We invite contributions that address:
• Advances in regional crop modeling frameworks (process-based, statistical, and hybrid)
• Integration of AI/ML techniques within a remote sensing and crop modeling framework
• Data assimilation techniques and model parameterization strategies
• Integration of remote sensing data into regional crop modeling systems
• Seasonal yield forecasting and the role of improved initial conditions via data assimilation
• Uncertainty quantification of regional crop model output
• Applications to water use, irrigation, and agro-hydrological monitoring
• Applications to assess and optimize climate change adaptation strategies
• Benchmarking and intercomparison of crop models with remote sensing data
This session brings together researchers in crop modeling, remote sensing, climate science, and data assimilation to advance integration across disciplines and tackle global challenges in agriculture and food security.
Over the past fifty years, tomato has become one of the most extensively cultivated horticultural crops in the Mediterranean region. Climate projections for Italy indicate that temperature increases and rainfall changes will cause a 15% yield reduction in processing tomatoes, requiring an additional 85-110 mm of irrigation and 20-30 kg N ha-1 to partially offset negative impacts. Mediterranean agriculture is particularly threatened by projected climate changes in temperature and precipitation patterns. Region-specific crop models, validated against local field data, are therefore critical tools for assessing yield risks and identifying effective agronomic adaptations. Conventional process-based crop models often rely on fixed transplanting or sowing dates and harvesting dates, which fail to reflect spatiotemporal variability in management practices. Such assumptions can lead to systematic biases in regional simulations and environmental assessments. Yet phenological observations (e.g., flowering, fruit set, harvest dates) are essential for parameterizing crop models, available data typically represent point locations or experimental stations rather than the field to regional scale resolution needed for spatially explicit modelling. Sentinel-2’s high temporal frequency and spatial resolution allow tracking of within-season crop development at field scale. This study aims to: (a) compare model performance using broad agricultural land masks versus pixel-level tomato identification; and (b) evaluate whether incorporating satellite-observed canopy development dynamics (greenness trajectories, growth stage timing) reduces uncertainty in simulated crop growth, water use, and nitrogen cycling processes including nitrate leaching risk.
We propose a simulation framework that combines the process-based model MONICA (Model for Nitrogen and Carbon dynamics in Agro-ecosystems) with earth observation data for processing tomatoes in the Emilia Romagna region, a major tomato production area in Italy. MONICA was calibrated and validated using four years field trials and two years on-farm data from 49 fields. We integrated two remote sensing inputs: (i) field scale processed tomato masks, and (ii) dynamic transplant and harvest dates extracted from Sentinel-2 EVI time series (validated against on-farm data, R²=0.90). We conducted regional simulations (2007-2023) comparing four model set-ups: fixed transplant and harvest dates with basic cropland mask, fixed dates with tomato masks, dynamic dates with tomato masks, and modified dynamic dates with tomato masks for sensitivity tests on transplanting date.
Our research results indicate that employing specific tomato field maps combined with dynamically determined growing periods significantly improved yield simulation accuracy compared to basic cropland mask (reducing RMSE by 24%) and specific maps without consideration of remotely sensed growing season dynamics (reducing RMSE by 10%). Incorporating remote sensing data and tomato maps into the MONICA crop model also improved the model’s ability to capture yield anomalies as an indicator of its sensitivity to climatic signals, with a 24% reduction in RMSE. Integrating remote sensing-derived growing periods into crop models resulted in a wider range of simulated values, enhancing the model’s capacity to simulate nitrate leaching under real-world conditions.
This study demonstrates that using remote sensing data to inform crop models significantly enhances the understanding of dynamic growth patterns, thereby supporting regional yield estimation and nitrate leaching simulations, while providing crucial insights for agricultural resource management.
How to cite: Yang, X., Rezaei, E. E., Farneselli, M., Croci, M., Tei, F., and Nendel, C.: Improving regional simulations of processing tomato using remote sensing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3532, https://doi.org/10.5194/egusphere-egu26-3532, 2026.
Process-based crop models embedded within land surface schemes provide a physically consistent framework for assessing crop–climate interactions. However, their application in semi-arid regions is often constrained by limited field data and simplified management assumptions. In particular, fixed planting dates remain a major source of uncertainty for rainfed systems where sowing decisions are strongly controlled by rainfall timing and intra-seasonal variability.
We develop and evaluate a framework for simulating rainfed sorghum growth using the Noah-MP-Crop model, with dynamic planting dates derived from satellite observations. Sowing timing is inferred from temporal trajectories of the GLASS Leaf Area Index (LAI) product, enabling spatially and interannually varying planting information to be incorporated. The approach is applied over the semi-arid eastern Nile basin, where sorghum production is highly sensitive to seasonal rainfall variability.
The model is implemented within the WRF-Hydro modeling system and driven by ERA5-Land atmospheric forcing and IMERG satellite-based precipitation. A stepwise calibration strategy is adopted, targeting crop phenology, leaf area development, and carbon allocation processes. Model performance is evaluated against satellite-derived LAI, independent energy flux estimates, and observed yield data, with comparisons between simulations using fixed and dynamic planting assumptions.
Results show that dynamic planting dates substantially improve the timing and magnitude of simulated LAI, particularly during early growth stages. In contrast, energy fluxes exhibit weaker sensitivity to planting date representation, reflecting the dominant control of atmospheric demand and radiation on surface energy partitioning in semi-arid conditions. Furthermore, simulations using dynamic planting dates show improved agreement with observed yields, indicating that a realistic representation of sowing variability translates into better seasonal productivity estimates. The findings highlight the importance of representing realistic sowing variability for crop growth simulation, while also illustrating the potential of combining open satellite products with process-based models in data-limited regions.
This work demonstrates a practical methodology for integrating dynamic planting information into land surface crop models, providing a transferable approach for improved crop–climate assessments and future seasonal yield prediction applications.
How to cite: Hageltom, Y., Arnault, J., Elagib, N., Laux, P., and Kunstmann, H.: Integrating dynamic planting dates into Noah-MP-Crop for sorghum simulation in semi-arid regions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14055, https://doi.org/10.5194/egusphere-egu26-14055, 2026.
