Embed Size (px)
Citation preview
1
Land use map of Rwanda
Prepared during a workshop held in Butare, Rwanda
at the CGIS centre,
from the 22nd
of November until the 3rd
of december Dec 2010
in the framework of the Kagera TAMP project,
implemented by FAO, funded by GEF
2
Workshop participants names (authors of the Rwanda maps are the member of the Rwanda team)
COUNTRY NAME INSTITUTE E-MAIL
RWANDA
Ernest Uwayezu CGIS NUR, HUYE (BUTARE) [email protected]
Elias Nyandwi CGIS NUR, HUYE (BUTARE) [email protected] [email protected]
Antoine Nsengiyumva Kagabo NATIONAL LAND CENTRE, KIGALI [email protected]
Theodomir Mugiraneza CGIS NUR, HUYE (BUTARE) [email protected]
Marie Christine Simbizi CGIS NUR, HUYE (BUTARE) [email protected]
Rachel Murekatete CGIS NUR, HUYE (BUTARE)
Jules Mutabazi ISAR, LIVESTOCK UNIT, KARAMA [email protected]
U. R. OF TANZANIA
Kipondya Winfrida
SCC-VI AGROFORESTRY KAGERA PROJECT, BUKOBA [email protected]
Godwin Safari Edmund Athanas
SCC-VI AGROFORESTRY MARA PROJECT, MUSOMA [email protected]
Frank Onesmo Mkiramwinyi
UKINGURU TRAINING INSTITUTE, MWANZA [email protected]
Elmens Kaboni
MINISTRY OF AGRICULTURE AND FOOD SECURITY, MWANZA [email protected]
UGANDA Dennis Babaasa
INSTITUTE OF TROPICAL FOREST CONSERVATION, BWINDI FOREST NATIONAL PARK [email protected]
Grace Nangendo WILDLIFE CONSERVATION SOCIETY, KAMPALA
[email protected] [email protected]
National Coordinator, Theobald Mashinga (2010), FAO
International Consultant: Dennis Babaasa
National Consultant: Ernest Uwayezu, CGIS NUR
Coordination of activities and facilitator Monica Petri, FAO
Kagera TAMP project contact: [email protected]
We thank the Nile Equatorial Lakes Subsidiary Action Program (NELSAP) of Nile Basin Initiative
(NBI), that provided a significant proportion of GIS inputs data used during the workshop. Input
data were prepared within the preparation of the « Monographie du bassin de la Kagera. Rapport su
le développement du bassin publiée » of the Kagera river basin transboundary water resources
management and development project of the Nile basin Initiative (NELSAP) in July 2008 in
collaboration with BRL Ingénierie.
We thanks Tim Robinson (FAO Senior Officer, NRC), Valentina Ercoli, (FAO Officer, NRC) and
Gianluca Franceschini (FAO Consultant) for the support during the preparation of livestock data.
The method used in this workshop has been developed within the project LADA FAO/UNEP GEF
and is available in the document: Freddy Nachtergaele, Monica Petri, 2011. Mapping Land Use
Systems at global and regional scales for Land Degradation Assessment Analysis. LADA technical
report ( http://www.fao.org/docrep/017/i3242e/i3242e.pdf ).
3
All intellectual property rights, including copyright, in the work performed under the
“Transboundary Agro-Ecosystem Management Programme for the Kagera River Basin” (thereafter
referred as “Kagera TAMP”) shall be vested in FAO, including without any limitations, the right to
use, publish, translate, sell or distribute, privately or publicly, any item or part thereof for non-
commercial purposes. FAO hereby grants to the delegated institution(s) of beneficiary country a
non-exclusive royalty-free license to use, publish, translate and distribute, privately or publicly, any
item or part of the work performed under Kagera TAMP for non-commercial purposes. Delegated
institutions are MINAGRI RAB, RNRA, REMA, GIS NUR, Kagera TAMP Districts, and Nile
Basin Initiative.
