Ionospheric modeling: data assimilation

and data ingestion

Sandro M. RadicellaAeronomy and Radiopropagation LaboratoryAbdus Salam International Centre for Theoretical Physics

First Lecture

IONOSPHERIC MODELS

Introduction

SOURCES OF ENERGY IN THE IONOSPHERE:

Solar photons (EUV and X) upper atmospheres neutrals

Excess energy of photoelectrons

Other sources of ionization and/or energy at middle and high latitudes:

ionize

transferred via collisional processes thermal electronsionsneutrals

magnetospheric electric field particle precipitation field-aligned currents

GEOMAGNETIC EFFECTS:chemical

transport } processesradiative

are modulated byGEOMAGNETIC FIELD

High latitudes

Middle latitudes

Low latitudes

polar cap tongues of ionizationauroral Ne enhancementspolar holestroughs

smooth diurnal variationslarge variations at

sunrise and sunset

Equatorial Anomaly

IONOSPHERIC BACKGROUND (CLIMATE)

!"#$%&&'(')*

!"#$%"!&'"&()*&(+,-../0

!"#$%&&'(')*

12345.,567589

+,-. /"012*3#0'2"

.:/:.5 %;45$2<=;>?5 @6A6?B;4<45 @?5 25 <43@6A5 675 @6A@?4C532?5 =;2=5 4D=4AC?5 7<6E5F8,5GE5 HIJ<43@6AK5 =65F.,,,5GE5 HLJ<43@6AK5 @A5 2M=@=NC45 HL@3N<45 .J.K:5 "=5 @EB2O=?5 6A5 =;45 C4?@3APB4<76<E2AO452AC56B4<2=@6A5675E2AQ5+L5?Q?=4E?56B4<2=@A35R4M6S5.5TUV52AC54W4A;2?5 ?6E45 @EB6<=2A=5 4774O=?5 NB5 =65 .,5TUV:5 %;45 C4A?@=Q5 675 @6A@?2=@6A5 W2<@4?5 2?5 27NAO=@6A5675 =@E45675C2QP5?42?6AP5Q42<P5 M2=@=NC454=O:5%<2C@=@6A2MMQP5 =;<4453463<2B;@O<43@6A?5 2<45 O6A?@C4<4C5 S;4A5 C4?O<@R@A35 =;45 @6A6?B;4<4X5 A2E4MQ5 ;@3;P5 E@C5 2AC4YN2=6<@2M5 M2=@=NC4?P52AC5 =;4?45E2Q5 @A5 =N<A5R45?NRC@W@C4C5 HL@3N<45.JZK:5[=NC@4?567=;45 3463<2B;@O5 2AC5 =4EB6<2M5 W2<@2=@6A5 675 =;45 @6A6?B;4<45 2AC5 @=?5 @EB2O=5 6A5 +L?Q?=4E?5 67=4A5 A4O4??@=2=45 M6A3J=4<E5 ?=NC@4?5 2AC5 @A=4<A2=@6A2M5 <4?42<O;5 O6J6B4<2=@6A:5 5 \CC@=@6A2MMQP5 ?Q?=4E?5 =;2=5 B<6B232=45 W@25 6<5 =;<6N3;5 =;45 @6A6?B;4<@O4AW@<6AE4A=5 67=4A5 A4O4??@=2=45 @A=4<A2=@6A2M5 BM2AA@A35 ?O;4E4?5 2AC5 O6J6<C@A2=@6A23<44E4A=?:

!"#$%&'()('*+&',-.&%&/'01%$21$%&'"3'1+&'453567+&%&

4/567

83121%$.9:%$

7123;<=2$%1

>%?

