.png){kind=icon}
.png){kind=icon}
Return
Predicting rainfed and irrigated yield responses under drought with soil-hydrological modelling and variety trials for economic evaluation
.png){kind=icon}
OA
AI
B
S
A
W
P
B
S
DOI：10.1186/s12302-026-01480-y.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Agricultural droughts in many humid regions such as Southern Germany are projected to become more frequent and severe, leading to a substantial rise in irrigation demand to secure crop production. As a result, irrigation is becoming an increasingly important factor in agricultural production, and operational and regulatory decision-makers are relying more on data on site-specific yield responses to severe drought, especially as spatial soil heterogeneity leads to pronounced yield variability. Statistical crop yield models are commonly used to predict yield responses to drought; however, their calibration is often based on aggregated yield data, which can bias predictions of site-specific yield responses. Reported aggregated yield data for drought years are difficult to interpret because it is unclear whether crops were irrigated and to what extent. The aim of this study is to introduce a methodological approach to forecast agro-economic impacts of increasing drought frequency, using site-specific soil properties, meteorological data and yield observations from past drought years to assess potential impacts of future consecutive drought events. High drought frequencies are represented by repeatedly applying the severe drought conditions of 2003 over time. The process-based hydrological model RoGeR was coupled with a linear mixed-effects model, combining simulated soil moisture data with spatially corresponding observed yield data from post-registration variety trials (Landessortenversuche) in the German state of Baden-Württemberg. Yield data from the variety trials are characterized by standardized management practices, particularly with respect to irrigation. We demonstrate the proposed approach in a case study in Southern Germany by forecasting yields under rainfed and irrigation scenarios as well as irrigation volumes for major arable crops and representative soils for the period 2000–2024. The results showed that both yield losses and irrigation-induced yield gains during drought vary significantly between crops and across sites and are associated with substantial differences in irrigation volumes and the number of irrigation events, which in turn affect the profitability of irrigation investments. The findings confirm that site-specific modelling of yield effects is essential for evaluating the profitability of irrigation investments. Our approach provides the necessary data basis and can be applied across spatial scales.
Keywords:
Agricultural drought
Irrigated agriculture
Crop yield model
Linear mixed effects model
Hydrological model
Soil moisture
Groundwater
Payback periods
Irrigation viability
Journal
IF:
6
Papers:
1.4K
Citations:
5.6K
Organization
Cited Papers