4
Index
Contents Index..................................................................................................................................................... 4 1. Introduction, Kagera TAMP project and national LADA method used at the Kagera basin scale . 5 2. Inputs, formats, scale, resolution and precision ............................................................................... 6 3. Africover re-classification................................................................................................................ 7
4. Validation of the land cover classes ................................................................................................. 9
5. Other data used to prepare the land use map ................................................................................. 12
5.1 Protected area ........................................................................................................................... 12 5.2 Classification of livestock statistical data ................................................................................ 13
5.2.1 Use of Agro – ecological zones of Rwanda as a base for livestock distribution .............. 13 5.2.2 Livestock intensity ............................................................................................................ 14
6. Land use map implementation ................................................................................................... 17
7. Land use map improvements ..................................................................................................... 17 8. Land use map database implementation .................................................................................... 19
8.1 Livestock natural region and livestock species .................................................................... 19 8.2 Annual rainfall ..................................................................................................................... 19
8.3 Annual temperature .............................................................................................................. 19
8.4 Elevation and slope .............................................................................................................. 19 8.5 Soil ....................................................................................................................................... 19 8.6 Population density ................................................................................................................ 22
8.7 Poverty ................................................................................................................................. 22
5
1. Introduction, Kagera TAMP project and national LADA method used at the Kagera basin scale
The Kagera TAMP project is funded through the Global Environment Facility (GEF) in partnership
and with co-funding from the governments, partner programmes and donors at country and regional
levels. The project is executed by the UN Food and Agriculture Organization (FAO). The project
goal is to adopt an integrated ecosystems approach for the management of land resources in the
Kagera basin that will generate local, national and global benefits including: restoration of degraded
lands, carbon sequestration and climate change adaptation and mitigation, protection of
international waters, agro-biodiversity conservation and sustainable use and improved agricultural
production, leading to increased food security and improved rural livelihoods.
Kagera Trans-boundary Agroenvironmental Management Programme is going to select its
intervention areas on the base of the more advanced technologies of definitions and assessment of
the land degradation and of the sustainable land management. The selection of most degraded zones
or of the areas with a higher impact of land degradation will provide the basis for the identification
of locations of the next project phases, particularly the implementation of the Farmer Field Schools.
This work will be realised in Rwanda, Uganda, and United Republic of Tanzania, and also in the
entire Kagera Basin (the workshop in Burundi was held from 9th
up to 16th
of November). The
method will be used fro the first time at a hydrographical basin scale.
The selection of intervention areas will be based on the method implemented by the project LADA
(FAO/UNEP GEF) that developed a national assessment method for the land degradation and
sustainable land management that has been tested in six countries (South Africa, Argentina, China,
Cuba, Senegal, Tunis). The method, very rapid, is composed by two main axes:
- the preparation of a land use systems (LUS) map by using all best available data,
- the preparation of the land degradation and sustainable land management map using the
method LADA WOCAT QM, that use the LUS as cartographic unit.
A workshop was organized by Kagera TAMP in Butare (Huye) between the 22nd
of November and
the 3rd
of December 2010 with the participation of GIS experts having different background
(agronomy, livestock, land cover, GRID, etc.) with the objective of preparing a LUS map of the
area. Data used are prepared by NELSAP (Nile Equatorial Lakes Subsidiary Action Program) in
2008 and have been provided to the Kagera TAMP team within the preparation of a Memorandum
of Understanding.
The present report give the complete description of the preparation of a land use map and also the
preparation of a national map of livestock intensity, needed as LUS input. The present method
allows the preparation of a multi-country land use map in two weeks with the work of 13 experts,
and is comparable to the continuous of 130 working days. Further to that, the LUS has been
validated with the support of GIS consultants.
The report details the work realised during the workshop describing the LUS preparation for
Rwanda.