@A3%021'%$

B'*CD%0'03*)

E8F

!"#$%&'()8'9-%"5$6':&#"536'"3'1+&'453567+&%&

.:/:Z5 U@?=6<@O2MMQP5 ;@3;5 7<4YN4AOQ5 HULK5 ?Q?=4E?5 S6NMC5 ;2W45 R44A5 @C4A=@7@4C5 2?5 =;4B<@E2<Q5 ?Q?=4E5 2774O=4C5 RQ5 =;45 @6A6?B;4<4:5 U6S4W4<P5 =;4<45 2<45 E2AQ5 6=;4<O6A=4EB6<2<Q5 ?274=QP5 BNRM@O5 ?4<W@O4P5 O6EE4<O@2M5 2AC5 E@M@=2<Q5 +L5 ?Q?=4E?5 S;@O;2<45 2M?65 2774O=4C:5 %;4?45 @AOMNC45 E4=46<5 RN<?=P5 ?2=4MM@=45 O6EENA@O2=@6A?P5 3M6R2MA2W@32=@6A5?2=4MM@=45?Q?=4E?5H?NO;52?5T1[52AC5T2M@M46KP5?B2O4JR2?4C5 <2C2<?52AC2M=@E4=4<?P5 M6A3P5 E4C@NE5 2AC5 ?;6<=S2W45 <2C@65 R<62CO2?=@A35 2AC5 E2<@=@E4O6EENA@O2=@6A?P:5%;4?45<2C@657<4YN4AOQ5?Q?=4E?5;2W45<4M4W2AO45=65R6=;5]'52ACA6AJ]'5O@W@M52AC5E@M@=2<Q52BBM@O2=@6A?:5

.:/:-5 %;45]'5;2?525S6<MC5 M42C@A3536W4<AE4A=P5 @ACN?=<Q52AC52O2C4E@O5O2B2R@M@=Q5 @A5 =;4@6A6?B;4<452AC5@6A6?B;4<@O5?Q?=4E5<4?42<O;:5"A5=;45B2?=5=;45\34AOQ5;2?5C<2SA56A

IONOSPHERIC WEATHER:

Superimposed to the ionospheric climate features there is a wide range of weather like characteristics: day-to-day, hour-to hour, and inter-hourly variations at any given location.

over-the-horizon radarHF communicationssatellite surveying and navigation systemssurveillancesatellite tracking and lifetimepower gridspipelines

the

abdus salam international centre for theoretical physics

united nations,

educational scientific

and cultural

organization

international atomic

energy agency

10

Diurnal and day-to-dayvariability of NmF2

Variation of NmF2 at Slough for

every day during four 2-month

periods in 1973-1974.

H. Rishbeth, M. Mendillo /

Journal of Atmospheric and

Solar-Terrestrial Physics 63

(2001) 1661–1680

Ionospheric weathereffects onhuman activities

Ionospheric modeling

The understanding of the behaviour of the ionosphere and its effects on human activity is determined by the ability to model at least the height, geographical and time distributions of the electron density.

There is no numerical code or model that is able, at present time, to describe accurately both the three-dimensional and time-dependent distribution of the ionospheric plasma.

In other words, no model is able to reproduce in a satisfactory way both the climate and the weather of the Earth ionosphere.

In addition, there is no well established experimental database that can be used to verify and test the existing models in order to generate the improvements needed.

At present most models are able only to reproduce consistently the climate of the ionosphere, as defined mostly by diurnal, seasonal and solar cycle variations.

TYPES OF MODELS

Theoretical or first-principle models

Empirical or semi-empirical models

Analytical “profilers”

THEORETICAL OR FIRST-PRINCIPLE MODELS

In these models conservation equations (continuity, momentum, energy, etc.) are solved numerically as a function of spatial and time coordinates to calculate plasma densities, temperatures and flow velocities.These models require magnetospheric and atmospheric input parameters and its accuracy depend on the quality of the input data. They can be powerful tools to understand the physical and chemical processes of the upper atmosphere. Examples of models:

The Sheffield Coupled Thermosphere Ionosphere Plasmasphere Model (CTIP)

Time dependent middle-latitudes model

 

SHEFFIELD CTIP MODEL

A three-dimensional, fully coupled, numerical model of the thermosphere, low-latitude plasmasphere and high-latitude ionosphere system based on solving the equations of continuity, momentum and energy balance (Fuller-Rowell et al., J. Geophys. Phys., 92, 7744, 1987). The current versions of CTIP have resulted from over twenty years of development. The model was enhanced at Sheffield to account for the mid and low-latitude ionosphere and to include a self-consistent plasmasphere.

