that technological trends, such as deep learning techniques, can be exploited for encountering several big data challenges (e.g. data-driven crop classification models, selection of training data sets per crops) and for advancing the effectiveness of traditional machine learning methodologies for the identification of possible crop damages/losses.






Machine learning methodology for crop modelling


https://www.interamerican.gr/


Table 28: Agriculture pilot C1.1 stakeholders and user stories


Farmer As a farmer I want to be insured against several types of risks (damages)

Be assured that I will be compensated in case of hazards for my crops

Insurance company (Interamerican)

As an insurance company I want to improve key business processes

Reduce operational costs related to damage assessment and collect abundant data in order to manage insurance claims more effectively





• to raise the awareness of the farmers, farmer cooperatives and insurance companies on how new technological tools could optimize farm profitability (by insuring the agricultural products) and offer a significant advantage on a highly competitive sector.

• to promote the usage of EO-data, IoT data, etc. for better risk management and support in respect to key business processes (visual/field inspection of loss claims) that require time and extra expenses.




12.3 Pilot modelling with ArchiMate The current section presents the "Agriculture C1.1 Insurance Greece modelling with

ArchiMate" view point described using the ArchiMate standard.

12.3.1 Agriculture pilot C1.1 Motivation view

This section presents the "Agriculture C1.1 Motivation view" view defined in the "Agriculture

C1.1 Insurance Greece modelling with ArchiMate" view point.

Figure 42: Agriculture pilot C1.1 Motivation view










12.3.2 Agriculture C1.1 Strategy view

This section provides the "Agriculture C1.1 Strategy view" view defined in the "Agriculture

C1.1 Insurance Greece modelling with ArchiMate" view point.

Figure 43: Agriculture pilot C1.1 Strategy view








system and collect them in an internal storage system for presentation of current




One relevant KPI has been identified so far, namely:

• %Accuracy in damage assessment.

• %Decrease in the required time for conducting an assessment.





Figure 44: Agriculture pilot C1.1 initial roadmap







Figure 45: Agriculture pilot C1.1 BDVA reference model



Pilot 11 [C1.2] Farm Weather Insurance

Assessment 13.1 Pilot overview


Agricultural production faces a myriad of risks. Nevertheless, two major risks are of concern

to the agricultural sector—price risk caused by potential volatility in prices and production

risk resulting from uncertainty about the levels of production that primary producers can

achieve from their current activities. It is likely that these major risks will increase in the

future—price risk due to liberalization of trade and production risk caused by the effects of

climate change. In this challenge environment, agricultural insurance is an important part of

ensuring long-term stability and growth of the agriculture sector, and facilitating access to

credit, helping to reduce the negative impacts of natural catastrophes, and encouraging

investment in innovative production techniques and technologies.


The objective of proposed pilot is the provision and assessment on a test area of services for

agriculture insurance market, based on the usage of Copernicus satellite data series also

integrated with meteorological data, and other ground available data.

Among the needs of the insurances operating in agriculture, one of the most promising in

terms of fulfilment with Earth Observation data is the evaluation of risk assessment and

damages estimation down to parcel level. For damage assessment, the operational adoption

of remotely sensed data based services will allow optimization and tuning of new insurance

products based on objective parameters, such as maps and indices, derived from EO data and

allowing a strong reduction of ground surveys, with positive impact on insurances costs and

reduction of premium to be paid by the farmers.

For the risk assessment phase, the integrated usage of historical meteo series and satellite

derived indices, supported by proper modelling, will allow to tune EO based products in

support to the risk estimation phase.

Services provided in the pilot will support:

• Risk assessment: service of historical spectral and meteorological analysis in support to risk assessment of crop failures

• Damage and loss assessment: provision of indexes through the integrated usage of meteorological data and multispectral and multi-temporal information for the identification of site-critical weather situations on a small reference area.




Use case title

Farm Weather Insurance Assessment


Author/company/email Coordinator: Antonella Catucci/e-GEOS/[email protected]


responsibilities

Satellite Service Providers and Research and technology Organization/ Added value maps and products providing information for risk and damage assessment to be used by insurances in the agriculture domain;

Meteorological and Environmental EO service provider/ Meteorological data and value-added product about the historical and actual status of the considered areas of interest;

End Users/ definition of requirements/provision of input crop data/ validation of the service.

Goals The objective of proposed pilot is the provision and assessment on a test area of services for agriculture insurance market (risk and damage assessment useful for premium and reimbursement definition), based on the usage of Copernicus satellite data series also integrated with meteorological data, and other ground available data.


