P2 - Use of plant growth models and seasonal weather forecast for yield stabilization under vulnerable crop cultivation conditions

Status: Ongoing

Duration: 2021 - 2025

Keywords: Biogeophysics, Crop Modelling, Uncertainty Analysis

Description

Climate change is expected to have an adverse impact on cereal crop production in Ethiopia (Abera et al., 2018; Mequanint et al., under review) and a major food security risk is projected by 2050 for this region (van Ittersum et al., 2016). In particular, increasing frequencies and magnitude of climate extremes, as already documented for Ethiopia since the 1980s (Esayaset al., 2018; Gebrechorkos et al., 2019), pose an existential risk to smallholder farmers, which rely on rain-fed agriculture. Consequently, identifying appropriate short- and medium-term adaptation measures is one of the major challenges for agriculture in this region (Tesfaye et al., 2015).

Process-based crop models have proven to be useful tools to predict crop responses to different climate scenarios (Asseng et al., 2015) and to evaluate agricultural adaptation options with respect to sowing dates, irrigation (Ruiz-Ramos et al., 2018) and fertilization dates/amounts. It has been found that crop model ensemble predictions are superior to single model studies (Martre et al., 2015, Mequanint et al. (under review); Wallach et al., 2019). A prerequisite for precise crop model predictions, however, are reliable input data, in particular realistic weather scenarios. To enable farmers to react to weather fluctuations, the coming weather must be known several weeks in advance. Feeding crop models with the predicted weather scenario would then allow to find the best management strategy to optimize yields under the expected circumstances. In particular, short-term weather projections could act as an ‘early warning‘-system for weather extremes.

The impressive progress that was made in the field of high-resolution weather predictions (downscaled from lower resolution global weather models) during the recent years gives a promising prospective to achieve this vision in near future. The envisioned combination of plant growth models with local seasonal weather forecast could help decision makers respond to impending yield failures with regionally specific adaptation recommendations.

The aim of the project is to find out what value seasonal weather forecasts may have for stabilizing yields in vulnerable regions such as Sub-Saharan Africa. The agroecosystem modelling platform XN (Expert-N 5.1) will be used to find optimized field management strategies with respect to weather projections until the end of the growing season for main cereal crops and the four main climate regions of Ethiopia.

In a first step, XN has to be extended for teff, based on the work by Paff and Asseng (2018). Multi-crop-model ensembles will be set up for wheat, maize, barley, and teff. For all simulated crop-species and varieties (e.g., high-land/low-land varieties), suitable model parametrizations with respect to the different climate zones in Ethiopia have to be established, based on phase I results (wheat, maize), field data (to be collected from research stations) and climate chamber experiments from (project P2, barley, wheat, phase I+II).

Simulations assuming ‘business as usual’ farming methods are compared with simulations, in which crop cultivation was adapted to the projected weather, to answer the question to what extent crop cultivation could have been optimized, if the season’s weather conditions had been known. The studies will be carried out for current climatic conditions and future climate projections (2050s) to take into account possible changes in weather extremes in the major growing regions of Ethiopia.

Involved persons

Prof. Dr. Thilo Streck, Dr. Sebastian Gayler
Kidist Abera Anteneh

Involved institutions

Institute of Soil Science and Land Evaluation, Biogeophysics (310d), University of Hohenheim

Involved persons

Involved institutions

Sponsors