The spatial resolution is 2 degrees latitude by 18 degrees longitude by 1 scale height (extending from a tidally-forced lower boundary at 80 km/1 Pa).

The temporal resolution in the standard version is normally 1 minute for the thermosphere and ionosphere, and 15 minutes for the plasmasphere. Species concentration solved include the atomic species O, H and N; the atomic ions O+, H+, N+ and He+; the molecular species O2 and N2, and the molecular ions NO+, O2+ and N2+.

Calculated and derived quantities include also three-dimensional ion and neutral winds, ion, neutral and electron temperature,Joule Heating, Lorentz Forcing, and Hall and Pedersen conductivity.

CTIP Model results of electron density for Day 355 during a geomagnetic disturbance.

http://www.aber.ac.uk/propag/theory.html

TIME DEPENDENT MIDDLE-LATITUDES MODEL:

It solves a set of equations involving different ions and electrons. Both dynamic and photo-chemical processes are considered and stable and meta-stable species are taken into account (Zhang S. R., et al., Annali di Geofisica, 36, 5-6, 105, 1993. and Zhang S. R. and S. M. Radicella, Annali di Geofisica, 36, 5-6, 111, 1993). The model depends on EUV91 model data for solar radiation flux, MSIS 86 for neutral concentration and temperature and HWM 90 for neutral horizontal winds.

Inputs are geographical latitude and longitude, day number, Ap index and daily and 81-day mean of F10.7.

Output are electron and ion concentrations at a function of time and altitude.

Model runs fast in any PC (one day run, every 15 min and at 2 km altitude interval, is done in about 2 min)

EXAMPLE OF MODEL OUTPUT: NOON AND MID-NIGHT CONCENTRATIONS

S.-R. Zhang et al.; Adv. Space Res. Vol 18, n° 6, pp. 165-173,1996

EMPIRICAL AND SEMI-EMPIRICAL MODELS

Based on a analytical description of the ionosphere with functions derived from experimental data or adapted from physical models. Two global scale models:

IRI: International Reference Ionosphere

PIM: Parameterized Ionospheric Model

IRI

An international project sponsored by the Committee on Space Research (COSPAR) and the International Union of Radio Science (URSI), which aimed at producing a reference model of the ionosphere based on available experimental data sources. IRI is updated periodically and has evolved over a number of years. (D. Bilitza, Radio Science 36, #2, 261-275, 2001)

IRI data sources are :

coefficients (foF2 and M(3000)) produced by ITU-R (former CCIR) from a large number of ground based sounders,

incoherent scatter radars (Jicamarca, Arecibo, Millstone Hill, Malvern, St. Santin) measurements,

ISIS and Alouette topside sounders ionograms,

in situ measurements by several satellites and rockets.

It can also use as input experimental values of F2 peak electron density (or foF2) and height .

IRI OUTPUT:

For given location, time and date, IRI gives in the altitude range from about 50 km to about 2000 km:

electron concentration,

electron temperature,

ion temperature,

ion composition

total electron content (TEC).

It provides monthly averages in the non-auroral ionosphere for magnetically quiet conditions.

IRI

The graph attempts to illustrate two parameters at once at 14.00 UT and 400 km altitude. The electron density is scaled by the intensity. The electron temperature is indicated by the hue. The sun position is approximately indicated by an asterisk. The map grid is Cartesian in Geodetic Coordinates. The dotted lines indicate Geomagnetic Coordinates.

From: powerweb.grc.nasa.gov/.../ info/movies/iri90.html

PIM

is a fast global ionospheric and plasmaspheric model based on a combination of the parameterized output of several regional theoretical ionosphere models and an empirical plasmaspheric model (Daniell et al., Radio Sci., 30, 1499-1510, 1995).

In 1997 the Gallagher plasmaspheric model (Gallagher et al., Adv. Space Res., 8, 15-24, 1988.), a fast empirical model of plasmaspheric H+, was incorporated into PIM.

From a given set of geophysical conditions (day of the year, solar activity index f10.7, geomagnetic activity index Kp) and

positions (latitude, longitude, and altitude), PIM produces electron density profiles between 90 and 25000 km altitude, corresponding critical frequencies and heights for the ionospheric E and F2 regions, and Total Electron Content (TEC).