Current

solutions

Compute(System) Non-existing system today. Traditional methods to assess the risk are related to statistical data on the insured parcels; while traditional methods for damage assessment and reimbursement payment are based on in field verification.

Storage Local system + web based information systems for general statistical data

Networking Manual assessment of Web, internal data about insured parcels

Software Multiple individual systems, not integrated processing and display




Centralized – related to crop information and in-situ data



Distributed/local: satellite meteorological and optical data

Volume (size) Meteorological: Each single data collection (e.g. precipitation from Meteosat) has a size of 3 MB per 100Km2 per 10 years considering a resolution of 4,5x4,5 km and frequency of 15 min

HR Optical S2 data: Each tile ~100x100km2 and ~0.5GB. 1 single tile cover each Pilot areas.

In situ data: Usually the local data can be estimated in 3 MB per year per 100km2 of interpolation area.

Velocity

(e.g. real time)

Not highly varying considering that: risk information is mainly related to historical data and actual but no-real time data. Damage assessment requires processing results after one two days after the event.

Variety


Multiple Datasets: the idea is to include satellite meteorological optical data together with in-situ data provided by the users (past events, crop and soil information).


Rate of change depends very much on data source/type.


analysis,

action)


Time-series analysis must contain a certain number of real measurement in order to give hints on the identified risks and damages.

Visualization Visualization of complex risk maps and processed images coupled with data analytics providing information about crop trends and eventual correlation results. Farmers (and insurance companies) are not used to analyse meteorological data in a geographic/spatially-distributed way. This underlines the need for a good visualisation component.

Data quality (syntax) Collected information must be linked to parcels database



Data types Need to analyse multi-dimension (explore) and multisource data.

Data analytics Aggregation mechanism (Drill Down at parcel level (polygon) and station level (point).

Plotting and visualization of processing results (multi modal dashboard).


There is a need for closing the gap between insurance daily activities and technological instruments able to provide historical and actual information able to support and optimize premium and reimbursement definition. Other need is to use Big Data techniques to investigate eventual correlation among weather data and crop damage in order to better estimate the spatial and temporal risk distribution.


economy

The Earth observation and weather data will be challenging in sense of dealing with the Volume of data. Whereas the farm data will be challenging in Variety of data coming from different sources, in different formats, using different semantics.



Reference data about loss from insurers or farmers will be used.

This would require a clause on confidentiality and a software

component handling security and privacy of certain datasets.

Moreover, the confidentiality of the services for the Insurance

could be an issue.



Correlation analysis of different and heterogeneous datasets (actual and historical data) as support to the risk assessment. Damage estimation from image analysis by using machine learning approach and techniques.


www.databio.eu



13.2 Pilot case definition The main end users are insurance acting in the agriculture domain. Nevertheless, farmers

have been considered as secondary users and beneficiaries of the services included in the

pilots. With regard to the Insurance activities two are the identified main components of the

insurance value chain that will be supported by the pilot activities:

• Premium definition: traditionally insurance estimation of the risks affecting the area, and linked to the premium requested, is realized by using a statistical approach. Ground true data together with Earth Observation and Meteorological information




collected and correlated with Big Data advanced methods will drastically increase the risk assessment capacity and will support a more appropriate definition of the premium.

• Damage assessment: the damage assessment process to pay indemnities requests is traditionally conducted by means of in field inspections. The use of Big Data instruments and techniques that integrate different data sources will support the insurance also in this phase reducing the in-field management costs. Here following some use cases derived from the above-mentioned activities.



Who (type of user) I want to (can you perform

some task)


Insurance company determine the regional

spreading of risks for each

type of bad weather (hail,

heavy rain, drought)

to evaluate their insurance

portfolio

Insurance company determine temporal trend for

each type of bad weather

(hail, heavy rain, drought)

to determine the possible

influence of climate change

on crop growth.

Insurance company determine the actual risk per

crop on field level

to determine the pricing of

the insurance package

Insurance company assess the damage caused by

a bad weather event

ensure non-erroneous

compensation to farmers

Farmers view the risk level for heavy

rain and drought on field level

(optionally crop specific)

to evaluate the business case

for prevention measures


Insurances need a two-step process of risk assessment, followed by damage assessment.

Relying on historical data of calamities (e.g. hail damage, frost damage, heavy rains),

(historical) meteorological data, soil data, crop data and height data a Big Data analysis

algorithm should provide the optimized risks pricing of insurance packages. The damage



assessment will be relying on satellite imagery (providing biomass indices and/or weather

event extend maps) before and after the bad weather event to ensure non-erroneous

compensation process. These data should be combined with insurance data on damages to

calibrated a Big Data analysis algorithm.

