Simulate Crop Yield Research Articles

Modelling the effects of soil organic content and pH on the yield responses of tea to nitrogen fertilizer

CONTEXTSustained high yields of tea rely on the supply of nitrogen (N) from soil reserves, typically maintained by N fertilisation from inorganic or organic sources. OBJECTIVEThis paper describes how soil N levels, including the effects of soil organic content and pH, were developed and incorporated into a crop yield simulation model called CUPPA-Tea. METHODSThe nitrogen dynamics are presented in terms of i) the initial nitrogen stocks, ii) the addition of nitrogen to the system, iii) the uptake, use and loss of nitrogen by tea plants, and iv) nitrogen flows within the soil. CUPPA-Tea was then calibrated and validated using measured tea yields from Tanzania and Kenya. RESULTS AND CONCLUSIONSAfter integrating a wide range of nitrogen algorithms, the model explained 79% of the variation in annual yields within a nitrogen and irrigation experiment in Tanzania and a fertilizer experiment in Kenya. The slope of the relationship was 0.84 and 0.73 respectively, the root mean square error was 660 kg ha−1 and 507 kg ha−1, and the modelling efficiency was 0.77 and 0.75 respectively. The model predicted that in the absence of N application, tea yields would be higher from a site with a high rather than a low soil organic content. By contrast, at high levels of mineral N application, the yield response in the model was not sensitive to the soil organic content. Hence within the model, a site in Tanzania with a low soil organic content of 1.6% showed a greater yield response to applied mineral N than a site in Kenya where the soil organic carbon was 4.0%. The model also predicted small losses of N from the cropping system through denitrification and leaching due to the acidic soil conditions (pH < 4.5) and an assumed tea rooting depth between 300 and 500 cm. In Tanzania, irrigation was predicted to result in around 10% higher nitrogen uptake than under unirrigated conditions. SIGNIFICANCEThe use of the CUPPA-Tea model can be useful in supporting decision making and improving the accuracy of tea yield estimates, as well as predictions of N fate within the soil-plant-atmosphere continuum.

Seasonal soil moisture and crop yield prediction with fifth-generation seasonal forecasting system (SEAS5) long-range meteorological forecasts in a land surface modelling approach

Abstract. Long-range weather forecasts provide predictions of atmospheric, ocean and land surface conditions that can potentially be used in land surface and hydrological models to predict the water and energy status of the land surface or in crop growth models to predict yield for water resources or agricultural planning. However, the coarse spatial and temporal resolutions of available forecast products have hindered their widespread use in such modelling applications, which usually require high-resolution input data. In this study, we applied sub-seasonal (up to 4 months) and seasonal (7 months) weather forecasts from the latest European Centre for Medium-Range Weather Forecasts (ECMWF) seasonal forecasting system (SEAS5) in a land surface modelling approach using the Community Land Model version 5.0 (CLM5). Simulations were conducted for 2017–2020 forced with sub-seasonal and seasonal weather forecasts over two different domains with contrasting climate and cropping conditions: the German state of North Rhine-Westphalia (DE-NRW) and the Australian state of Victoria (AUS-VIC). We found that, after pre-processing of the forecast products (i.e. temporal downscaling of precipitation and incoming short-wave radiation), the simulations forced with seasonal and sub-seasonal forecasts were able to provide a model output that was very close to the reference simulation results forced by reanalysis data (the mean annual crop yield showed maximum differences of 0.28 and 0.36 t ha−1 for AUS-VIC and DE-NRW respectively). Differences between seasonal and sub-seasonal experiments were insignificant. The forecast experiments were able to satisfactorily capture recorded inter-annual variations of crop yield. In addition, they also reproduced the generally higher inter-annual differences in crop yield across the AUS-VIC domain (approximately 50 % inter-annual differences in recorded yields and up to 17 % inter-annual differences in simulated yields) compared to the DE-NRW domain (approximately 15 % inter-annual differences in recorded yields and up to 5 % in simulated yields). The high- and low-yield seasons (2020 and 2018) among the 4 simulated years were clearly reproduced in the forecast simulation results. Furthermore, sub-seasonal and seasonal simulations reflected the early harvest in the drought year of 2018 in the DE-NRW domain. However, simulated inter-annual yield variability was lower in all simulations compared to the official statistics. While general soil moisture trends, such as the European drought in 2018, were captured by the seasonal experiments, we found systematic overestimations and underestimations in both the forecast and reference simulations compared to the Soil Moisture Active Passive Level-3 soil moisture product (SMAP L3) and the Soil Moisture Climate Change Initiative Combined dataset from the European Space Agency (ESA CCI). These observed biases of soil moisture and the low inter-annual differences in simulated crop yield indicate the need to improve the representation of these variables in CLM5 to increase the model sensitivity to drought stress and other crop stressors.