It represents the climatological portion of the Parameterized Real-time Ionospheric Specification Model (PRISM).

PIM fc contours at 390 km, 18 March 03

Zhang Rui Xue, et al., Workshop on Application on Radio Science, 18-20 February 2004, 18-Feb-04. DSO National Laboratories, Singapore

DIFFERENCES BETWEEN EMPIRICAL AND PARAMETERIZED THEORETICAL MODELS:

Both type of models give the climatology of the ionosphere

Empirical climatology gives an “average” ionosphere over potentially different ionospheric conditions that could be smeared by the averaging process.

Empirical climatology is limited by the amount of data and their spatial and temporal distribution.

Parameterized theoretical climatology yields a “representative” ionosphere with features similar to those that might be observed on any given day under specific geophysical conditions.

Theoretically derived climatology is limited by the accuracy and completeness of the physics and chemistry included in the models.

“PROFILERS”

Another way to estimate the electron concentration distribution with altitude is to express an altitude profile in terms of simple mathematical functions, adjusted to ionospheric characteristics routinely scaled from the ionograms. The advantage of this type of "profile modelling" is that it can in principle use as inputs simply experimental values of basic ionospheric characteristics for both quiet and disturbed conditions. Examples of profilers:

Bradley and Dudeney 1973

Dudeney 1978

Di Giovanni and Radicella 1990

Family of Trieste-Graz profilers

BRADLEY AND DUDENEY 1973

Describes the electron concentration profile up to the peak of the F2 region with two parabolic layers for the E and F2 layer and a linear segment in between. This profile generation is still used by the ITU-R HF propagation prediction method (Bradley, P.A. and J. R. Dudeney, J. Atmos. Terr. Phys., 35, 2131, 1973).

DUDENEY 1978

A more refined profiler that uses routinely scaled characteristics, incorporates combinations of trigonometric functions segments, and provides optionally valley and F1 description. Electron concentration gradient with altitude is continuous (Dudeney, J. R., J. Atmos. Terr. Phys., 40, 195, 1978)

DI GIOVANNI AND RADICELLA (DGR) 1990

Describes the electron concentration profile in the E-F1-F2 regions of the ionosphere by using simple analytical expressions. It is constructed as the sum of three Epstein layers that are formally identical considering the existence of characteristic points in the profile with co-ordinates (values of electron concentrations and its height) calculated by means of empirical expressions. It gives the electron concentration profile above the F2 peak making use of an effective shape parameter empirically derived for the topside ionosphere. Total electron content is computed with an analytical expression. This model has been adopted by the European Commission COST 238 (PRIME) action (Di Giovanni, G. and S. R. Radicella, Adv. Space Res., 10, No. 11, 27-30, 1990, and Radicella, S. M. and M. L. Zhang, Annali di Geofisica, 38, 1, 35-41, 1995).

DGR PROFILES

M.M. de Gonzalez, Adv. Space Res. vol 18, n° 6, pp. 53-56, 1996

TRIESTE-GRAZ FAMILY OF MODELS

NeQuick, a quick-run model particularly tailored for transionospheric propagation applications;

COSTprof, a more complex model for ionospheric and plasmaspheric satellite to ground applications;

NeUoG-plas, a model to be used particularly in assessment studies involving satellite to satellite propagation of radio waves.

COSTprof model was adopted by the European COST 251 action. NeQuick model has been used by the ESA EGNOS program and adopted by Recommendation P.531-6 of the International Telecommunication Union-Radiocommunication sector (ITU-R) as a suitable method for TEC modelling.

Are particularly suitable for assessment studies in connection with advanced satellite navigation and positioning systems. They can be used to assess various radiopropagation effects observed in received satellite signals.

They can simulate space and time regional ionospheric variations produced by geomagnetic storms, TIDs or trough, and their effects on transionospheric propagation (S. M. Radicella, and R. Leitinger, , Adv. Space Res., Vol. 27, n° 1, Pages 35-40 (2001)).