The risk is based on:

● Local/regional weather and the frequency and intensity of bad weather conditions

(extreme rainfall, hail, drought)

● Sensitivity of the crop for the weather-related risks

● Topography (height / height differences) and

● Soil type

For the risk assessment, the geographic distribution of the risk areas will be presented, based

on a.o. historical (bad) weather conditions, like intense rainfall, hail and drought. It should

also be possible to present a temporal trend in order to determine the possible influence of

climate change on crop growth. For damage assessment satellite imagery (combined with

meteorological data) will be used for evaluation of hail, flooding and drought. Combined with

the crop data (e.g. potatoes die after a few days of flooding, other crops can survive) the

actual damage can be determined. The risk assessment data can also be used for services for

farmers to benchmark their risk and evaluate the business case for prevention. Privacy of

results needs to be discussed with users. Guarantee of confidentiality through a written

agreement at the very beginning of the pilot phase.

13.3 Pilot modelling with ArchiMate The current section presents the "Agriculture C1.2 Insurance Netherlands modelling with

ArchiMate" view point described using the ArchiMate standard. It lists the views and

nomenclatures composing the view point.


This section provides the "Agriculture C1.2 Motivation view" view defined in the "Agriculture

C1.2 Insurance Netherlands modelling with ArchiMate" view point.













13.3.2 Agriculture pilot C1.2 Strategy view


C1.2 Insurance Netherlands modelling with ArchiMate" view point.













Two preliminary KPIs that has been identified so far:

● Information correctness: risk assessment success ratio and damage assessment success ratio having an acceptable error rate when tested on historic data that it was not trained on.

● System usage: Number of users of the services, and number of users visiting the website with information about.









Pilot 12 [C2.1] CAP Support 14.1 Pilot overview


In the framework of EU Common Agriculture Policy (CAP), farmers can have access to

subsidies from the EU, that are provided through Paying Agencies or Authorized SMEs (the

Greek case) operating at National or Regional level according to the Member State. For the

provision of the subsidies, Paying Agencies must operate several controls over farmer parcels

in order to verify the compliance of the cultivation with EU regulations. These controls are

carried out with a large support from remote sensing. Today, due to the cost of remote

sensing data, the controls are limited to a sample of the whole amount of farmers

declarations, and the control is often focused on a specific time window, not covering the

whole lifecycle of the agriculture parcel during the year. With the availability of the Sentinel

satellite data, it is now possible to have access to a large amount of free of charge satellite

images providing frequent coverage of the whole EU with a resolution compliant with the

average agriculture field size for many farmers. Therefore, it is possible, through the

processing of satellite time series, to provide services in support to the Paying Agencies and

the authorized collection offices for a more accurate and complete control of the farmers’

declaration. Foreseen services will allow a more complete and efficient management of EU

subsidies, strongly enhancing their procedure for combating frauds or not compliant

behaviours, thus guaranteeing an evident economic return also in terms of saving for ground

surveys and optimization of control management.


Reference service situation and user needs

Member States must take the necessary measures to ensure that transactions financed by

the EAGF (European Agricultural Guarantee Fund) are implemented correctly. Furthermore,

Member States, through the National Paying Agencies, must prevent irregularities and take

the appropriate actions if they do occur. For this purpose, the national authorities are

required to operate an Integrated Administration and Control System (IACS) in order to

ensure that payments are made correctly, irregularities are prevented, revealed by controls,

followed up and amounts unduly paid are recovered. Controls are currently operated on the

basis of workflows relying on the use of EO data (aerial data, VHR and HR satellite data)

integrated with additional geodata and databases. High and very-high resolution satellite

imagery (HR and VHR) are currently used to check farmer’s declarations, and increasingly to

verify the compliance of their farming practices with agro-environmental rules. In general,

National Paying Agencies operate the workflows through service integrators, that process EO

data and tune the operational workflow. Service integrators are often local enterprises, that

operate with different software solutions based on a common EO data feed. Main

stakeholders, external to the National Paying Agencies and involved in the process, are:

satellite image providers, aerial image providers, value adders on image data, service



integrators. The value of the market for satellite image providers is in the order of 6-7 Milion

euros per year, with a strong improvement foreseen for 2015.