Performance evaluation of vertical bifacial and single-axis tracked agrivoltaic systems on arable land

This study evaluates and compares two types of interspersed bifacial agrivoltaic systems in Belgium: a fixed vertical system and a dynamic single-axis tracker, focusing on sugarbeet cultivation. Additionally, the impact of different tracking algorithms on crop yield and quality is investigated. The main approach involves developing an empirical crop model based on radiation use efficiency (RUE) and utilizing a radiation tool to simulate crop yield and quality. Field measurements of electricity and crop output are conducted over a span of two years, 2021 and 2022. The findings reveal that the dynamic solar tracker outperforms the fixed vertical bifacial setup in both years. The smart-tracking algorithm, applied under optimal watering conditions in the 2021 season, leads to a significant increase in energy yield (＋30%) and land use efficiency (＋20%) at a lower cost, while maintaining comparable crop yields to the vertical setup. However, limitations are observed in the empirical crop growth model’s ability to account for climatic variability in the dry 2022 year, reducing its usefulness during the design phase. Notably, the practical implementation of the systems highlights challenges during construction and exploitation, emphasizing the significance of considering these practical factors together with climate, structure and crop choice assessing the effectiveness and to de-risk investments.

Assessment of the sustainability of groundwater utilization and crop production under optimized irrigation strategies in the North China Plain under future climate change

Over-exploitation of groundwater due to intensive irrigation and anticipated climate change pose severe threats to the water and food security worldwide, particularly in the North China Plain (NCP). Limited irrigation has been recognized as an effective way to improve crop water productivity and slow the rapid decline of groundwater levels. Whether optimized limited irrigation strategies could achieve a balance between groundwater pumping and grain production in the NCP under future climate change deserves further study. In this study, an improved Soil and Water Assessment Tool (SWAT) model was used to simulate climate change impacts on shallow groundwater levels and crop production under limited irrigation strategies to suggest optimal irrigation management practices under future climate conditions in the NCP. The simulations of eleven limited irrigation strategies for winter wheat with targeted irrigations at different growth stages and with irrigated or rainfed summer maize were compared with future business-as-usual management. Climate change impacts showed that mean wheat (maize) yield under adequate irrigation was expected to increase by 13.2% (4.9%) during the middle time period (2041–2070) and by 11.2% (4.6%) during the late time period (2071–2100) under three SSPs compared to the historical period (1971–2000). Mean decline rate of shallow groundwater level slowed by approximately 1 m a−1 during the entire future period (2041–2100) under three SSPs with a greater reduction for SSP5–8.5. The average contribution rate of future climate toward the balance of shallow groundwater pumping and replenishment was 62.9%. Based on the simulated crop yields and decline rate of shallow groundwater level under the future climate, the most appropriate limited irrigation was achieved by applying irrigation during the jointing stage of wheat with rainfed maize, which could achieve the groundwater recovery and sustainable food production.

Direct and indirect simulating and projecting hydrological drought using a supervised machine learning method

There is a trend in using Artificial Intelligence methods as simulation tools in different aspects of hydrology, including river discharge simulations, drought predictions, and crop yield simulations. The motivation of this work was to assess two various concepts in applying these methods in simulations and projections of hydrological drought. In this study, Standardized Runoff Index (SRI) was simulated and projected using Artificial Neural Networks (ANNs). Maximum and minimum temperature, precipitation, and meteorological drought indicators (the Standardized Precipitation Index (SPI)) were selected as predictors. A direct approach (directly simulating and projecting SRI) and an indirect approach (simulating and projecting river discharge, then calculating SRI) were assessed. Our results show that the indirect approach performs better than the direct approach in simulations of SRI in four discharge stations in the Odra River Basin (a transboundary river basin in Central Europe) from 2000 to 2019. Moreover, a considerable difference between these two approaches was detected in projections of hydrological drought under the RCP8.5 emission scenario for two horizons (near future: 2021–2040, and far future: 2041–2060). Based on the run theory, both approaches show somewhat similar drought conditions for future projections.

