linked in logo | IS-ENES | Contact

Exploring climate model data

Impacts of climate change on crop yields in the tropics

Goal:

Assess the impacts of climate change on crop yields in the tropics.

Overview

Agriculture is considered as the most weather-dependent of human activities. Although average levels of crop productivity across the world mostly depend on local non-climatic factors, the temporal variability in crop yields often reflects the variability in weather conditions. Fluctuations in climate can lead to severe socio-economic impacts in developing countries in the tropics for three main reasons:
(i) crop production is often the main source of food and income
(ii) levels of crop management technology are low and
(iii) in many cases these countries are exposed to high variability in climate.

We thus propose to translate climate outputs from GCMs into crop yields in the tropics. We will first validate the accuracy of the simulated yields by using yield database at various scales and then propose scenarios of crop yields evolution in the future.

Author:

Benjamin Sultan (LOCEAN, France)

Actors:

IPSL

Related use cases:

None

Data needs:

  • data type
    daily maximum and minimum temperatures, daily potential evapotranspiration, global radiation, daily rainfall
    period : 1968-2100,
    spatial covering: global data and regional data at higher horizontal resolution for Africa
  • data set
    Simulated data from the stream 2 data set of the European project ENSEMBLE. It includes 7 models and several simulations per models. RCM data from ENSEMBLES in Africa Observation gridded data for model correction. Three available datasets: the the reanalyses ERA40, ERAinterim, or NCEP. Rainfall datasets at stations (Hulme, IRD). Yield database from FAO at the national scale, district-scale and local scale yields database in West Africa

Typical course of events:

  • Correction of modelling data

    Selection of a correction method functions of the data biases, the geographical zones of interest and the data availability. Selection of a data set to perform the correction : observations, re-analyses Development or adaptation of correction code source Validation of corrected data

  • Yields simulation in the actual period

    Run the crop model with bias-corrected actual outputs from GCMs. Aggregate simulated yields at various scale corresponding to yields data base Compare simulated and observed yields

  • Yields simulation in the futur period

    Run the crop model with bias-corrected future outputs from GCMs. Aggregate simulated yields at various scale corresponding to yields data base Analyse the trends of future yields Identify vulnerable regions in the tropics

References:

P. Oettli, B. Sultan, C. Baron, M. Vrac (2011). ' Are regional climate models relevant for crop yield prediction in West Africa?'. Environ. Res. Lett., 6 , doi: 10.1088/1748-9326/6/1/014008

Methods:

  • Quantile-quantile correction method
  • Crop modelling
  • Method to compare several data-sets (trend, seasonal cycle, inter-annual variability)
  • Multi-model analysis
  • Kendall test for trend significativity evaluation

Software:

  • ORCHIDEE-MIL (crop modelling tool in the tropics)
  • IDL to change format data from ascii to netcdf.
  • Fortran program for the data correction and the indicator calculation
  • CDO, NCO, ferret, for data treatment and data analyses (DTR, indicator..)
  • NCL, EXCEL for graphical display

File format:

NetCDF Ascii

 

 

The IS-ENES project has received funding from the European Union's Seventh Framework Programme for research, technological development and demonstration.

Disclaimer