For its controls, the National Paying Agencies adopt a national agriculture-oriented land cover

reference map (updated in general on a three-years basis with aerial data), and performs

detailed and tailored checks by means of satellite data over risk based and random sample

zones covering 5% of the farms on a yearly basis.

The controls are aimed at:

• To check the presence of cultivated land and specific crops in case of coupling

• To check the diversification of crops according to greening criteria

• To check the presence and maintenance of permanent pastures

• To check the presence and maintenance of land lying fallow

All these controls are operated only on the 5% of the farms with fresh VHR and HR EO data.

On all the other farms, the control must rely on EO data that could have been collected one

or two years before (due to LPIS updating cycle), with a negative impact on farm compliance

evaluation. In addition to this situation, it must be noticed that most of the checks could

provide more reliable results if based on multi-temporal time series, since adopted data could

introduce bias in the interpretation.

Objective



temporal series of free and open EO data, with focus on Copernicus Sentinel 2 data. Products

and services will be tuned in order to fulfil requirements from the 2015-20 EU CAP policy, and

will be general information layers and indicators on EU territory with different level of

aggregation and detail up to farm level. Therefore, the services, to be tuned and confirmed

with end users, will:

• Identify parcels (monitoring objects) over which the declared crop is potentially different from the one that extracted from the EO models (outliers). The service is based on Sentinel data and machine learning methods for the description of the crop and analytics methods for the identification of the outliers. The service will allow the performing of big data analytics to various crop indicators on parcel level.

• Identify different crops present inside a single farm when the global size of declared surface is exceeding a specific threshold. This is due to the fact that CAP requires crops diversification such that farmers should cultivate at least two/three different crops. The service will be based on the management of optical satellite data together with farmer declarations information and limited ground measures if any, and will provide an indication of possible compliance/not compliance of the farmer vs. EU CAP requirements.




Use case title

CAP Support


Author/company/email Coordinator: Antonella Catucci/e-GEOS/[email protected]


responsibilities

Satellite Service Providers / Added value maps and products;

Technological and Agricultural Service Company/ providing validation data and information;

End Users/ definition of requirements/provision of crop data/ validation of the service.

Goals The objective of proposed pilot is to provide services in support to the National and Local Paying Agencies and the authorized collection offices for a more accurate and complete control of the farmers’ declaration related to the obligation introduces by the current Common Agriculture Policy.

Use case description In the framework of EU Common Agriculture Policy (CAP), farmers

can have access to subsidies from the EU, that are provided

through Paying Agencies or Authorized SMEs (the Greek case)

operating at National or Regional level according to the Member

State. For the provision of the subsidies, Paying Agencies must

operate several controls over farmer parcels in order to verify the

compliance of the cultivation with EU regulations.

Current

solutions

Compute(System) Different agriculture declaration

management systems are today

available at National or Regional level.

Today, due to the cost of remote

sensing data, the controls are limited

to a sample of the whole amount of

farmers declarations, and the control

is often focused on a specific time

window, not covering the whole

lifecycle of the agriculture parcel

during the year. These controls are

carried out with a large support from

remote sensing.

Storage Local system + web based information systems for general agriculture data management



Networking Manual assessment of Web, internal data about insured parcels

Software Multiple individual systems, not integrated processing and display




Centralized: related to parcel information and provided by the users. The system of the user is a legacy centralized system.

Distributed: satellite optical and SAR data are processed in separate processing platform and then stored in a big-data store and delivered using web services to the user legacy system.

Volume (size) HR Optical S2 data: each tile ~100x100km2 and ~0.5GB. 1 single tile cover each Pilot areas.

Landsat 8: each tile ~185x170km2 and ~0.8GB

SAR S1 data: each Burst ~20x20km2 and ~0.25GB

Velocity

(e.g. real time)

Periodic update (see later on variability)

Variety


Multiple Datasets: the idea is to include satellite optical and SAR data together with parcel data provided by the users.


Rate of change depends very much on data source/type:

- Satellite S1/S2: 5 days - Landsat 8: 14 days - Parcels: yearly update


analysis,

action)


Time-series analysis must contain a certain number of real measurement in order to give hints on the potential inconsistencies.

Visualization Visualization processed images coupled with data analytics providing information about the current status of the parcel with expected status (compliant with farmer declaration)



able to detect potential inconsistencies.

Need of plotting and visualization of processing results (multi modal dashboard).

Data quality (syntax) Collected information must be linked to parcels database (can be different across different countries).

Data types Need to analyse multi-dimension (explore) and multisource data.