Understanding when major crops face water deficits and the magnitude of the resulting yield impact is becoming increasingly important; however, large-scale, crop-specific evaluations of when drought stress occurs and how severe it becomes remain limited, particularly those that connect stress timing and severity to the physiological processes determining yield. Our study addresses this gap by using a process-based model integrated with remote sensing data to derive a physiologically grounded drought indicator from point scale to grids and ultimately to district-level resolution. More specifically, we used gridded historical and projected climate data, along with crop, soil, and terrain information. Our first step was to examine how stress timing and severity have historically influenced silage maize and winter wheat yields across Germany. The analysis revealed that drought during shooting-tasselling and tasselling to flowering for silage maize, and grain filling for winter wheat had the strongest association with major yield losses. These crop-specific windows highlighted the importance of stage-dependent stress assessment. The next step involved benchmarking of our physiologically based drought indicator against Sentinel-3 based drought hazard products to compare the simulated and remotely sensed drought-affected areas. Finally, we conducted scenario-based exploration of climate and irrigation conditions to assess how different management and environmental scenarios alters future drought exposure and yield outcomes. In this process, we incorporated Sentinel-2 derived irrigation maps to spatially distinguish irrigated from rainfed areas, improving the representation of actual water management practices. By combining process-based crop models with Earth observation data, our framework provides a foundation for digital twin applications in agriculture showcasing a virtual replication of crop-climate interactions that enables systematic evaluation of how future stress patterns, management decisions and policy interventions may shape agricultural productivity at a larger scale.
How to cite: Bondad, J. G., Ghazaryan, G., Schwarz, M., Augscheller, I., Escueta, R., and Nendel, C.: Upscaling drought stress detection through integrated crop model and remote sensing data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6511, https://doi.org/10.5194/egusphere-egu26-6511, 2026.
Reliable regional crop yield forecasts are increasingly challenged by climate variability, extreme weather events, and growing pressure on land and water resources. Process-based crop and agro-ecosystem models provide a physically consistent framework to assess these impacts, yet their predictive skills at regional scales remain limited by uncertainties in initial conditions, parameterization, and the representation of in-season stress dynamics. At the same time, Earth observation (EO) data provide spatially explicit information on crop phenology and vegetation status that can help constrain and update model simulations.
This study investigates hybrid modeling and data assimilation strategies to improve seasonal yield predictions by integrating satellite-derived vegetation indicators (e.g., fraction of absorbed photosynthetically active radiation (FAPAR)) with the process-based agro-ecosystem model LPJmL. The focus is on regional-scale applications, using Bavaria, Germany as a case study representative for a heterogeneous and hydrological complex landscape, and on assessing how EO-informed initial states and in-season updates influence yield predictions throughout the growing season.
Time series of FAPAR observations are used to characterize crop phenology and canopy dynamics during the growing season and are integrated with LPJmL simulations through different coupling strategies. As LPJmL does not natively support continuous EO data assimilation, several integration pathways are explored, including parameter forcing, ensemble-based approaches, and hybrid extensions that combine process-based modeling with machine learning components trained on model outputs, EO, and meteorological inputs. These hybrid elements are designed to leverage EO and meteorological information to account for non-linear effects and growth-stage-dependent responses that are difficult to capture in purely process-based algorithms.
Meteorological forcing is derived from ERA5-Land reanalysis and C3S seasonal forecast data, with sensitivity experiments exploring the role of seasonal forecast information. Particular emphasis is placed on the role of climate extremes during critical phenological phases and their implications for seasonal yield variability. Model calibration and evaluation are conducted using historical yield statistics and regionally consistent land-use information, allowing an assessment of uncertainty related to parameter choices, assimilation strategies, and hybrid model components.
The presented framework contributes to ongoing efforts to link regional crop models with EO vegetation dynamics data through scalable and transferable methods. By combining process understanding with data-driven constraints, this work aims to improve the robustness of seasonal yield forecasting and to support future applications in agricultural and food security monitoring, climate impact assessment, and adaptation planning.
How to cite: Jörges, C. and Hank, T.: Regional Seasonal Crop Yield Forecasts Through Hybrid Crop Modeling and Remote Sensing Data Assimilation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4037, https://doi.org/10.5194/egusphere-egu26-4037, 2026.
Root-zone soil moisture (RZSM) critically controls crop development, yet satellite missions observe only near-surface soil moisture, which poses challenges for its incorporation into crop models. In AquaCrop, the soil water module exhibits limited vertical coupling between computational soil compartments due to the abrupt effects of wilting point and field capacity thresholds, restricting the downward propagation of surface information. To address this limitation, an exponential filtering approach can be used to transform surface soil moisture into temporally smoothed estimates that are more representative of deeper soil layers. We assess the assimilation of SMAP Level-2 surface soil moisture into AquaCrop over European croplands (2015–2023, 0.1° resolution) with the aim of improving RZSM under these structural constraints. We compare direct assimilation of SMAP retrievals with assimilation of exponentially filtered datasets representing effective target depths of 30, 60, and 100 cm, using seasonally varying CDF matching within an ensemble Kalman filter.
The assimilation consistently improves topsoil (0–30 cm) moisture, but gains in subsoil (30–100 cm) moisture are strongly affected by the weak internal vertical coupling of the soil water balance. Specifically, while the direct assimilation of surface observations has limited impact below 30 cm, that of filtered products leads to improvements in RZSM. The best performance is obtained for a 60 cm target depth, with widespread increases in correlation against in situ observations. The impact of improved soil moisture is also evaluated for canopy cover and biomass using satellite-based reference data. Vegetation improvements remain weak and inconsistent, influenced by several factors including biases in the reference data and limitations in soil–plant coupling, for example, due to the use of a generic crop parameterization that is not spatially explicitly calibrated. Our results highlight the value of exponential filtering for soil moisture assimilation in weakly coupled crop models and point to joint soil moisture–vegetation assimilation as a promising pathway for further improvements.
How to cite: Bechtold, M., Busschaert, L., Heyvaert, Z., Kumar, S., Raes, D., Massari, C., and De Lannoy, G.: Surface Soil Moisture Data Assimilation in AquaCrop: Overcoming Limited Vertical Coupling with an Exponential Filter, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12692, https://doi.org/10.5194/egusphere-egu26-12692, 2026.