CHARACTERISTICS OF THE MODELS

They consist in two height regimes:

The bottomside model for the height region below the peak of the F2-layer use a modified DGR profile formulation which includes 5 semi-Epstein layers with modelled thickness parameters and is based on foE, foF1, foF2 and M(3000)F2 values. Different sources for these ionosonde parameters can be considered depending on the purpose.

The topside model for the height region above the F2-layer peak uses different approaches: Nequick: a semi-Epstein layer with a height dependent thickness parameter. COSTprof: three physical parameters: oxygen scale height at the F2 peak, its height

gradient, O+–H+ transition height, modelled according to solar activity, season, local time and ”modified dip latitude”. NeUoG-

plas: uses a magnetic field aligned formulation for an H+ diffusive equilibrium for plasmaspheric heights above 2000 km.It takes over the (hydrogen) scale height found at the field line foot point at 2000 km.

NeUoG-plas has an additional geomagnetic field aligned third part for the “plasmasphere”.

INPUT PARAMETERS:

ITU-R (former CCIR) coefficients for foF2 and M(3000)F2 and R12 and/or monthly mean F 10.7,

Measured values of foE, foF1, foF2 and M(3000)F2 or F2 peak concentration and peak height,

Regional maps of foF2 and M(3000)F2 based on grid values constructed from data obtained at given locations.

OUTPUT PARAMETERS:

Electron concentration vertical profile to a given height (including the GPS satellite altitude).

Electron concentration along arbitrary ground to satellite or satellite to satellite ray paths.

Vertical Total Electron Content (vTEC) up to any given height.

Slant Total Electron Content between a location on earth and any location in space.

PROFILES

NeQuick model profile is indicated with dashed line. COSTprof with dotted line and NeUoG-plas with solid line.

SLANT PROFILES OBTAINED WITH NeQuick MODEL

Nequick electron concentration along arbitrary groundto satellite ray paths for a satellite at 20000 km altitude.

Ground location: Lat: 30° N and Long. 10° E,elevation angles 10°, 30°, 60° and 90°.

VERTICAL TEC MAPS OBTAINED WITH NeQuick MODEL

IONOSPHERIC CORRECTION ON SLANT TEC USING NeQuick

Experimental single satellite slant TEC (green) and model calculation (blue) for one day as observed from Ankara.

The RMS ionospheric correction for the day reaches 82.4 %.

TROUGH MODULATION ON SLANT NeUoG-plas MODEL TO THE HEIGTH OF A GPS SATELLITE

FINAL REMARKS

Theoretical, empirical or semi-empirical models and “profilers” in their present state are not able to describe the variability of the ionosphere that define the “ionospheric weather”.

The way to follows is to assimilate or ingest data in the models to specify the ionospheric conditions globally or regionally at a given time.

THIS IS THE TOPIC OF THE SECOND LECTURE

Ionospheric modeling: data assimilation

and data ingestion

Sandro M. RadicellaAeronomy and Radiopropagation LaboratoryAbdus Salam International Centre for Theoretical Physics

Second Lecture

DATA ASSIMILATION AND INGESTION IN MODELS

Introduction

The ionospheric models mentioned in the previous lecture are able to reproduce reasonably well the climate of the ionosphere.

For most of the applications what is required is the “ionospheric specification” that can be defined as the state of the ionosphere globally or regionally at a given time in terms of the ionospheric characteristics of interest. These characteristics could be: 3D electron density profile parameters (peak values), actual electron density profile, total electron content.

In other words what is required is a good representation of the ionospheric weather.

ABOUT IONOSPHERIC SPECIFICATION

If the time of the “specification” is the present, then “ionospheric nowcasting” is needed.

If the time of “specification” is ahead of the present, then “ionospheric forecasting” is requiered.

A full 3D “ionospheric specification” cannot be obtained only from experimental data due to the randomness and scarceness of the observation sites, regardless of the type of observations involved.

Techniques that should be able to assimilate or ingest experimental data into models, in order to give the full 3D representation of the ionosphere at the time required, are needed.

DATA ASSIMILATION AND DATA INGESTION

Data assimilation and data ingestion are terms normally used to express the same concept.

In this lecture a different meaning is given to the two terms.