Data analytics Aggregation mechanism (Drill Down at parcel level (polygon) and station level (point).

Plotting and visualization of processing results (multi modal dashboard).


There is a need for correlating remote sensed crop status and expected behaviour for the crop typology communicated by the farmer.


The possibility to provide support to the National and Local paying agencies will allow the optimization of economic resources dedicated from the EC to the agriculture domain. These resources could be allocated to other activities in support to the agriculture activities development.



Reference data about potential frauds will be used. This require a clause on confidentiality and a software component handling security and privacy of the pilot results.



Verification of the compliance of the cultivation with EU regulations.


www.databio.eu



14.2 Pilot case definition The proposed pilot project has been tailored on the specific needs of three end users, one

operating at National level (Romania Agriculture Ministry), one operating at Regional level

(AVEPA Paying Agency) in one of the most important agricultural regions in Italy and one

operating in Spain. The services that will be provided in the pilot project will rely on the

processing of big amount of data such those provided by Copernicus Sentinel-1 and Sentinel-




2 satellite, collecting SAR and multispectral image data with a 10-days frequency (that will be

increased to 5-days with the full constellation Sentinel-2A Sentinel-2B Sentinel 1B will be fully

available). Data will also be integrated by time series of Landsat8 satellite, providing images

with a 16-days frequency but at lower resolution.

Geographical location of the pilot:


Romania Italy

Location: South-East Romania

Area size: 10.000 km2

Targeted Crops: 3-10 crop types (Wheat,

corn, sun-flower) TBC

Location: North Italy (Veneto)

Area size: 50.000ha

Targeted Crops: two open crop types

TBC

End-User: National/regional Agencies in charge for the agriculture controls in order to verify

the compliance of the cultivation receiving subsidies with EU regulations. In particular:

Italy: AVEPA is the regional entity in charge for the regional payments in agriculture of

European funds (including FEAGA and FEASR funds) in the Veneto Region. The AVEPA agency

also acquired more responsibility from the Regional Administration to manage the entire set

of international and national funds in the agriculture domain including the CAP

implementation and the payment of the public insurances in case of weather related critical

events. Veneto Region is one of the most important region in the Agriculture and Food

domain in Italy and also the most proactive in terms of international activity related with the

innovation (semi-finalist for the European Communication award PAC 2014).

Romania: APIA (National Subsidy Agency for Agriculture, Ministry of Agriculture) holds

responsibility in Romania of the implementation of CAP mechanisms for direct payments. The

entire procedure is handled by the Integrated System of Administration and Control (IACS)

that also deals with the verification of the compliance of the declarations submitted by the

farmers. Currently, a minimum of 5% from the applications is crossed-checked either by field

sampling or by remote sensing.





Who (type of

user)

I want to (can you perform some

task)


1 – National /

Regional

Subsidy Agency

for Agriculture

I want to verify that farmers fulfil

the CAP criteria related to crop

presence and diversification.

In order verify farmers that receive

simple or decoupled green

payment/subsidies per ha from

European Commission.

2 – National /

Regional

Subsidy Agency

for Agriculture

I want to verify that farmers fulfil

the criteria related to the

maintenance of permanent

pastures.

In order verify farmers that receive

decoupled green

payment/subsidies per ha from

European Commission.




temporal series of free and open EO data, with focus on Copernicus Sentinel data. In

particular, the output of the project will support the National CAP in the verification of the

compliance of farmer declarations and the automatic identification of possible frauds; this

information could be used by the Agencies also to better plan focused controls. The general

methodology will be based on the comparison of the real crop behaviour (detected by

Remote sensing techniques) with the expected trends for each crop typology.

14.3 Pilot modelling with ArchiMate The current section presents the "Agriculture C2.1 CAP support Italy Romania modelling with





C2.1 CAP support Italy Romania modelling with ArchiMate" view point.















C2.1 CAP support Italy Romania modelling with ArchiMate" view point.