In agrarian economies like India, anticipating crop yield shocks before harvest is crucial for managing climate risks, stabilizing markets, and safeguarding food security. As extreme weather events become more frequent, policymakers need not only early warnings but also interpretable insights that explain where and why failures may occur. Yet, a key limitation remains: while remote sensing provides fine-scale information, yield data are usually available only at coarse administrative levels, and common averaging approaches erase the local variability that often drives yield losses. To bridge this gap, we introduce INDRA-Net (Interpretable Network for District Residual Aggregation), a weakly supervised Multiple Instance Learning (MIL) framework that directly predicts 38 years (1980-2017) of district-level yield residuals from high-resolution pixel-level time series. Unlike conventional methods that rely on naive spatial aggregation, the architecture employs a shared Temporal Fusion Transformer (TFT) backbone to independently encode the complex interactions between static drivers (e.g., soil properties, topography) and dynamic inputs (e.g., weather, vegetation indices) at the individual grid-cell level. These local embeddings are then synthesized via a learnable Gated Attention mechanism, which dynamically assigns higher weights to agriculturally relevant pixels while suppressing noise and non-crop signals. The framework is trained with a quantile regression objective to forecast yield anomalies, enabling explicit uncertainty estimates (P10–P90) essential for operational risk management. Extensive evaluation on wheat and maize yields across Uttar Pradesh, Punjab, Madhya Pradesh, and Bihar demonstrates that INDRA-Net reduces forecasting error (RMSE) by 12–14% compared to state-of-the-art machine learning baselines (Random Forest, XGBoost) and deep learning models (LSTM). By preserving pixel-level variability, the model captures localized extreme events—such as heatwaves or moisture stress- that are typically smoothed out by spatial aggregation. Crucially, the model’s three-dimensional interpretability aligns with crop physiology, correctly identifying maximum temperature during wheat grain-filling and precipitation anomalies during maize silking as the dominant temporal drivers, while isolating sub-district clusters responsible for yield failures. This enables the generation of granular yield anomaly maps without pixel-level labels, offering policymakers a scalable and operational tool for precision monitoring and targeted risk intervention.
How to cite: Srivastava, A. K., Halder, K., Muduchuru, K. R., Pires Barbosa, L. A., Rahaman, K., Lanka, K., Alsafadi, K., Maerker, M., Gaiser, T., Behrend, D., Zhao, G., Zheng, W., Han, L., Singh, M., and Ewert, F.: INDRA-Net: A Weakly Supervised Multiple Instance Learning Framework for Spatio-Temporally Interpretable and Extreme-Aware Crop Yield Forecasting in India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19039, https://doi.org/10.5194/egusphere-egu26-19039, 2026.
We introduce NeuralCrop, a differentiable hybrid global gridded crop model (GGCM) that combines the strengths of an advanced process-based GGCM, resolving important processes explicitly, with data-driven machine learning components. The model is first trained to emulate a competitive GGCM before it is fine-tuned on observational data. We show that NeuralCrop outperforms state-of-the-art GGCMs across site-level and large-scale cropping regions. Across moisture conditions, NeuralCrop reproduces the interannual yield anomalies in European wheat regions and the US Corn Belt more accurately during the period from 2000 to 2019 with particularly strong improvements under drought extremes. When generalizing to conditions unseen during training, NeuralCrop continues to make robust projections, while pure machine learning models exhibit substantial performance degradation. Thanks to optimization to graphical processing units (GPUs), NeuralCrop is more than 80 times faster on a single GPU than a state-of-the-art competitor on 128 CPU cores. Our results show that our hybrid crop modelling approach offers overall improved crop simulations and more reliable yield projections under climate change and intensifying extreme weather conditions.
How to cite: Lin, Y., Bathiany, S., Badri, M., Gelbrecht, M., Hess, P., Groenke, B., Heinke, J., Müller, C., and Boers, N.: NeuralCrop: Combining physics and machine learning for improved crop yield predictions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6744, https://doi.org/10.5194/egusphere-egu26-6744, 2026.
Accurate retrieval of wheat phenological stages is fundamental for crop monitoring, yield forecasting, and understanding climate-crop interactions, particularly in heterogeneous landscapes such as the Indo-Gangetic Plains. Conventional field-based observations, although reliable, are labour-intensive, spatially limited, and often unsuitable for regional-scale assessments. Satellite remote sensing offers a valuable alternative, yet current phenology monitoring is constrained by observational gaps driven by cloud interference, uneven temporal sampling, and signal noise in vegetation indices. These limitations create uncertainty in identifying critical phenological stages, such as emergence, jointing, heading, and maturity, during the entire winter wheat growing season. To address these challenges, this study presents a refined and transferable phenology extraction approach that integrates multisatellite observations from Sentinel-2A and Landsat-8 using a data assimilation-based fusion technique. Daily, gap-free wheat NDVI trajectories at high (10m) spatial resolution were generated by combining the strengths of both sensors through pixel-level data assimilation and Savitzky–Golay (SG) filtering. A double logistic curve-based phenology detection algorithm was then applied to extract key inflection points from the wheat NDVI seasonal profile. This allowed the retrieval of five major phenological stages: Start of Season, Active Greenup, End Greenup, Peak, and Senescence. The satellite-derived stages were compared with field-observed growth stages at the Department of Water Resources Development and Management, Indian Institute of Technology Roorkee experimental farm. These five satellite-derived phenological stages corresponded closely to emergence, crown root initiation, jointing, heading, and maturity, respectively. Validation showed strong performance, with a mean absolute error of 7 days and a Kling-Gupta efficiency of 0.92. Spatial patterns highlighted pronounced early and mid-season variability across the study region. The Siwalik–Bhabar uplands exhibited delayed emergence and slower Greenup due to shallow, gravel-rich soils and restricted moisture availability, while lowland floodplains demonstrated earlier and more uniform phenological progression. Despite variability in early stages, final maturity dates converged across districts, reflecting regionally synchronized harvest timing. This approach enhances large-scale phenological assessment for supporting better management decisions in data-scarce agroecosystems.
Keywords- Data assimilation, double logistic, wheat, phenology
How to cite: Singh, P. and Kothari, K.: Refined Retrieval of Winter Wheat Phenological Stages in the Indo-Gangetic Plains Using Fused Sentinel-2A and Landsat-8 NDVI Time Series Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-711, https://doi.org/10.5194/egusphere-egu26-711, 2026.