Impact of different quantity and quality of irrigation water on crop yield and biomass of winter maize using FAO-Aqua crop model

Irrigation has a major role to play in the productivity of winter maize. Precise information about the quantity and quality of irrigation water is the key for higher productivity of winter maize. In the present study attempt has been made to asses the impact of different depth of irrigation water on crop yield and biomass of winter maize using FAO-Aquacrop Model. In the first case crop yield and biomass was simulated for irrigation water depth varied from 20 mm to 80 mm, keeping the irrigation water quality constant. Similarly, in another case the optimum irrigation depth was kept constant and irrigation water quality varied from 1 to 10 ds/m. The simulated crop yield and biomass increases up to 40 cm depth of irrigation water application for all three seasons. When a similar comparison was made for 30 cm depth of irrigation water application the simulated yield reduction was only 0.79%, 2.2% and 2.4 % for the year 2016-17, 2017-18 and 2018-19 respectively. The analysis suggested that this yield reduction can easily be compromised for saving 10 cm of irrigation water. This study indicated that 30 cm depth of irrigation water is optimum for Winter maize in BurhiGandak river basin of North Bihar In case of deficit irrigation of 20 cm depth of irrigation water application the simulated yield reduced by 14.4 %, 25.4 % and 11.4 % for the year 2016-17, 2017-18 and 2018-19 respectively. Assessment of response of different quality irrigation water on simulated crop yield and biomass of winter maize using FAO-Aquacrop model suggests that simulated yield was found maximum with 1 ds/m. The reduction in simulated yield with 10 ds/m water quality was observed maximum with a value of 41.3 %, 44.4 % and 38.4 % respectively for the year 2016-17, 2017-18 and 2018-19. FAO-Aquacrop model can be used as an important tool for efficient planning of irrigation water under diminishing water supply and deteriorating water quality.

Simulating crop yield using the DSSAT v4.7-CROPGRO-soyabean model with gridded weather and soil data

Simulating crop yield using the DSSAT v4.7-CROPGRO-soyabean model with gridded weather and soil data

Simulation of crop yield using the global hydrological model H08 (crp.v1)

Abstract. A better understanding of the food–water nexus requires the development of an integrated model that can simultaneously simulate food production and the requirements and availability of water resources. H08 is a global hydrological model that considers human water use and management (e.g., reservoir operation and crop irrigation). Although a crop growth sub-model has been included in H08 to estimate the global crop-specific calendar, its performance as a yield simulator is poor, mainly because a globally uniform parameter set was used for each crop type. In addition, the effects of CO2 fertilization and vapor pressure deficit on crop yield were not considered. Here, through country-wise parameter calibration and algorithm improvement, we enhanced H08 to simulate the yields of four major staple crops: maize, wheat, rice, and soybean. The simulated crop yield was compared with the Food and Agriculture Organization (FAO) national yield statistics and the global dataset of historical yield for major crops (GDHY) gridded yield estimates with respect to mean bias (across nations) and time series correlation (for individual nations). Our results showed that the effects of CO2 fertilization and vapor pressure deficit had opposite impacts on crop yield. The simulated yield showed good consistency with FAO national yield. The mean biases of the major producer countries were considerably reduced to 2 %, 2 %, −2 %, and −1 % for maize, wheat, rice, and soybean, respectively. The capacity of our model to capture the interannual yield variability observed in FAO yield was limited, although the performance of our model was comparable to that of other mainstream global crop models. The grid-level analysis showed that our model showed a similar spatial pattern to that of the GDHY yield in terms of reproducing the temporal variation over a wide area, although substantial differences were observed in other places. Using the enhanced model, we quantified the contributions of irrigation to global food production and compared our results to an earlier study. Overall, our improvements enabled H08 to estimate crop production and hydrology in a single framework, which will be beneficial for global food–water nexus studies in relation to climate change.

Crop yield prediction algorithm (CYPA) in precision agriculture based on IoT techniques and climate changes

Agriculture faces a significant challenge in predicting crop yields, a critical aspect of decision-making at international, regional, and local levels. Crop yield prediction utilizes soil, climatic, environmental, and crop traits extracted via decision support algorithms. This paper presents a novel approach, the Crop Yield Prediction Algorithm (CYPA), utilizing IoT techniques in precision agriculture. Crop yield simulations simplify the comprehension of cumulative impacts of field variables such as water and nutrient deficits, pests, and illnesses during the growing season. Big data databases accommodate multiple characteristics indefinitely in time and space and can aid in the analysis of meteorology, technology, soils, and plant species characterization. The proposed CYPA incorporates climate, weather, agricultural yield, and chemical data to facilitate the anticipation of annual crop yields by policymakers and farmers in their country. The study trains and verifies five models using optimal hyper-parameter settings for each machine learning technique. The DecisionTreeRegressor achieved a score of 0.9814, RandomForestRegressor scored 0.9903, and ExtraTreeRegressor scored 0.9933. Additionally, we introduce a new algorithm based on active learning, which can enhance CYPA's performance by reducing the number of labeled data needed for training. Incorporating active learning into CYPA can improve the efficiency and accuracy of crop yield prediction, thereby enhancing decision-making at international, regional, and local levels.