Data assimilation is considered the process of assimilating data from different instruments and techniques into theoretical or first-principle physical models of the ionosphere. Data kinds can be from ionosonde derived electron density profiles to in situ measurements to slant path total electron content to neutral winds and neutral densities,

Data ingestion is called the process of ingesting single-technique ionospheric data like vertical or slant total electron content into empirical, semi-empirical or “profilers” models of the ionosphere.

Both assimilation and ingestion techniques require an initial guess provided generally by a “climate” description of the ionosphere.

Assimilation

GAIM: Global Assimilative Ionospheric Model (Chunming Wang, et al, Radio Sci., 39, RS1S06, doi:10.1029/2002RS002854, 2004)

Is the more ambitious and complex effort to assimilate data in ionospheric theoretical models.The research efforts on ionospheric data assimilation by the groups involved in GAIM began in 1996 as a collaborative research project involving: Jet Propulsion Laboratory (JPL), University of Southern California (USC), Cornell University, University of Sheffield.

GAIM HISTORY

As an exploratory research project, the goal was to look at the feasibility of combining ground and space-based GPS total electron content (TEC) measurements with the theoretical Sheffield Plasmaspheric and Ionospheric model to produce ionospheric specifications.

In 1999 a Multidisciplinary University Research Initiative project supported the development of the Global Assimilative Ionospheric Model (GAIM) by two consortia, one led by USC/JPL and the other led by Utah State University.

GAIM OBJECTIVE AND EXPECTED OUTPUT

GAIM’s objective is to assimilate multiple types of ionospheric measurements including ground and space-based GPS, ionosonde profiles, UV airglow radiances, in situ electron and neutral densities, plasma drifts, neutral winds, neutral densities, or other types of measurements creating the equivalent of numerical weather prediction (NWP) models.

It is expected that assimilation of ionospheric measurements into mature first-principles ionospheric models will produce physically consistent accurate ionospheric weather forecasts.

The current operational objective of GAIM is to provide real time or near-real time ionosphere specification (i.e., the ionospheric condition at the time when measurements were taken) including short-term (1 hour to 1 day) ionospheric forecast.

MAIN COMPONENTS OF GAIM ASSIMILATION MODEL

Driving forces are neutral wind, ExB drift, etc.

The forward ionospheric model is a global three-dimensional time-dependent model in which the main model variables are the ion densities. An operational environment where diverse types of ionospheric data are available that are first preprocessed and edited to reduce the outliers and noise. The main tool for automatic data quality control is statistical analysis. The optimization subsystem is responsible for adjusting the values of the model variables as well as the driving forces so that the predicted measurements based on the adjusted values are closer the actual measurement.

POTENTIAL DATA SOURCES FOR GAIM

SOME RESULTS:

F2 peak densities,

13 June 2001

CONCLUSIONS BY THE GAIM TEAM

“The development of GAIM is still work in progress. Although the main components of the system are implemented, considerable work still needs to be carried out to fine-tune the algorithms and error statistics.”

Ingestion

DATA INGESTION IN NeQuick MODEL (1)(Bruno Nava, et al, Proceedings of Atmospheric Remote Sensing Using Satellite Navigation Systems, Matera, Italy, 2003)

A new ingestion technique to reconstruct the electron concentration of the ionosphere has been developed using the profiler model NeQuick.

The relevant NeQuick feature for this use is that, for a given location and at a given epoch, the model is driven by an “ionisation level index” Az, which determines the vertical electron density profile characteristics by anchoring the profile to the F2 region peak.

Since for the given epoch and location, NeQuick is a function of this ionization parameter alone, it is possible to define Az as the value that minimizes the difference between an experimental and modelled vertical TEC.

THE TECHNIQUE USED

Applying the concept indicated to all vertical TEC (vTEC) values of a global experimental vTEC map, it is possible to retrieve an Az grid, which drives NeQuick to generate a 3D representation of the ionosphere over the globe that exactly reproduces the starting vTEC map.

Starting from each Az value, the model is able to construct a full vertical profile. In such a way it is possible to compute, by adequate interpolation, the electron density in any point of the ionosphere, and also slant or vertical TEC, at any given time.