Precise monitoring of crop canopy cover (CC) is crucial for evaluating growth and health under diverse water and nutrient conditions. Although nano liquid urea has been promoted in India as an eco-friendly alternative to conventional nitrogen (N) fertilizers. Its effectiveness in potato cultivation, particularly for canopy development and yield, remains unclear. To address this gap, a field experiment was conducted during the 2024–25 winter season using three N treatments under a micro-irrigation system: T1-recommended granular urea (46% N), T2-without N, and T3-IFFCO nano liquid urea (4% w/v). Images were captured using a downward-looking smartphone camera positioned 2.5 meters above the crop near each treatment, serving as the primary input for estimating canopy cover. The images were processed using both a prompt-based segmentation model (SAM3) and an image-processing pipeline (Modified Excess Green Index + Otsu thresholding) to estimate CC. The SAM3 occasionally overestimated or failed to detect the potato canopy with the prompt “Green plants, leaves, vegetations, canopy cover”, whereas the image-processing approach consistently provided accurate CC estimates and was therefore used for subsequent analysis. The result revealed that the CC clearly showed differentiation among the treatments after 25 days after sowing (DAS). With this most of the treatment comparisons depict the CC peaking after 45 DAS, where the T1 recorded the highest canopy cover (~71%), indicating a healthy crop, while the pairwise comparisons showed CC values of ~55% (T1–T2), ~66% (T1–T3), ~33% (T2–T2), ~36% (T3–T3), and ~36% (T2–T3), depicting the nitrogen deficit. Similarly, the yield followed the same trend, with T1 producing the highest yield (26.78 t/ha), compared to 10.89 t/ha in T2 and 11.90 t/ha in T3. The results indicate that nano liquid urea does not supply sufficient nitrogen to support optimal potato canopy growth and productivity, resulting almost similar response to the no nitrogen application treatment. Increasing the nitrogen concentration in nano liquid urea formulations may improve their effectiveness. This study provides evidence to guide farmers in selecting appropriate nitrogen fertilizers for potato cultivation. In the future, such fertilizers should be evaluated across different crops to ensure their efficacy and to prevent farmers from adopting products that may not meet crop nutrient requirements.
How to cite: Maurya, A. K. and Pathak, A.: Evaluating SAM3 and Conventional Image Processing Method for Potato Canopy Cover Estimation as an Indicator of Crop Health, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1681, https://doi.org/10.5194/egusphere-egu26-1681, 2026.
Crop residue burning in smallholder farming systems represents a critical source of atmospheric pollution and greenhouse gas emissions. However, current operational active fire monitoring products from coarse-resolution MODIS/VIIRS, restrict their application to mapping and monitoring crop stubble burning in smallholder farms. These smallholder farming systems have field sizes that may vary between 0.5 and 2 hectares, resulting in an underestimation across ~40% of the global croplands. This current limitation necessitates the need for high-resolution alternatives that can help track and monitor crop burn practices. This enables the accurate quantification of GHG emissions and the implementation of regulations in densely populated areas. To address this limitation, we developed a machine-learning approach for high-resolution mapping and monitoring of stubble burning using PlanetScope (3-5 m resolution) and Sentinel-2. Our results demonstrate that the burn detection model applied to PlanetScope achieved an accuracy of 81%, outperforming the Sentinel-2-based detection model, which had an accuracy of 69%. We attribute this to the finer resolution of Planetscope, which even compensated for the spectral limitation in detecting the burn events. The predicted PlanetScope burn detection product further enabled the delineation of burn patterns within individual farm boundaries, allowing us to classify whether a farm is entirely burned or partially burned based on the percentage of burnt area per field. Random Forest feature importance indicated that Global Environmental Monitoring Index (GEMI) consistently outperformed as the optimal spectral predictor, compared to the traditional indices, including the Normalized Burn Ratio and the Normalized Difference Vegetation Index. We also found that GEMI can effectively discriminate between burnt signatures and spectrally similar agricultural activities, such as post-harvest tillage and crop residue management operations. Our results demonstrate that high-resolution commercial imagery can significantly enhance operational agricultural monitoring. Moreover inspiring confidence in policymakers and researchers by enabling the accurate quantification of emissions, effective policy enforcement, and environmental health protection across smallholder regions globally. However, a significant challenge persists in the scalability of research-grade studies to operations due to extremely higher costs associated with PlanetScope's commercial data acquisition (exceeding $200,000 annually for district-level continuous monitoring). These costs present significant barriers for resource-constrained governmental agencies and research institutions in developing countries, despite their demonstrable technical superiority. Future studies should address these challenges by developing data fusion-based hybrid frameworks that offer a scalable solution, striking a balance between technical needs and fiscal realities, while supporting climate mitigation and sustainable agricultural practices that strategically leverage complementary sensor capabilities.
Keywords: remote sensing, machine learning, PlanetScope, Sentinel-2, crop residue burning, burnt area detection
How to cite: Konkathi, P., Torbick, N., Ajmera, I., Reba, M., and Hall, J. V.: Monitoring Crop Residue Burning in Smallholder Farms at Sub-Field Scale Using High-resolution Satellite Imagery, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15156, https://doi.org/10.5194/egusphere-egu26-15156, 2026.
PhenoCams have become a common tool for vegetation phenology monitoring in North America and Europe, but their use in Indian agriculture remains very limited. Most crop phenology research in India still depends on satellite imagery or field surveys. These approaches provide valuable information, but they often lack high temporal resolution and cannot capture short-term changes in crop conditions. To address this gap, we evaluated the use of a PhenoCam to monitor wheat and rice phenology at a field site in Roorkee, India across two growing seasons. Our aim was to assess how high-frequency imagery can complement field observations, satellite, and meteorological data for crop phenology assessment.
We installed an infrared-enabled PhenoCam on a 6.5 m tower overlooking a winter wheat field during the rabi season (2023-24 and 2024-25) and a rice field during the kharif season (2024). Images were captured automatically at fixed intervals and were processed using the PhenoAI framework, which is a deep learning Python framework designed for automated time-series data processing. Greenness indices such as GCC and NDVI were derived from the processed images. For the wheat study, we conducted a cross-platform evaluation over two consecutive seasons (2023–2025) by combining PhenoCam data with in-situ observations, Sentinel-2, and PlanetScope imagery. PhenoCam achieved the highest timing agreement with field observations, with a mean absolute error (MAE) of 2.6–3.5 days. Sentinel-2 followed with MAE values of 2.4–4.2 days, while PlanetScope showed larger errors of 4.1–5.6 days due to radiometric noise and cloud cover. GCC was most sensitive to early green-up, whereas NDVI provided stable tracking of the full growth cycle (R² > 0.90).