Measured and modeled nitrogen balances in lowland rice-pasture rotations in temperate South America

Rotational rice systems, involving pastures, other crops and/or livestock, are common in temperate South America, exemplified by the rice-pasture-livestock system of Uruguay which combines very high rice yields with tight nitrogen (N) balances. The generally good nutrient use efficiency in these systems provides a template for nutrient management in other mixed farming systems, if the underlying processes can be sufficiently well quantified and understood. Here, we studied N balances in rice–non-rice rotations in a long-term experiment in Uruguay, with the aim of parameterizing and testing the DNDC model of N dynamics for such systems for use in future work. The experiment includes three rotations: continuous rice (RI-CONT), rice-soybean (RI-SOY) and rice-pasture (RI-PAST). We considered 9 years of data on N balances (NBAL), defined as all N inputs minus all N outputs; N surplus (NSURP), defined as all N inputs minus only N outputs in food products; and N use efficiency (NUE), defined as the fraction of N inputs removed in food products. We parameterized DNDC against measured yield and input and output data, with missing data on N losses inferred from the N balance and compared with literature values. The model performance was assessed using standard indices of mean error, agreement and efficiency. The model simulated crop yields and rice cumulative N uptake very well, and soil N reasonably well. The values of NBAL were +45 and−20 kg N ha−1 yr−1 in RI-CONT and RI-SOY, respectively, and close to zero in RI-PAST (−6 kg N ha−1 yr−1). Values of NSURP decreased in the order RI-CONT &amp;gt;&amp;gt; RI-SOY &amp;gt; RI-PAST (+115, +25 and +13 kg N ha−1 yr−1, respectively). Values of NUE (84, 54, and 48% for RI-SOY, RI-PAST, and RI-CONT, respectively) decreased as NBAL increased. The sensitivity of DNDC's predictions to the agronomic characteristics of the different crops, rotations and water regimes agreed with expectations. We conclude that the DNDC model as parameterized here is suitable for exploring how to optimize N management in these systems.

Modelling maize yield impacts of improved water and fertilizer management in southern Africa using cropping system model coupled to an agro-hydrological model at field and catchment scale

AbstractThis study quantifies the effect of fertilizer and irrigation management on water use efficiency (WUE), crop growth and crop yield in sub-humid to semi-arid conditions of Limpopo Province, South Africa. An approach of coupling a cropping system model (DSSAT) with an agro-hydrological model (SWAT) was developed and applied to simulate crop yield at the field and catchment scale. Simulation results indicated that the application of fertilizer has a greater positive effect on maize yield than irrigation. WUE ranged from 0.10–0.57 kg/m3 (rainfed) to 0.84–1.39 kg/m3 (irrigated) and was positively correlated with fertilizer application rate. The combined application of the variants with deficit irrigation and fertilizer rate (120:60 kg N:P/ha) for maize turned out to be the best option, giving the highest WUE and increasing average yield by up to 5.7 t/ha compared to no fertilization and rainfed cultivation (1.3 t/ha). The simulated results at the catchment scale showed the considerable spatial variability of maize yield across agricultural fields with different soils, slopes and climate conditions. The average annual simulated maize yield across the catchment corresponding to the highest WUE ranged from 4.0 to 7.0 t/ha. The yield gaps ranged from 3.0 to 6.0 t/ha under deficit irrigation combined with 120N:60P kg/ha and ranged from 0.2 to 1.5 t/ha when only applying deficit irrigation but no fertilizer. This information can support regional decision makers to find appropriate interventions that aim at improving crop yield and WUE for catchments/regions.

Modelling coffee yield in Dak Lak province using machine learning algorithms

Application of machine learning algorithms in simulating crop yield has attracted more attention from plenty of scientists in recent years. The objective of this study is to estimate the coffee yields in Dak Lak province by using three machine learning algorithms, namely, artificial neural network (ANN), support vector regression (SVR), and random forest (RF), respectively. Input data in simulating processes includes maximum and minimum temperature, effective rainfall, reference evapotranspiration, and crop water requirement in the period 2000-2020. In which, the percentage of data in the training and testing phases is 70% and 30%, respectively. The results indicated that three machine learning models (i.e., SVR, ANN, and RF) have reasonable performance in simulating the coffee yield, out of which, the RF model performs best with NSE values of approximately 0.918 for the training phase and 0.818 for the testing phase.

