Agrometshell

1
Agrometshell

• Workshop 15-17 September Rome

2
Peter Hoefsloot?

• Dutch National
• Married, 2 children, 7 sheep
• Msc. In Agronomy/Meteorology/Comp. Science in
  Wageningen, The Netherlands
• Do management of geo-info projects for Dutch
  consultancy firm (Haskoning)
• Have my own (little) company working for FAO and
  Dutch ministries writing software
• 1989 1991 Agrhymet, Niger
• 1991 1994 RRSU, Harare, Zimbabwe
• Many missions (most SADC and CILSS countries,
  Djibouti)

3
Parts of this demo

• A bit of history
• Objectives of AMS
• AMS seen from different perspectives
• General structure and functions
• Demonstration

4
A bit of history

• Drought sub-Sahara Africa (mid-seventies)
  desertification
• 1974 - CILSS founded Agrhymet in Niamey (West
  Africa)
• 1986 - Intergovernmental Authority on Development
  (IGAD) formed Intergovernmental Authority on
  Drought and Development (IGADD) for East Africa
  in Djibouti
• Mid-eighties SADC (Southern Africa) founded the
  Regional Early Warning Unit in Harare
• Now - New frontiers Afghanistan, IRAQ,
  Bangladesh

5
Assisting early warning

• FAO (United Nations)
• GIEWS (Global Information and Early Warning
  System) and support for national and regional EW
  units
• ARTEMIS (Africa Real Time Environmental
  Monitoring Information System)
• USAID (United States)
• FEWS - Famine Early Warning System
• European Union
• Many institutes
• University of Venice , Italy
• University of Reading, UK (TAMSAT)
• ITC, The Netherlands
• USGS, United States
• ()

6
What do you need for EW?

• Data and information
• Methods and models
• Software
• Facilities (computers, communication)
• Skilled staff

7
Data for early warning

• Stocks on staple foods
• Pricing of foods on markets
• Crop Yields
• Weather data (e.g. rainfall) through GTS
  (worldwide) and Met Services (national)
• Satellite data (mainly METEOSAT and NOAA)
• Static support data maps, census data,
  agro-ecological zones, soil maps

8
Where do we get data from?

• Ministries and other government institutions
• Met Services
• Many internet sources
• ARTEMIS and AGROMET data information
• http//metart.fao.org/
• Africa Data Dissemination Service
• http//edcw2ks21.cr.usgs.gov/adds/

9
Methods and models

• NDVI (Vegetation Greenness Monitoring) 1x1 km
  and 7x7 km
• Cold Cloud Duration (CCD)
• Rainfall estimates (RFE)
• Water Balance Models
• Statistics
• Interpolation

10
Software

• IDA- Windisp
• Agman Priceman Spaceman (USGS)
• FAOINDEX, FAOMET and others
• ADDAPIX
• spatial and temporal analysis of satellite
  imagery
• MADAM
• generation of multi-image statistics
• IGT
• GIS and interpolation tools for IDA
• FAOCLIM
• software and large agro-climatic database

11
Why write EW software?

• License free
• Moderate computing requirements
• Ease of use (WB in Excel is possible..)
• Possibility to create new analysis methods (SEDI,
  ADDAPIX)
• GIS systems require a lot of training, use large
  and complex data models, are not license free,
  use heavy computers

12
AgrometShell
13
AMS History

• 1989 Niamey Niger SUIVI
• 1992 Harare Zimbabwe SEDI and IGT
• 1995 2000 SEDI updates
• From 2001 AMS
• Promotors and sponsors FAO Rome, Aghrymet, REWU
  Harare, IGADD

14
AgrometShell objectives

• Facilitate monitoring of growing season
• For national and regional EW units and
  international bodies like FAO
• Available license free
• Easy to use and well-documented
• Bridging the gap between agromet, remote sensing
  and socio-economic datasets
• Flexible toolbox to which others can contribute
  with code (e.g. Univ. of Louvain interpol.)
• Exchange with other relevant software
• Windows rewrite of DOS software
• AMS will not provide functions other packages
  offer

15
AgrometShell in a nutshell

• FAO Crop Water Balance model
• Database for Agromet point data (SUIVI)
• Interpolation (SEDI, Inverse distance, Co-Kriging
  etc)
• Statistics useful for Agromet
• Provide conversion functions between data files
• Viewer (every function ends with viewing results)
  
• Automation
• Some functions are in because unavailable in
  other EW software

16
AMS technically

• Programmed in Delphi (Pascal)
• Contributions by others in form of DLLs
• Access database (through ADO)
• Executable that does not require any other
  software
• Share database on network

17
Main Window
18
DEMO 1

• The agromet database

19
SUIVI Database for agromet data

• Daily, Dekadal and Monthly weather station
  (point) data
• Every operation through flexible stationlist
• Add parameters easily
• Flexible ASCII import

20
Database technically

• Database in Microsoft ACCESS 2000 format
• Accessible from outside AMS
• Very common database format
• Query generator in Access
• Database can be placed on network
• Exchange of data with large database (Oracle SQL
  Server etc.)
• Early versions had Paradox tables

21
Demo

• Inventory
• Lists and base list
• Parameters
• Data entry
• View data on map, graph, report
• Formulas and calculation
• Aggregation
• Import from image and ASCII file

22
DEMO 2

• Calculating a Water Balance

23
Water Balance

• Model based on the work of Frere/Popov and Rene
  Gommes
• Improvements so far
• Irrigation (amount at planting or dekad by dekad)
• Phenological stages initial, vegetative,
  flowering, ripening
• Crop coefficients based on 9 rather than 4 graph
  points Daily Time Steps
• More sets of crop coefficients per crop
• Run file approach

24
Water Balance (2)

• AMS does not operate directly on database, but
  on ASCII files.
• ASCII files are first exported from the database
• Two possibilities
• 1. Monitoring (1 year many stations)
• 2. Risk Analysis (1 stations many years)

25
DEMO

• Close look at crops
• Dekadal and daily time steps
• With and without irrigation
• View results
• Make images from results
• Automation

26
DEMO 3

• Integrating and analyzing data

27
Data integration (1)

• Technically data come as
• Points
• Areas
• Images (or grids)
• Images are best for analysis
• Very visual (easy to check results)
• A picture tells more than a 1000 words
• Easy arithmetic with pixels

28
Demo

• Integrate Water Balance results with Yield data
• WB results point data
• Yields From ministry (area aggregated)
• Turn both into images
• Study the relation between Yield and Water
  Satisfaction Index geographically