6
2. Inputs, formats, scale, resolution and precision
Table 1 Land use systems data for map
Used for Name Year of
measure
Year / periodicity
of publication
Scale / resolution /
extent Format
Availability or copyright
Producer Metadata Unity of
measurements
Admin units RWA (District) 2006 - 1:250.000 Shp free MINALOC none classes
Land cover based ecosystsems
Africover 1999 2002 1:100.000 shp country FAO and
MINITERE www.africover.org
classes
Livestock density
Livestock statistics
2006 yearly Old
provinces (2001)
stats free National
Institute of Statistics
report heads per species
Livestock natural distribution regions
Agroecological zones
1980 - 1:250.000 shp free MINAGRI none classes
Protected areas
World database protected areas
2009 yearly 1:1.000.000 shp
(polygons) free UNEP
www.wdpa.org
classes
Elevation SRTM Hydrosheds
2000 2007 90 m GRID free NASA
Hydrosheds Hydrosheds.er.usgs.org
7
Table 2 Information on resolution, scale, format and precision
Resolution 90 mètres (basée sur le DEM SRTM)
Scale 1 : 250.000 based on the data (based on agroecological zones)
Format ESRI GRID
Précision In the tab general, insert the working directory
In the tab extent, select the analysis extent (probably “as …” the baseline
layer)
In the tab extent, Snap extent to (probably “as …” the baseline layer)
In cell size, select the cell size (probably “as …” the baseline layer).
Correction by using Generalization commands have been used as needed.
Validation Results are based on experience of the work group. Results are not validated.
3. Africover re-classification
The LUS map is based on land cover, which represents the ecosystems where that are used by
human being to establish the land use, or to obtain goods and services from an area. The LUS need
to be prepared over a very generalized land unit. The reclassification has been realised by using the
class name (LC_NAME) of the Aggregated Africover version.
The simplification took place in a participatory manner; with the collaboration of national experts
that use Africover in their regular work or that participated on Africover preparation. Thus, new
classes combining two or more Africover classes as summarized in the table below.
One more improvement has been done for the area alongside the Kivu Lake where Coffee and
Bananas plantations are the main crops and should be considered as perennial crop and not
seasonal. For deriving the abovementioned area, map algebra was used as a geo-processing tool.
The following conditions were proposed as input query:
To be at the elevation range of 1460 to 1540 metres. The later was selected from the DEM
Grid in order to delineate the eastern hillside of Kivu Lake. The motivation for choosing this
elevation range is attributed to the fact that the elevation value at lake level is 1460 whereas
its eastern hillside occupied by perennial crops has an estimated denivellation of 100m from
lake level;
That elevation range should be within districts bordering Kivu Lake (including Rubavu,
Rutsiro, Karongi and Nyamasheke) and
The previous land use type should be seasonal crops.
Table 3 Reclassification of Africover classes in Rwanda
LCNAME new_class
Combination of Shrub Plantation and Rainfed Herbaceous Crop (approx. 60-70% and
30-40%)
Forest Plantation
Forest Plantation - (Eucalyptus) - (or Pinus and Cypress)
Forest Plantation (Eucalyptus) or Pinus and Cypress (mixed unit with natural
vegetation or other) (field area approx. 60% polygon area)
Scattered (in natural vegetation or other) Forest Plantation (Eucalyptus) or Pinus and
Cypress (field density 20-40% polygon area)
Scattered (in natural vegetation or other) Shrub Plantation - Undifferentiated (field
density 20-40% polygon area)