An operational application of this reconstruction method was implemented using available experimentally derived vTEC maps, taking into account that various institutions around the World produce such maps using GPS data.

SOME RESULTS

For evaluation of the results of the reconstruction technique using data ingestion, the authors have used ionosonde data and slant TEC values.

SLANT TEC

COMPARISONS

Christiansted

(Virgin Islands)

GLOBAL MAP OF foF2 RECONSTRUCTION FROM THE INGESTION TECHNIQUE

DATA INGESTION IN NeQuick MODEL (2)

( C. Brunini, et al., COSPAR Scientific Assembly, Paris, July 2004)

An additional approach to the ingestion of GPS derived TEC data into NeQuick model has been introduced by the authors.

Data ingested are directly slant TEC values.

Instead of tuning the “ionization level index” Az , the parameters that are now optimized for any given location and time are the F2 anchor point ionospheric characteristics: electron density and height. ITU-R climatic values are used as initial guess.

Corrections of ITU-R F2 electron density and height are calculated and introduced in the NeQuick model.

SPHERICAL WAVELETS APPROACH

The authors present a spherical wavelets approach to model regionally the corrections to ITU-R values.

The spherical wavelet approach allows the handling of regional data and also a multi-resolution representation.

Regional corrections of F2 peak characteristics and their decomposition into different frequency bands or components are obtained.

A full theory of this ingestion technique and preliminary results has been given by the authors.

These preliminary results correspond to a European sub-region and are based on slant TEC from 5 stations.

MAP OF NmF2 ITU-R CORRECTIONS

ELECTRON DENSITY VERTICAL PROFILE

VERTICAL TEC

Comments

ASSIMILATION AND INGESTION IN MODELS

Both techniques depend on:

The correct estimates of the errors of the data assimilated or ingested (data quality control)

and

In the first case (assimilation in theoretical models): how well the physical description of the ionosphere adopted by the model represents reality.

In the second case (ingestion in empirical- semi-empirical or profilers models): How well the formal description of the ionosphere by the model represents reality.

ARE THESE TECHNIQUES ALREADY SUCCESSFUL?

Preliminary results of both assimilation and ingestion techniques have shown successful results to specify middle-latitudes ionosphere where spatial and temporal gradients of electron density are small.

However the results are less satisfactory when the techniques have been used in low-latitudes regions where the highly dynamic and complex Equatorial Anomaly is present.

A QUESTION

The question that can be posed at this time is:

Represents data assimilation and ingestion in ionospheric models the future of ionospheric research?

My answer is: YES!

ANOTHER QUESTION

IS IONOSPHERE SCIENCE AND AERONOMY AT THE END OF THE ROAD?

The present situation of our particular science that is dealing with the ionosphere - and of all of the field of Aeronomy – is such that many consider that there is nothing new to learn about it.

The reality is that, what we know about the upper atmosphere now, is not much more of what meteorologists knew about the weather some forty years ago.

We can say that we are able to see and understand about aeronomic weather only 10% of it. It is like an Iceberg: we are far from seeing and understanding the 90% of it, that is under the water.

The future

At present our science is based on experts that deal with: experiments, theories and models.

In the future a new category of expertise will have to develop: the science of data assimilation and ingestion in models .

Such models have to be constantly improved and tuned with experimental information.

THE NEW GENERATION OF AERONOMERS AND IONOSPHERISTS

The new generation of aeronomers or ionospherists will have to deal, in a way that was not clearly understood until now, with:

- Experimental data errors

- Physics of the Earth atmosphere as a whole and its uncertainties

- Mathematics of advanced least-squares fitting and similar techniques

- High performance computing techniques.

A multidisciplinary mentality and approach will be essential !!!

THE OUTCOME

Assimilation and ingestion schemes able to represent the weather condition of the upper atmosphere and the ionosphere in particular.

Maximum return from the huge data flow of TEC data derived from GPS and later GALILEO and other data sources like ionosonde.

MY ANSWER TO THE QUESTION

The future of our science leans upon the ability to assimilate and ingest data into theoretical or empirical and semi-empirical or profilers models.

This is the message I wish to leave to the younger ones

Thank you for your attention !!!