PROMISE

Predictability and variability of monsoons, and the agricultural and hydrological impacts of climate change

WP 3100 Report for April-September 2001 (months 13-18)

Contract number:                                    VK2-CT-1999-00022

Project reporting period:                       April to September 2001 (months 13-18)

Project co-ordinator:                              Florent Maraux, Cirad-Amis, Progr. Agronomie, France

Work-package number:                          3100

Objectives

Workpackage 3100 is entitled "Seasonal prediction of crop yields and assessment of climate change impacts on crop productivity", with deliverables 3101 (Identification of parameters for crop models and evaluation of seasonal prediction and crop models), 3102 (Integrated system for interpretation of seasonal forecasts in terms of crop development and yield) and 3103 (Assessment of impacts on crop productivity). The objective of WP 3100 is therefore to develop tools to derive agronomic scenarios (crop performance) from seasonal and long-term climate simulations, and to evaluate these scenarios. Target areas for these activities are the Sahel (main annual crops) and India (groundnut).

Progress made during the reporting period

Model development

The conceptual crop model described in the previous report was realised in DELPHI programming language. Two very similar versions were developed and parameterised, one for cereals (sorghum and millet) and one for peanut. The models appear to accurately simulate crop growth, ground cover and yield under water and radiation limited conditions. Further validation is under way, particularly with respect to population density and photoperiod effects, as needed to apply the models to PROMISE climate scenarios.

Acquisition of historical weather data

Partner 2 (Cirad) has formally sub-contracted Agrhymet (Niamey, Rep. Niger) for (1) the use of specific Sahel weather data (N-S and E-W gradients) for PROMISE crop simulations, and (2) the joint development of crop modelling softwares for yield forecasting in the context of PROMISE. The historical climate database will become available to PROMISE at the end of this reporting period (September 2001).

Acquisition of crop data for model validation

Extensive, experimental datasets for sorghum have been generated in collaboration with ICRISAT in Mali, and are now available. Experiments in Senegal on drought responses of sorghum in 2000/2001, however, failed due to technical problems. For groundnut, experimental datasets from Senegal for 1992 and 1995 have been obtained, but data generated in 2000 are not yet available. The data available to date have permitted model calibration and validation for a small set of environments.

Scaling exercises

A methodological sub-project was initiated in May 2001 to compare the temporal variability of simulated climate at the pixel scale (200 x 200 km2) with that of historical point data (weather stations falling into the pixel), in collaboration with Partner 3 (Meteo-France, Dr. JF Royer). Comparisons being  made on the basis of weather variables (e.g., precipitation and calculated potential evapo-transpiration) and on the basis of simulated crop yields derived from the data. Groundnut in the Sahel is being used as test crop, on the basis of the crop model described above. This sub-study will be completed by the end of 2001.

Spatial analyses

A tool for the spatial analysis and representation of PROMISE results has been developed and is available. It will be coupled with the new crop model once it is fully validated.

Problems encountered

None

Overview of outputs as of  September 2001 (month 18)

- Crop model for sorghum, millet and groundnut developed (but not yet fully validated) - Tool for spatial analysis and representation of  weather/crop data developed, but not yet coupled with crop model