Shrub Plantation - Undifferentiated (mixed unit with natural vegetation or other)
(field area approx. 60% polygon area)
8
Shrub Plantation - Undifferentiated
Combination of Forest Plantation and Rainfed Herbaceous Crop - Two crop per year
(approx. 30% each; remaining polygon surface natural vegetation)
Seasonal Crops
Combination of Rainfed Herbaceous Crop - Two crop per year - and Forest Plantation
(approx. 70-80% and 20%-30%)
Combination of Rainfed Herbaceous Crop - Two crop per year and Shrub Plantation
(approx. 40-60% and 20-40%; remaining Forest Plantation - (Eucalyptus) - (or Pinus
and Cypress))
Combination of Rainfed Herbaceous Crop - Two crop per year and Shrub Plantation
(approx. 40-60% and 20-40%; remaining natural vegetation)
Combination of Rainfed Herbaceous Crop and Forest Plantation (approx. 40-60% and
20-40%; remaining natural vegetation)
Combination of Rainfed Herbaceous Crop and Shrub Plantation (approx. 40-60% and
20-40%; remaining natural vegetation)
Combination of Shrub Plantation and Rainfed Herbaceous Crop - Two crop per year
(approx. 40-60% and 20-40%; remaining Forest Plantation - (Eucalyptus) - (or Pinus
and Cypress))
Combination of Shrub Plantation and Rainfed Herbaceous Crop - Two crop per year
(approx. 40-60% and 20-40%; remaining natural vegetation)
Irrigated Herbaceous Crop
Isolated (in natural vegetation or other) Forest Plantation (Eucalyptus) or Pinus and
Cypress (field density 10-20% polygon area)
Isolated (in natural vegetation or other) Rainfed Herbaceous Crop - Two Crop Year -
(field density 10-20% polygon area)
Isolated (in natural vegetation or other) Rainfed Herbaceous Crop (field density 10-
20% polygon area)
Post Flooding Herbaceous Crop (mixed unit with natural vegetation or other) (field
area approx. 60% polygon area)
Post Flooding Herbaceous Crop
Rainfed Herbaceous Crop - Two Crop Year (mixed unit with natural vegetation or
other) (field area approx. 60% polygon area)
Rainfed Herbaceous Crop
Rainfed Herbaceous Crop -Two Crop Year
Scattered (in natural vegetation or other) Rainfed Herbaceous Crop - Two Crop Year
- (field density 20-40% of polygon area)
Scattered (in natural vegetation or other) Rainfed Herbaceous Crop (field density 20-
40% of polygon area).
Closed Herbaceous Vegetation
Grasslands Savannah (shrub or tree and shrub)
Sparse Herbaceous Vegetation
Rice Fields Irrigated Crops
Closed Shrubs
Natural_forest
Closed Trees
Multilayered Trees Broadleaved Evergreen
Open Broadleaved Deciduous Trees
Open Shrubs
Combination of Banana Plantation and Rainfed Herbaceous Crop (approx. 60-70%
and 30-40%) Perennial Crops
Tea Plantation
Closed Herbaceous Vegetation
Savanna
Open Shrubs (on temporarily flooded land - fresh water)
Open Shrubs
Savannah (shrub or tree and shrub)
Sparse Herbaceous Vegetation
Water Bodies Surface water
Urban And Associated Areas Urban and built up areas
Closed Herbaceous Vegetation On Permanently Flooded Land - Fresh Water
Wetland Closed Herbaceous Vegetation With Sparse Trees in Temporarily Flooded Land -
Fresh Water
9
4. Validation of the land cover classes
The validation was prepared by using the following GIS data layers obtained from Rwanda Natural
Resource Authority/Land and Mapping Department: Forest Plantations
Protected wetlands
Wetlands
Irrigated crops
Vegetation and land cover
Natural forests
Built-up area
Agriculture zones Settlements
Also, the validation team was able to use the Land and Mapping Department office, the recent
(2008), high resolution aerial photographs (25 cm) for areas where the land cover type was not clear
or confusing. We therefore did not find it necessary to make field trips to validate the land cover
map.
Main modification of the land cover classes are presented in table 4.