Two preliminary KPIs that has been identified so far:

• Information correctness: inconsistencies success ratio having an acceptable error rate when tested on historic data

• System usage: Number of users of the services, and number of users visiting the website with information about

• Satellite on-the-spot checks %: percentage of agricultural parcels covered with satellite on-the-spot checks. This can “significantly increase the efficiency of on-the-spot checks necessary for CAP payments” (https://ec.europa.eu/agriculture/newsroom/286_en)



https://ec.europa.eu/agriculture/newsroom/286_en




Figure 53: Agriculture pilot C2.1 BVDA reference model



Pilot 13 [C.2.2] CAP Support (Greece)


With an annual budget of roughly €59 billion, that accounts for 38% of the EU-28 budget, the

Common Agricultural Policy (CAP) aims to strengthen the competitiveness and sustainability

of agriculture in Europe by financing a range of support measures [REF-10]. CAP effectiveness

is crucial for at least 22 million farmers and agricultural workers but is limited by

administrative burdens, complexity and high implementation costs. The identification of best

practices that lead to the reduction of delivery costs without affecting the effectiveness is a

key priority not only for the National Payment Agencies but also for the EU Parliament. Earth

Observation (EO) has been frequently suggested as the best possible tool for the effective and

efficient implementation of the CAP. However, until now, EO has been limited to performing

“Controls with Remote Sensing” (CwRS) for the purposes of the annual verification of

subsidies claims. This “business process” is applied only to the 5-7% of EU farms and has a

total estimated cost of approximately €40 million per year, which is insignificant when

compared to the €2654.6 million per year of the official annual CAP delivery cost[REF-11]. The

EC and the JRC [REF-12], have together stressed the need for EO-based agricultural

monitoring able to support not only the assessment checks but in total the CAP

implementation and its instruments such as the Good agricultural and Environmental

Monitoring (GAEC) or the Farm Advisory System (FAS). Recent technological improvements in

terms of big data handling, available computing power and the Copernicus Sentinel data and

imagery allows the continuous and automated provision of agri-environmental information’s

for objects being monitored such as the agricultural parcels. Precision and/or Smart Farming

is a sector that relies for many of its key business process on the Earth Observation

technology. It is also a key concept that CAP has promoted as a necessity after 2020 [REF-13],

[REF-14] for the improvement of agricultural production and efficient farm management. In

the upcoming Common Agricultural and Food Policy, which is currently being designed,

Precision and/or Smart Farming and EO are the most valuable tools because their combined

use leads to an optimal and sustainable production and allows the provision of advisory

services based on facts.


The main scope of this pilot is to evaluate a set of EO-based services designed appropriately

to support key business processes and need of the CAP value chain stakeholders. The pilot

activities will focus on two (2) open crop types (dry beans and peaches) and will be held in

the Northern part of Greece. The services to be tested will rely on innovative tools and

technologies, that will sustain the interconnection with IoT infrastructures and EO platforms,

the collection and ingestion of spatiotemporal data, the multidimensional deep data

exploration and modelling and the provision of meaningful insights, thus, supporting the

simplification and improving the effectiveness of CAP. NP and GAIA Epicheirein will support

the activities of this pilot demonstration.




Pilot Site

Location North Greece

Area Size 50000ha

Targeted Crops 2 open crop types (dry beans and peaches etc.)

End-Users GAIA Epicheirein

Method

The overall objective of the pilot is to validate a set of EO value-added products and services

designed appropriately to support key business processes and the needs of CAP value chain

stakeholders. The methodology of the pilot activities involves the integration of high power

computing, machine learning methods, geospatial data analytics with data coming from EO

platforms and will allow for integration with IoT based data streams and services. The

convergence of high computing power, machine learning, and satellite imagery is “a perfect

storm that’s just beginning to peak” [REF-15] and as such the ambition of the current pilot is

to exploit the “produced power” for dealing effectively with CAP demands for agricultural

crop type identification, parcel monitoring, collaboration, transparency and analytics. This

way the value chain stakeholders (GAIA Epicheirein, farmers, farming cooperations, etc.) will

benefit from the EO data, supporting the simplification and improving the effectiveness of

CAP.










optimal big data management methodologies.

Benefit of pilot

The pilot activities aim at providing EO-based products and services designed to support key

business processes and needs of the CAP value chain stakeholders, including:

• The farmer decision-making actions during the submission of aid application. More specifically, for each agricultural parcel, the developed services will allow the a) automated validation of the declared crop, b) accurate definition of the eligibility area,



c) monitoring of the good agricultural and environmental conditions. Each year the Farmer, the beneficiary, must provide evidence to document his/her eligibility. His/her choices during the one-off submission process have great financing impact and may lead to losses or, even worse, trigger penalties. The offered services will boost the farmer’s decision-making power [REF-16] and help them maximize the benefits and minimize the financial risks in relation to the agricultural land for which direct support is requested.

• The Farmer transition towards Smart Farming, i.e. the provision of tools that support not only the compliance with CAP but also assist the adoption and implementation of Smart Farming practices. The proposed services for agricultural and environmental monitoring at parcel level, provide streams of data and facts that can be used for automated irrigation, crop protection, actual crop status and crop variability identification.