During the rice season, we focused on how crop phenology responds to local weather conditions. We collected meteorological data from a co-located automated weather station. We examined climate–phenology relationships using a combination of exploratory correlations and mixed-effects model analysis. Minimum air temperature and PAR showed the strongest overall negative correlations with canopy greenness (r = −0.42 and r = −0.37). Stage-wise analysis indicated that tillering responded positively to temperature (r = 0.45), while booting and heading showed negative responses. A log response ratio (lnRR) meta-analysis identified flowering as the most climate-sensitive stage, with significant lnRR effects for 4 out of 8 climate variables, followed by tillering (3/8) and germination (2/8).
Overall, these results show that PhenoCam imagery can resolve inter-annual shifts in wheat phenology, identify climate-sensitive stages in rice, and validate satellite-derived phenology at daily scale. As one of the first agricultural PhenoCam deployments in India, this work demonstrates the value of near-surface imaging for bridging field and satellite observations. It reduces temporal gaps during cloudy conditions, provides ground reference for satellite calibration, and reveals stage-specific climate responses relevant for climate-resilient crop management in India.
How to cite: Kumar, A. and Khare, S.: Establishing PhenoCam-based monitoring of wheat and rice phenology in India with satellite and meteorological data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14674, https://doi.org/10.5194/egusphere-egu26-14674, 2026.
This poster introduces the open-source AquaCrop v7.2 model as a new process-based crop model within NASA's Land Information System Framework (LISF) v7.5. Through two showcases, we demonstrate the current capabilities of AquaCrop in the LISF, along with topics for future development. In a first showcase, coarse-scale crop growth simulations with various crop parameterizations are performed over Europe. Satellite-based estimates of land surface phenology are used to inform spatially variable crop parameters. These parameters improve canopy cover simulations in growing degree days compared to using uniform crop parameters in calendar days. The second showcase aims at improving fine-scale agricultural simulations via satellite data assimilation. Specifically, the crop state is updated for winter wheat fields in the Piedmont region of Italy, through assimilation of fine-scale canopy cover satellite data with an ensemble Kalman filter. The state updating is beneficial for the intermediary biomass estimates, but leads to only small improvements in yield estimates. This is due to the strong model (parameter) constraints, and limitations in the assimilated satellite observations and reference yield data. The showcases highlight pathways to improve the current constraints in the crop model and observations, and to advance future crop estimates, e.g. through crop parameter updating and multi-sensor and multi-variate data assimilation.
How to cite: De Lannoy, G., Busschaert, L., Bechtold, M., Lanfranco, N., de Roos, S., Heyvaert, Z., Mortelmans, J., Scherrer, S., Bielinis, M., Van den Bossche, M., Kumar, S., Mocko, D., Kemp, E., Heng, L., Steduto, P., and Raes, D.: Advancing Crop Modeling and Data Assimilation Using AquaCrop v7.2 in NASA's Land Information System Framework v7.5 , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9338, https://doi.org/10.5194/egusphere-egu26-9338, 2026.
Wheat crop provides a quarter of global calorie consumption. In dryland regions like Israel, spring wheat is grown under rain-fed conditions across a wide diversity of soils and agroclimatic zones. As a result, wheat grain yields suffer from high year-to-year and regional variability. With the projected climate warming intensifying water scarcity in the Eastern Mediterranean region and the global food demand rising, there is a need to develop new crop strategies for future needs.
Regional crop models allow us to assess yield and water use efficiency under future regional projected climate conditions, and thus can be used to develop such crop strategies. The current study uses a climate model (WRF) coupled to a crop model (Noah-MP-Crop) to simulate at high spatial resolution (3 km2) wheat crop growth in Israel. This approach allows accounting for feedback between the climate and the annual crop, which, in the case of widespread crops like wheat, might be significantly important. After calibration of model parameters for Israel’s commercial spring wheat fields, we run the coupled model over a 30-year period, finding a good match between model predictions and recent field observations.
Our results reveal a strong non-linear dependency of yield and water use efficiency on soil moisture. Notably, water stress exceeding 30% can trigger a rapid decline in the potential yield. Clayey soils show more resilience to moisture variability, whereas sandy soils can sometimes outperform clayey soils under greater water stress if other growth factors are optimal. This apparent yield advantage of sandy soils can be attributed to more optimal agroclimatic conditions of these soil locations. Overall, these findings demonstrate that climate-informed, site-specific management strategies, including the selection of appropriate crops and cultivars, can substantially improve yield efficiency under future climate conditions.
How to cite: Strobach, E., Ray, A., Farhi, D., and Ben-David, R.: Grain yield efficiency of dry land wheat in Israel: a high-resolution coupled crop-climate modeling approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9000, https://doi.org/10.5194/egusphere-egu26-9000, 2026.
Water productivity, defined as yield per unit consumptive water use, remains low in many marginal farming communities with landholdings of 1 ha or less, as defined by the Government of India. This persists despite the wide availability of field-scale weather forecasts, seasonal climate outlooks, and remote-sensing-based crop condition and soil moisture products, along with irrigation advisories. The challenge is not only limited information, but also the weak connection between top-down climate and crop data and bottom-up day-to-day irrigation and crop management decisions made by marginal farmers. This study develops an integrated framework to model water productivity for climate-resilient agriculture under high climate variability. In such conditions, rainfall and temperature vary strongly across seasons and years, creating uncertainty about when and how much to irrigate. This increases the risk of crop water stress or over-irrigation. Water productivity becomes critical under these conditions because it reflects how efficiently limited and uncertain water supplies are converted into yield, rather than focusing only on the volume of water applied. The research focuses on marginal farmers in semi-arid villages of Nashik District, Maharashtra, India. Top- down remote sensing data from MODIS and Sentinel-2 are used to derive evapotranspiration and NDVI, CHIRPS is used for rainfall, and ERA5 for temperature to generate initial local-scale estimates of water productivity. These estimates are then interpreted and refined using bottom-up field data. Bottom-up data collected from household surveys, focus group discussions, and participatory need assessment mapping capture farmer irrigation practices, perceived stress periods, soil moisture conditions, and decision rules. Seasonal and sub-seasonal patterns of water productivity are analysed and related to rainfall variability, temperature stress, irrigation timing, and NDVI-based crop growth dynamics. NDVI and temperature time- series fields are used to identify short stress windows and link fluctuations in water productivity to irrigation timing and crop growth stages, without overstating final yield outcomes. The framework links remote-sensing-based water productivity estimates with farmer-reported irrigation timing, irrigation method, and perceived stress, allowing fields to be grouped into short-term stress categories and relative performance classes that directly inform irrigation decisions. Comparison of satellite observations with farmer responses shows that mismatches between satellite-derived signals and farm-level outcomes arise mainly in small and fragmented plots, during short irrigation decision windows, and when advisory information lacks local relevance or trust. Results show strong variation in water productivity within small areas, driven by differences in irrigation decisions and access to usable information rather than by total consumptive water use. The study provides an integrated framework that reformulates water productivity based on remote sensing into indicators that are suitable for decision-making and are influenced by farmer participation. The framework demonstrates how combining top-down climate data with bottom-up participation can support more adaptive and equitable water use under increasing climate variability.