Table 4 Changes made to the Land Cover map of Rwanda during validation exercise
Administrative Unit (former
Provinces)
Changes made to Land Cover
Classes
Remarks
Kigali More area under Built up areas
mapped
Kigali City has greatly expanded
East Woodland ecosystem class
introduced
Were previously classified as
Natural Forest in the Land Cover
map. We reclassified them as
Woodlands as they are mostly
dominated by nearly same size/age
Acacia trees and are not layered like
the natural forests. Occur along river
courses and in the south in Bugesera
Grassland ecosystems removed These areas have same classification
in Africover map as savanna. The
grassland ecosystems were therefore
merged with savanna
More Built up Areas mapped Large settlements especially in the
north east were lacking in the Land
Cover map
More areas under Perennial crops
mapped
Perennial crops especially bananas
and coffee are being adopted,
especially in the south east and north
east eating up areas that used to be
under savanna and/or seasonal crops
More areas under Seasonal crops
mapped
The Savanna areas have been
reduced in recent past by converting
them to seasonal crops
More areas under Forest Plantations
mapped
Because majority of the individual
forest plantations are small in area,
they are normally not mapped in low
resolution maps. But because they
are many and cover a wider
landscape
More areas of Perennial crops
mapped
Areas under pyrethrum plantations,
and tea outside the irrigated areas
10
had not been mapped
More areas under Forest plantation
mapped
Because majority of the individual
forest plantations are small in area,
they are normally not mapped in low
resolution maps. But because they
are many and cover a wider
landscape
North Protected wetlands mapped Protected wetlands were not
originally mapped. There is a recent
Government policy to protect
wetlands that had not yet been
drained for agriculture
Grassland ecosystems removed Areas classified as grasslands were
successional stages of natural forest
or areas under herbs/shrubs in forest
plantations that were yet to be
planted or areas under fallow or
uncultivated areas in seasonal crops
ecosystem
More areas under Forest plantation
mapped
Because majority of the individual
forest plantations are small in area,
they are normally not mapped in low
resolution maps. But because they
are many and cover a wider
landscape
West More areas of Perennial crops
mapped
Plantations of coffee, and some
bananas are being established,
especially along the shores of Lake
Kivu
Protected crops class removed These were areas that had been
cultivated within the boundaries of
protected areas like Gishwati forest
but now are being restored to natural
forest or forest plantation
More areas under Forest plantation
mapped
Because majority of the individual
forest plantations are small in area,
they are normally not mapped in low
resolution maps. But because they
are many and cover a wider
landscape
More area under Built up areas
mapped
Settlements and towns have greatly
expanded e.g. Gisenyi
South More areas under Forest plantation
mapped
Because majority of the individual
forest plantations are small in area,
they are normally not mapped in low
resolution maps. But because they
are many and cover a wider
landscape
More areas under Perennial crops
mapped
Banana plantations are being
established
A map of reclassified land cover of Rwanda at 1/100000 was prepared by local experts. Ten (10)
new classes were created as shown on the figure below.
11
Figure 1Reclassified land cover map
12
5. Other data used to prepare the land use map
5.1 Protected area
The protected areas were derived from the World database of protected areas, 2008.
Figure 2 Rwanda protected areas
13
5.2 Classification of livestock statistical data
Available livestock data of 2003 (2008) have been used as baseline.
Table 5 Livestock heads in Rwanda in 2003
NAME/Province Cattle Sheep Goats Pigs
Butare 85396 8270 310340 43391
Byumba 76392 110885 193543 10294
Cyangugu 60092 29068 169889 53871
Gikongoro 69674 82171 134606 68355
Gisenyi 49350 123299 290140 93436
Gitarama 165416 42068 213160 60142
Kibungo 62581 3118 380120 19988
Kibuye 66358 76327 223620 38393
Kigali Ngari 120209 42000 386029 41505
Kigali Ville 0 0 0 0
Ruhengeri 84152 174053 188988 77108
Umutara 282559 4108 165363 21048
In table above, Kigali as the capital city was not covered by the inventory.
Those data has been converted to tropical livestock units (TLU) using the same conversion factors
as defined by NBI-NELSAP for the Kagera basin: cattle 0.75, goats 0.1, sheep 0.1, pigs 0.2. With
the aim of having a data that is comparable between species and districts of different dimension, the
results have been calibrated basing on extent, obtaining an indication of TLU/km2
Table 6 Tropical livestock units (TLU) per Km2
NAME/Province Cattle_TLU/km2 Sheep_TLU/km
2 Goats_TLU/km
2 Pigs_TLU/km
2
Butare 64047 827 31034 8678.2
Byumba 57294 11088.5 19354.3 2058.8
Cyangugu 45069 2906.8 16988.9 10774.2
Gikongoro 52255.5 8217.1 13460.6 13671
Gisenyi 37012.5 12329.9 29014 18687.2
Gitarama 124062 4206.8 21316 12028.4
Kibungo 46935.75 311.8 38012 3997.6
Kibuye 49768.5 7632.7 22362 7678.6
Kigali Ngari 90156.75 4200 38602.9 8301
Kigali Ville 0 0 0 0
Ruhengeri 63114 17405.3 18898.8 15421.6
Umutara 211919.3 410.8 16536.3 4209.6
5.2.1 Use of Agro – ecological zones of Rwanda as a base for livestock distribution
Livestock map using statistics per administrative unit does not reflect the realistic spatial
distribution. Using our experience we managed to improve spatial pattern, distribution of livestock
14
using agro –ecological zones of Rwanda. The agro –ecological zones maps has been created by
Ministries of Agriculture (MINAGRI) in 80’s at a scale of 1/250 000 with ten classes as illustrated
in the figure bellow.