Use case title CAP Support




responsibilities

GAIA Epicheirein – Supporting role in the farmers’ declaration process

Farmers from the engaged agricultural cooperatives in the pilot area

Goals Crop identification using EO data to check against the declared crop type


Current

solutions

Compute(System) Non-existing system today. The application process relies only on static background maps. The farmer’s decision is not supported by multi-temporal EO data.

Storage -

Networking -

Software -



Centralized (within GAIA Cloud): Field sensing data from GAIAtrons, Remote



sensing (Earth observation) data, anonymized IASC data

Volume (size) ● ~7.5 TB/year for remote sensing data, including raw data and extracted biophysical and vegetation indices for the pilot areas,


Velocity

(e.g. real time)


Variety







analysis,

action)


Need for a system that can constantly provide support to the farmers’ declaration process

Visualization Spatio-temporal information visualization for facilitating the declaration process

Data quality (syntax) The quality of field sensing data is being controlled by several filtering, outlier detection and stream processing mechanisms. The integrity of remote sensing data quality is being assessed by a hash check upon product download.

Data types Remote sensing data provided in raster format (.jp2). Field sensing data



provided as time series unstructured data with configurable frequency

Data analytics Descriptive and diagnostic analytics for CAP support.


There is a need for VHR data for validation and optimization of methodologies. There is a need for smarter fusion of the heterogenous data types that are being collected towards providing accurate insights. To this end, it is important to explore mechanisms that could combine raster and vector data at parcel level (polygon) and station level (point).


It has been identified that technological trends, such as deep learning techniques, can be exploited for encountering several big data challenges (e.g. data-driven crop classification models, selection of training data sets per crops) and for advancing the effectiveness of traditional machine learning methodologies for crop classification.






Machine learning methodology for crop modelling




GAIA Epicheirein

GAIA needs services to support farmers in the declaration processes

Ensure the validity of the declarations and offer a reliable tool/product to the farmers that would strengthen its position as a service provider for the agri-food sector.



Farmers Farmers need a reliable supporting/consulting tool for the crop type declaration process

Minimize the risk that the process will trigger financial penalties



• to demonstrate the added-value of EO-based products and services designed appropriately to support key business processes and the needs of CAP value chain stakeholders

• to raise the awareness of the farmers, agronomists, agricultural advisors, farmer cooperatives and organizations (e.g. group of producers) on how new technological tools could facilitate the crop declaration process.




15.3 Pilot modelling with ArchiMate The current section presents the "Agriculture C2.2 CAP support Greece modelling with





C2.2 CAP support Greece modelling with ArchiMate" view point.













C2.2 CAP support Greece modelling with ArchiMate" view point.



Figure 55: Agriculture pilot C2.2 strategy view












Two relevant KPIs have been identified so far, namely:

● %Decrease in false crop type declarations following the supporting services vs what

would be expected based on historical data (information correctness measured as

inconsistencies ratio).

● %Accuracy in crop type identification















Conclusion D1.1 provides analysis of agriculture pilots and presents initial pilot description and models.

This description will be used for further development inside WP1, but they are also inputs

important for WP4 and WP5. In initial stage, the document will be used for match making

activities among pilots and technology providers, for modelling and analysis of synergies

among pilots and for definition of DataBio reference architecture.

The work in WP1 will now continue with focus on testing of components and building first

version of pilot applications. After first round of pilots testing of different components from

different producers (technology partners) the pilot experience will be used for definition of

DataBio Reference Big Data Architecture.



References Reference Name of document (include authors, version, date etc. where applicable)

[REF-01] DataBio website. www.databio.eu. Retrieved 2017-06-20.

[REF-02] ArchiMate® 3.0 Specification.

http://pubs.opengroup.org/architecture/ArchiMate3-doc/toc.html

[REF-03] Schellberga J, Hill MJ, Gerhards R et al., 2008. Precision agriculture on

grassland: Applications, perspectives and constraints. Europ. J. Agronomy 29:

59-71

[REF-04] Segarra E, 2002. Precision agriculture initiative for Texas high plains. Annual

Comprehensive Report. Lubbock, Texas, Texas A&M University Research and

Extension Center

[REF-05] Shibusawa S, 2002. Precision farming approaches to small farm agriculture.