How to cite: Singh, M., Chinnasamy, P., and Mishra, T.: Integrating top-down remote sensing and bottom-up participatory approaches to model water productivity in marginal farming communities in Maharashtra, India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20964, https://doi.org/10.5194/egusphere-egu26-20964, 2026.
We investigated the evolution of surface-atmosphere interactions in two strategic irrigated agricultural frontiers within the Brazilian Cerrado: The Alto Rio Preto and São Marcos basins. Driven by agricultural expansion, regional landscapes underwent transformations over the last four decades. In Alto Rio Preto, savanna cover decreased from 42% to 30 %, while agriculture expanded to occupy 42% of the basin. In São Marcos, native vegetation retreat was more severe, dropping from 42% to 17%, yielding space to an agricultural matrix that dominates 43% of the total area, with irrigated agriculture already consolidating 10% of the territory. The central objective was to quantify how the replacement of native vegetation with rainfed systems and the subsequent implementation of irrigation altered regional ecosystem carbon and water dynamics. The methodology employed Landsat time series (30 meters) for field-scale vegetation mapping. Gross Primary Productivity (GPP) was estimated using an adapted Light Use Efficiency (LUE) model, and Evapotranspiration (ET) was obtained via the geeSEBAL model. Ecosystem Water Use Efficiency (WUEeco) was calculated as the ratio of carbon uptake to water loss (GPP/ET). The analysis was stratified into three levels: (i) regional spatiotemporal dynamics; (ii) trends in constant Land Use/Land Cover (LULC) areas; and (iii) impacts in technological transition zones. Regionally, a robust growth trend in GPP was observed, with mean annual values rising to unprecedented levels in both basins. Notably, over the last decade, the extent of high-productivity areas expanded significantly, becoming the dominant landscape feature in the São Marcos Basin. Water consumption followed this dynamic but with distinct regional behaviors: while ET in São Marcos remained stable at elevated levels, Alto Rio Preto underwent a structural shift, marked by a drastic reduction in low-consumption areas and a transition toward a regime of higher mean evapotranspiration. Consequently, mean annual WUEeco in both basins rose from <1.0 to >2.0 gC/mm, indicating that carbon uptake increments proportionally outpaced ET rates. The analysis of constant land use areas revealed distinct intensification strategies. Although forests maintained the highest absolute GPP and ET averages, anthropogenic systems showed the highest acceleration rates. The São Marcos basin was distinguished by the efficiency of rainfed agriculture, recording the highest relative productivity leap (+169%) while operating with stable water consumption, culminating in a 189% rise in WUEeco. In technological transition areas (rainfed to irrigated), the year 2000 marked a clear inflection point. From this date onwards, transition areas consistently outperformed rainfed GPP. Post-2010, WUEeco values converged between irrigated and rainfed areas, suggesting technical maturity. It is concluded that agricultural modernization has established a new regional paradigm: cultivated systems have attained water use efficiency levels that significantly contrast with historical baselines, resulting in a highly productive landscape that maintains resilience despite the extensive replacement of native vegetation.
How to cite: Alves, W., Ruhoff, A., Ramalho, N., and Laipelt, L.: Dynamics of carbon-water coupling in the Brazilian Cerrado: A long-term comparison of natural and agricultural systems (1985–2024) , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15617, https://doi.org/10.5194/egusphere-egu26-15617, 2026.
Rainfed agriculture is widely practiced across Mediterranean landscapes; however, its strong dependence on seasonal weather conditions makes it particularly vulnerable to drought and heat stress. Under projected increases in the frequency and intensity of these extremes due to climate change, the agricultural sector requires timely and reliable information to support planning and adaptation strategies. The main objective of this study is to investigate the response of a rainfed crop to climate variability at both seasonal and decadal timescales. The crop examined is barley, a major rainfed cereal cultivated in semi-arid and Mediterranean regions. Cyprus, located in the Eastern Mediterranean, is used as a case study. Crop growth is simulated using the Noah Land Surface Model with multi-parameterization options and a crop module (Noah-MP-Crop). The model is calibrated and evaluated against observations of evapotranspiration and net ecosystem exchange measured by an eddy covariance flux tower, soil moisture from sensors at multiple depths, leaf area index, and crop yield for the period 2020 - 2025 at an agricultural site in the central plain of the island. The long-term average rainfall for the site is 315 mm. To assess climate impacts on crop growth and yield, the calibrated model is subsequently applied across Cyprus, focusing on areas under rainfed barley cultivation. Climate impacts are analysed at seasonal and decadal scales using two simulation experiments driven by bias-corrected and statistically downscaled climate datasets. Seasonal simulations are forced by ECMWF SEAS5, while decadal simulations are based on EC-Earth3 DCCP CMIP6. All simulations are conducted at a spatial resolution of 0.1° for the period 1982–2016.
This research received financial support from the European Union under the PREVENT Project (GAP 101081276).
How to cite: Sofokleous, I., Zittis, G., Strobach, E., Djuma, H., Christodoulou, N., Savvides, A., and Bruggeman, A.: Rainfed agriculture under a changing climate: Investigating barley crop growth at seasonal and decadal timescales, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16845, https://doi.org/10.5194/egusphere-egu26-16845, 2026.
As climate change reshapes environmental and human systems, climate analogues — present-day locations whose climate resembles projected future conditions1 — are increasingly used to support climate mitigation or adaptation. However, most existing applications rely on coarse climate model outputs and raw variables, often neglecting microclimatic variability and sector-specific climatic constraints.
Using European vineyards as a case study2, we investigate how incorporating problem-specific refinements affects climate analogue identification. We enhance the classical analogue framework by introducing bioclimatic indices tailored to vine growth and pathogen development, applying sub-grid temperature corrections based on elevation, slope, and aspect, and using Principal Component Analysis to weight their contributions and thereby reduce redundancy among indices. A systematic sensitivity analysis quantifies the individual impact of each refinement on the spatial distribution of the selected analogues.