Figure 3 Agro-ecological zones of Rwanda defined by MINAGRI
5.2.2 Livestock intensity
The livestock map has been prepared following 2 phases:
Creating a density map of livestock distribution per region using tropical livestock unit
based on 2003 statistics
Creating livestock intensity map based on livestock map per administrative unit and spatial
distribution using agro-ecological zones judging using our experience.
The scoring table was created as follow:
Table 7 Scoring livestock dominance per province in 2003
les savanes de l'Est les hautes terres de
Burebuka les cones et hautes plaines
volcaniques les cretes et plateaux
bordant les savanes de l'Est
Province CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS
Butare
Byumba y y y Y Y Y
Cyangugu
Gikongoro
Gisenyi y y
Gitarama y y
Kibungo y y Y y
15
Kibuye
Kigali Ngari Y Y y
Kigali Ville
Ruhengeri y y
Umutara Y y y Y Y
la crete Zaire-Nil le Plateau Central le bord du lac Kivu les hautes terres de
Burebuka
Province CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS
Butare y Y
Byumba Y Y Y
Cyangugu Y Y Y
Gikongoro Y Y Y Y
Gisenyi Y y y y Y y
Gitarama y Y Y
Kibungo
Kibuye Y y y Y
Kigali Ngari y
Kigali Ville
Ruhengeri y y y
Umutara
le Mayaga et Bugesera
peripheriques l'arriere pays de Cyangugu la Plaine de Bugarama
Province CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS CATTLE GOATS SHEEP PIGS
Butare Y y
Byumba
Cyangugu y y y
Gikongoro
Gisenyi
Gitarama Y Y
Kibungo y y
Kibuye
Kigali Ngari Y Y
Kigali Ville
Ruhengeri
Umutara
16
Figure 4 Livestock intensity map (source data also include livestock statistics)
17
6. Land use map implementation
The baseline for land use preparation is land cover. The preparation of land use has been prepared
using livestock intensity, protected areas and wetlands areas. Data have been classified using the
conditional command (“CON”) of ArcGIS Spatial Analyst. In total twenty seven (27) major classes
for land use systems have been created as shown in the table below.
Table 8 Main land use systems
Code/Value Land Use Classes
1 Protected natural forest
3 Natural forest with livestock
4 Protected forest plantation
6 Forest plantation with livestock
7 Protected Savana
8 Savana with high livestock
9 Savana with moderate livestock
10 Protected Grassland
12 Grassland with livestock
13 Protected wetland
15 Wetland with livestock
16 Crop in protected area
18 Perennial Crops with livestock
19 Seasonal Crops with high livestock
20 Seasonal Crops with moderate livestock
22 Irrigated Crops with livestock
23 Protected Surface water
25 Surface water with other uses
27 Urban area
Land use classes qualified as having:
High livestock intensity were those having a threshold of above forty seven (47) tropical
livestock units per km2,
Moderate livestock intensity were those having a threshold of below forty seven (47)
tropical livestock units per km2,
No livestock
7. Land use map improvements
The land use map is needed in the second part of the LADA (FAO/UNEP GEF) method, that
include the assessment of the land degradation and of the sustainable land management practices.
This exercise is realised using the participatory method LADA WOCAT QM throughout the
involvement of local experts. During the assessment LUS units are calibrated and improved basing
on participants suggestion. The final LUS map is presented in Figure 5.
18
Figure 5 Land Use System Map of Rwanda
19
8. Land use map database implementation
The database will be utilized during land degradation and sustainable land management assessment
as supporting material of the LADA WOCAT QM method. It gives additional information for the
groups of assessment. The use of the GRID formal allows converting all data to the baseline
resolution. The list of input data is available in table 10.