Agro-Chemicals Report, 2(4): 13-20

[REF-06] FAO 2009 How to feed the World

https://www.google.cz/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=

0ahUKEwjD07O-

vuLUAhUKYVAKHUVKCcoQFggmMAA&url=http%3A%2F%2Fwww.fao.org%2Ff

ileadmin%2Ftemplates%2Fwsfs%2Fdocs%2Fexpert_paper%2FHow_to_Feed_t

he_World_in_2050.pdf&usg=AFQjCNEvZBB2Wrep78tku_vCige3UqTm3Q&cad

=rja

[REF-07] World Bank. 2011. The World Bank Annual Report 2011: Year in Review.

Washington, DC. © World Bank.

https://openknowledge.worldbank.org/handle/10986/2378 License: CC BY

3.0 IGO

[REF-08] https://www.euractiv.com/section/agriculture-food/infographic/farming-4-0-

the-future-of-agriculture/

[REF-09] E. Coleman, W. Dick, F. Rispoli, A. Stoppa, Remote sensing for index insurance

Findings and lessons so far, International Fund for Agricultural Development

and World Food Programme, 2015

[REF-10] http://ec.europa.eu/agriculture/cap-overview_en

[REF-11] Bruno CHAUVIN, AGRI.DDG4.J.1, 39th Conference of Directors of EU Paying

Agencies, CAP Delivery costs, AMSTERDAM, 25 - 27 May 2016

[REF-12] Arie van der Greft (DG AGRI), Philippe Loudjani (DG JRC), 2016, “Spotlight on

technology in IACS”, 22nd CAP/IACS conference, 25 November 2016

https://ec.europa.eu/jrc/sites/jrcsh/files/1_loudjani.pdf

[REF-13] Cork 2.0 Declaration 2016. A Better Life in Rural Areas. Prepared by the

participants of the European Conference on Rural Development in Cork,

Ireland, 5-6 September 2016

http://ec.europa.eu/agriculture/events/2016/rural-development/cork-

declaration-2-0_en.pdf


http://pubs.opengroup.org/architecture/archimate3-doc/toc.html

https://www.google.cz/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwjD07O-vuLUAhUKYVAKHUVKCcoQFggmMAA&url=http%3A%2F%2Fwww.fao.org%2Ffileadmin%2Ftemplates%2Fwsfs%2Fdocs%2Fexpert_paper%2FHow_to_Feed_the_World_in_2050.pdf&usg=AFQjCNEvZBB2Wrep78tku_vCige3UqTm3Q&cad=rja






https://openknowledge.worldbank.org/handle/10986/2378

https://www.euractiv.com/section/agriculture-food/infographic/farming-4-0-the-future-of-agriculture/

https://www.euractiv.com/section/agriculture-food/infographic/farming-4-0-the-future-of-agriculture/

http://ec.europa.eu/agriculture/cap-overview_en

https://ec.europa.eu/jrc/sites/jrcsh/files/1_loudjani.pdf

http://ec.europa.eu/agriculture/events/2016/rural-development/cork-declaration-2-0_en.pdf

http://ec.europa.eu/agriculture/events/2016/rural-development/cork-declaration-2-0_en.pdf



[REF-14] Louise O. Fresco and Krijn J. Poppe (2016), “Towards a Common Agricultural

and Food Policy”, WUR

http://www.wur.nl/upload_mm/6/b/c/11791580-8cfd-4f29-a8ad-

2d9748c787d0_Towards_CAFP_LR.pdf

[REF-15] Pavel Machalek, co-founder and CEO of Spaceknow

[REF-16] Direct Payments. Eligibility for Direct Payments of the Common Agricultural

Policy, September 2016

http://ec.europa.eu/agriculture/sites/agriculture/files/direct-support/direct-

payments/docs/direct-payments-eligibility-conditions_en.pdf

http://www.wur.nl/upload_mm/6/b/c/11791580-8cfd-4f29-a8ad-2d9748c787d0_Towards_CAFP_LR.pdf

http://www.wur.nl/upload_mm/6/b/c/11791580-8cfd-4f29-a8ad-2d9748c787d0_Towards_CAFP_LR.pdf

http://ec.europa.eu/agriculture/sites/agriculture/files/direct-support/direct-payments/docs/direct-payments-eligibility-conditions_en.pdf

http://ec.europa.eu/agriculture/sites/agriculture/files/direct-support/direct-payments/docs/direct-payments-eligibility-conditions_en.pdf

Documents

D1.1 Agriculture Pilot Definition · D1.1 – Agriculture Pilot Definition Dissemination level PU -Public Type of Document Report ... H2020 Contract No. 732064 Final – v1.0, 30/6/2017