All three refinements exert a significant impact on analogue identification with similar magnitude. While the generalized distance statistics between reference sites and their analogues remain relatively stable when changing parametrizations, the geographic location of analogues can shift by several hundred to over a thousand kilometres, in some cases altering matches at the continental scale. These results emphasise the significant impact of variable selection, their interdependence, and the local climate variability on climate analogue outcomes. Consequently, problem-specific considerations are essential to ensure that the identified analogues are truly relevant to the application of interest3. While developed for viticulture, the proposed framework is readily transferable to other climate-sensitive systems, including agriculture, ecosystem management, and urban planning, underscoring the need for problem-specific climate analogue methodologies.
1 G. Rohat, S. Goyette, J. Flacke, Characterization of European cities’ climate shift– an exploratory study based on climate analogues, International Journal of Climate Change Strategies and Management (2017)
2 H. Allaman, S. Goyette, P.-H. Dubuis, J. Kasparian, Future viability of European vineyards using bioclimatic climate analogues, Agricultural and Forest Meteorology (2026)
3 H. Allaman, S. Goyette, P.-H. Dubuis, J. Kasparian, Sensitivity of Climate Analogues to Problem-Specific Adjustments: A Case Study, Manuscript submitted for publication, Under review
How to cite: Allaman, H., Goyette, S., Dubuis, P.-H., and Kasparian, J.: Sensitivity of Climate Analogues to Problem-Specific Adjustments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12455, https://doi.org/10.5194/egusphere-egu26-12455, 2026.
Regional crop production is increasingly affected by climate variability, creating a need for operational monitoring and early-warning systems based on Earth observation (EO). In this study, we present an end-to-end EO-driven framework for regional maize monitoring in Lombardy (Northern Italy), combining annual maize mapping (2017–2025) with early-season yield forecasting over the same period.
Maize distribution is mapped annually at 10 m resolution using Sentinel-2 imagery. A LightGBM classifier is trained on phenology-based NDVI features derived from seasonal composites. Training data are obtained from official crop-type raster maps of the Lombardy Regional Agricultural Information System and supplemented with provincial parcel data for 2024. To reduce commission errors, classification is restricted to cropland using the DUSAF “arable land” mask provided by Regione Lombardia.
Maize yield forecasting relies exclusively on early-season information defined in thermal time (GDD < 1200). Field-level features are extracted by GDD stages from multiple EO and meteorological sources, including Sentinel-2 L2A spectral indices, Sentinel-1 GRD VV/VH backscatter, MODIS land surface temperature, evapotranspiration (ET/PET), and LAI/FPAR, ERA5-Land daily temperature, precipitation, radiation and soil moisture (with vapor pressure deficit derived), SMAP surface and root-zone soil moisture, and static terrain and soil properties from NASADEM and SoilGrids.
A stacking ensemble model (Random Forest, Gradient Boosted Decision Trees, and XGBoost with a ridge regression meta-learner) is trained on an independent field-level maize yield dataset from Spain, linearly calibrated, and transferred to Lombardy. Regional and provincial yield estimates are further bias-corrected using standardized early-season anomaly features and an independent drought indicator (PDSI). When evaluated against official Lombardy maize yield statistics (7-province average), the anomaly- and PDSI-based correction substantially improves interannual performance, reducing RMSE from 1.20 to 0.53 t ha⁻¹ and increasing explained variance to R² ≈ 0.73.
Overall, the proposed framework shows how phenology-based crop mapping and early-season, multi-source EO information can be integrated into a practical regional system for maize monitoring and yield forecasting, supporting climate risk assessment and adaptation planning.
How to cite: Chen, J., Franch, B., Mariani, S., and Corbari, C.: Maize yield forecasting in Lombardy region in Italy using a machine learning model driven by remote sensing data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16378, https://doi.org/10.5194/egusphere-egu26-16378, 2026.
High-resolution soil moisture (SM) information is critical for irrigation decision-making, crop modelling, flood and drought prediction, and water resources management. However, satellite products only provide coarse-resolution data that cannot capture farm-scale spatial variability influenced by factors such as soil heterogeneity, topography, and anthropogenic activities. While downscaling methods offer a potential solution, they currently struggle in data-scarce regions, such as India, where the absence of dense observation networks limits their effectiveness. In this study, we present an irrigation optimisation framework that downscales satellite-derived soil moisture (SM) data to field-scale root zone soil moisture (RZSM) to support data-driven irrigation decision-making in Nashik District, Maharashtra, India. Utilising a Convolutional Long Short-Term Memory (ConvLSTM) network, we integrated sparsely located in-situ data from ground-based sensors with remote sensing predictors, including precipitation, vegetation indices, land surface temperature, and terrain attributes. The ConvLSTM architecture captures non-linear spatial and temporal interactions governing the field-scale SM variability. The models achieved strong performance, with Root Mean Square Error (RMSE) values from 0.02 to 0.08 m³/m³, Mean Absolute Error (MAE) values from 0.02 to 0.06 m³/m³, Correlation Coefficient (r) values ranging from 0.79 to 0.92, and Coefficient of Determination (R²) values between 0.61 and 0.88. These results validate the potential of deep learning for accurate field-scale SM estimation without requiring dense ground networks. Building on this, we are currently extending the framework by coupling the ConvLSTM architecture with a farm-scale ecohydrological model. This hybrid approach enables generalised, field-scale mapping at ungauged locations without in-situ sensors, offering a scalable, scientifically grounded solution for precision agriculture in water-stressed regions. This work can support farmers in making informed irrigation decisions and contribute to improved water management practices.
How to cite: Patoo, U. H., Arora, C., and Ghosh, S.: Deep Learning Based Soil Moisture Downscaling Framework for Precision Agriculture in Data-Scarce Regions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1082, https://doi.org/10.5194/egusphere-egu26-1082, 2026.