8.1 Livestock natural region and livestock species
The map of livestock region was prepared basing on the description in chapter 4.2. In the
preparation of the land use map, the carte has been used as input. With a slight modification and
following the same method than is chapter 4.2, the map can list the dominant livestock species. The
groups of dominant livestock species are the following:
Cattle and goats
Sheep
Cattle, sheep and goats
Sheep and pigs
Cattle and sheep
Goats and pigs
Cattle
Goats
8.2 Annual rainfall
The annual average rainfall amount is calculated by summing up monthly values downloaded from
the World Clim database, published in 2005. Although the data is at 30 arc seconds resolution, the
data is considered the best available to have an indicative representation of the rainfall of the area.
The data is presented in Figure 6.
8.3 Annual temperature
The annual temperature range is calculated from data downloaded from the World Clim database,
published in 2005. Although the data is at 30 arc seconds resolution, the data is considered the best
available to have an indicative representation of the rainfall of the area. The data is presented in
Figure 6.
8.4 Elevation and slope
The elevation (meters) at a resolution of 90 meters is obtained using SRTM 2000 data included in
the Hydrosheds database published in 2007. The slope (degrees) is interpolated from the same
dataset. The data are presented in Figure 7.
The elevation map use the following elevation classes in the database: 900-1000,1000-1400,1400-
1800,1800-2200,2200-3000,> 3000. The slope classes are: 0-2,2-8,8-16,16-32,>32.
8.5 Soil
The soil map is downloaded from the Harmonized would soil database with a resolution of 30 arc
seconds and is presented in Figure 8.
20
Table 8 Input data for LUS database preparation
Used for Name Year of
measure
Year / periodicity
of publication
Scale / resolution
/ extent Format
Availability or
copyright Producer Metadata
Unity of measurements
Admin units RWA (District) 2006 - 1:250.000 Shp free MINALOC none classes
Temperature Wordclim - 2005 30 arc
seconds GRID free www.worldclim.org www.worldclim.org degree
centigrads
Livestock density
Livestock statistics
2006 yearly Old
provinces (2001)
stats free National Institute
of Statistics report
heads per species
Livestock natural distribution regions
Agroecological zones
1980 - 1:250.000 shp free MINAGRI none classes
Rainfall Wordclim - 2005 30 arc
seconds GRID free www.worldclim.org www.worldclim.org mm
Elevation SRTM Hydrosheds
2000 2007 90 m GRID free NASA
Hydrosheds Hydrosheds.er.usgs.org m
Slope SRTM Hydrosheds
2000 2007 90 m GRID free NASA
Hydrosheds Hydrosheds.er.usgs.org degree
Soil HWSD 2008 2008 30 arc
seconds GRID free
FAO
IIASA
ISRIC
ISSCAS
JRC
http://www.iiasa.ac.at/Research/LUC/External-World-soil-database/HTML/ -
Population density
population statistics
2002 yearly - stats country ? inhabit /sq km
Poverty Poverty heard index
2005-6 - - stats free
National
institute of
statistics of
Rwanda
none EICV
21
Figure 6 Annual rainfall and temperature range
Figure 7 Elevation and slope
22
Figure 8 Soil map
8.6 Population density
Population data (inhabitants per districts) are available from the national census for the year 2002.
To be able to compare data between district, the population density has been calculated by dividing
the population for the area in square kilometres. The list of classes used in the database is the
following: 55-100,100-300,300-500,500-800,>800. The map is presented in figure 9.
8.7 Poverty
In 2007 the National institute of statistics of Rwanda published the report “EICV poverty analysis
for rwanda's economic development and poverty reduction strategy” including the assessments
derived from the Enquete Intégrale sur les Conditions de Vie des ménages de Rwanda, that provide
as a result the indicator Poverty headcount by province (EICV2). From this report, the values in
percentages have been converted to the map presented in figure 9. The classes used in the map are:
20-30,30-60,60-65, >65.
23
Figure 9 Map of socio-economical indicators including population density and poverty