Posters virtual: Tue, 5 May, 14:00–18:00 | vPoster spot 2
EGU26-16034 | ECS | Posters virtual | VPS5
Modelling the groundwater pumping for agriculture in the Noah-MP model to support sustainable water management over the North China PlainTue, 05 May, 15:06–15:09 (CEST) vPoster spot 2
The intensive irrigation-linked groundwater abstraction in North China Plain (NCP) is dramatically affecting the hydrological processes and regional climate. Impacts from these anthropogenic groundwater withdrawals are evident in the fluctuation of each component in the terrestrial water cycle, the lack of groundwater sustainability, and regional climate extremes. Ensuring future groundwater security within this context will largely depend on how accurately the human activities in the Human-Earth system model were represented. However, to date, most hydrological models and land surface models either ignore the representation of human intervention or realistically model sophisticated human activity processes. In this study, we incorporated two groundwater-fed irrigation schemes in the Noah-MP model and further used realistic irrigation water use results constraining irrigation water withdrawals. We evaluate the influence of the groundwater pumping representation on the simulation of evapotranspiration and groundwater water table depth using Fluxnet-MTE ET data and observational groundwater well data, respectively. The Noah-MP simulation with groundwater-fed irrigation produced ET that matched the magnitude of observations-based Fluxnet-MTE ET values. Observational well-depth anomaly fluctuations can be reproduced in irrigated areas within the groundwater-fed simulation. In addition, the improvement of groundwater pumping also helps to improve terrestrial water storage estimates in higher resolution. We estimated that, over a seasonal cycle, groundwater-fed irrigation in the model can account for 80% of the declining terrestrial water storage trend from 2003 to 2016. Our approach and results reinforce the importance of parameterizing human activities in the Human-Earth system model and better address the water security challenges under climate change and human interventions.
How to cite: dai, D.: Modelling the groundwater pumping for agriculture in the Noah-MP model to support sustainable water management over the North China Plain, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16034, https://doi.org/10.5194/egusphere-egu26-16034, 2026.
EGU26-21818 | ECS | Posters virtual | VPS5
Deciphering Olive Yield Determinants under Contrasting Water Regimes: A Multi-Site Machine Learning Approach in Morocco Agro-EcosystemsTue, 05 May, 15:09–15:12 (CEST) vPoster spot 2
Improving olive yield in Moroccan agro-ecosystems requires a better understanding of the interactions between water availability, soil properties, and management practices. The complexity and non-linear nature of these interactions limit the effectiveness of conventional analytical approaches. This study applies machine learning methods to predict olive yield and to assess how the importance of yield determinants varies under contrasting water regimes. A multi-site dataset from Moroccan olive groves, including more than 2,000 observations, was analyzed. Machine learning models showed high predictive accuracy across water regimes. Under rainfed conditions, CatBoost achieved the best performance (R² = 0.845), indicating that yield variability is mainly driven by soil properties and spatial context. Under irrigated conditions, XGBoost provided the highest accuracy (R² = 0.855), highlighting the increasing role of management practices such as planting density and nitrogen fertilization. Under intensive irrigation, fruit-related variables, particularly 100-fruit weight, became the dominant predictors, while the influence of edaphic constraints decreased.
Overall, the results demonstrate that irrigation does not simply increase olive yield but fundamentally alters the hierarchy of factors controlling production. These findings emphasize the need for data-driven, site-specific management strategies to enhance the sustainability and efficiency of olive production in Morocco.
How to cite: Azamz, R., Elyoussfi, H., Benzhair, F., El Mousadik, R., and Belaqziz, S.: Deciphering Olive Yield Determinants under Contrasting Water Regimes: A Multi-Site Machine Learning Approach in Morocco Agro-Ecosystems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21818, https://doi.org/10.5194/egusphere-egu26-21818, 2026.
Posters virtual: Thu, 7 May, 14:00–18:00 | vPoster spot 2
EGU26-231 | ECS | Posters virtual | VPS6
vLeaf@DSSAT: integrating leaf energy balance and biochemistry into CERES-Maize to reassess water-efficient ideotypesThu, 07 May, 14:45–14:48 (CEST) vPoster spot 2
Future food security will increasingly depend on the development of crop ideotypes that produce higher yields per unit of water used. Stomata are central to developing water-efficient crop ideotypes, as they serve as the primary gateway for carbon and water exchange. Process-based crop models are essential tools for testing crop phenotypes with favorable stomatal traits, as they can explain how changes in stomatal traits propagate to whole canopy carbon gain and water use. Yet, current models still struggle to connect leaf-level physiology to season-long canopy performance (e.g., yield) under realistic climate variability.
Current process-based models have one of these limitations: (i) lack of explicit biochemical photosynthesis module for C3 or C4 crops, preventing mechanistic analysis of crop phenotypes; and (ii) models that explicitly represent biochemistry ignore leaf energy balance dynamics and assume leaf temperature (Tleaf) equal to air temperature (Tair), ignoring the feedback between stomatal conductance, transpiratory cooling. As a result, they require extensive empirical calibration and are not recommended for exploring novel stomatal phenotypes, such as lower stomatal density and pore size. In particular, current efforts to manipulate stomatal traits in crops cannot be reliably evaluated using these simplifications, as they do not account for how changes in stomata affect CO2 diffusion and canopy energy balance.
This study presents a novel cross-scale framework, vLeaf@DSSAT, where we couple a process-based leaf model with the CERES-Maize growth model and introduce a two-leaf (sunlit–shaded) canopy representation. The explicit consideration of energy balance makes this framework distinct from similar attempts in the past. CERES-Maize provides daily crop state variables such as leaf area index (LAI), phenology, soil water status, and nitrogen status. Using these, vLeaf then computes hourly net assimilation and transpiration rates for both sunlit and shaded leaf areas. It computes photosynthesis, stomatal conductance, boundary-layer conductance, and leaf energy balance simultaneously in an iterative loop. Root water uptake from CERES-Maize constrains canopy transpiration; vLeaf then reruns under these constraints and updates Tleaf and gas exchange rates. The resulting canopy-scale assimilation from vLeaf drives the biomass accumulation in CERES-Maize on the next day, closing the loop between leaf biophysics and crop growth.
Simulations for climates based on a US Midwest reference site show that neglecting leaf energy balance results in sizeable errors in both carbon gain and water use. For cooler climates, forcing Tleaf = Tair underestimates seasonal carbon gain by ≈ 9% and transpiration by ≈ 30%. For warmer climates, the bias in carbon gain is small, but transpiration is overestimated by 5–10%. These errors can create uncertainty in ranking crop phenotypes with favorable stomatal traits. vLeaf@DSSAT provides a practical approach to testing stomatal manipulation, irrigation strategies, and climate-resilient ideotypes under realistic climate conditions, while also connecting leaf biophysics to field-scale yield and water use.
How to cite: Srivastava, A.: vLeaf@DSSAT: integrating leaf energy balance and biochemistry into CERES-Maize to reassess water-efficient ideotypes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-231, https://doi.org/10.5194/egusphere-egu26-231, 2026.