A 258-year-long data set of temperature and precipitation fields for Switzerland since 1763

Abstract. Climate reconstructions give insights in monthly and seasonal climate variability in the past few hundred years. However, for understanding past extreme weather events and for relating them to impacts, for example through crop yield simulations or hydrological modelling, reconstructions on a weather timescale are needed. Here, we present a data set of 258 years of daily temperature and precipitation fields for Switzerland from 1763 to 2020. The data set was reconstructed with the analogue resampling method, which resamples meteorological fields for a historical period based on the most similar day in a reference period. These fields are subsequently improved with data assimilation for temperature and bias correction for precipitation. Even for an early period prior to 1800 with scarce data availability, we found good validation results for the temperature reconstruction especially in the Swiss Plateau. For the precipitation reconstruction, skills are considerably lower, which can be related to the few precipitation measurements available and the heterogeneous nature of precipitation. By means of a case study of the wet and cold years from 1769 to 1772, which triggered widespread famine across Europe, we show that this data set allows more detailed analyses than hitherto possible.

Evaluation of MODIS, Landsat 8 and Sentinel-2 Data for Accurate Crop Yield Predictions: A Case Study Using STARFM NDVI in Bavaria, Germany

The increasing availability and variety of global satellite products and the rapid development of new algorithms has provided great potential to generate a new level of data with different spatial, temporal, and spectral resolutions. However, the ability of these synthetic spatiotemporal datasets to accurately map and monitor our planet on a field or regional scale remains underexplored. This study aimed to support future research efforts in estimating crop yields by identifying the optimal spatial (10 m, 30 m, or 250 m) and temporal (8 or 16 days) resolutions on a regional scale. The current study explored and discussed the suitability of four different synthetic (Landsat (L)-MOD13Q1 (30 m, 8 and 16 days) and Sentinel-2 (S)-MOD13Q1 (10 m, 8 and 16 days)) and two real (MOD13Q1 (250 m, 8 and 16 days)) NDVI products combined separately to two widely used crop growth models (CGMs) (World Food Studies (WOFOST), and the semi-empiric Light Use Efficiency approach (LUE)) for winter wheat (WW) and oil seed rape (OSR) yield forecasts in Bavaria (70,550 km2) for the year 2019. For WW and OSR, the synthetic products’ high spatial and temporal resolution resulted in higher yield accuracies using LUE and WOFOST. The observations of high temporal resolution (8-day) products of both S-MOD13Q1 and L-MOD13Q1 played a significant role in accurately measuring the yield of WW and OSR. For example, L- and S-MOD13Q1 resulted in an R2 = 0.82 and 0.85, RMSE = 5.46 and 5.01 dt/ha for WW, R2 = 0.89 and 0.82, and RMSE = 2.23 and 2.11 dt/ha for OSR using the LUE model, respectively. Similarly, for the 8- and 16-day products, the simple LUE model (R2 = 0.77 and relative RMSE (RRMSE) = 8.17%) required fewer input parameters to simulate crop yield and was highly accurate, reliable, and more precise than the complex WOFOST model (R2 = 0.66 and RRMSE = 11.35%) with higher input parameters. Conclusively, both S-MOD13Q1 and L-MOD13Q1, in combination with LUE, were more prominent for predicting crop yields on a regional scale than the 16-day products; however, L-MOD13Q1 was advantageous for generating and exploring the long-term yield time series due to the availability of Landsat data since 1982, with a maximum resolution of 30 m. In addition, this study recommended the further use of its findings for implementing and validating the long-term crop yield time series in different regions of the world.

Adaptation Strategies Strongly Reduce the Future Impacts of Climate Change on Simulated Crop Yields

AbstractSimulations of crop yield due to climate change vary widely between models, locations, species, management strategies, and Representative Concentration Pathways (RCPs). To understand how climate and adaptation affects yield change, we developed a meta‐model based on 8703 site‐level process‐model simulations of yield with different future adaptation strategies and climate scenarios for maize, rice, wheat and soybean. We tested 10 statistical models, including some machine learning models, to predict the percentage change in projected future yield relative to the baseline period (2000–2010) as a function of explanatory variables related to adaptation strategy and climate change. We used the best model to produce global maps of yield change for the RCP4.5 scenario and identify the most influential variables affecting yield change using Shapley additive explanations. For most locations, adaptation was the most influential factor determining the projected yield change for maize, rice and wheat. Without adaptation under RCP4.5, all crops are expected to experience average global yield losses of 6%–21%. Adaptation alleviates this average projected loss by 1–13 percentage points. Maize was most responsive to adaptive practices with a projected mean yield loss of −21% [range across locations: −63%, +3.7%] without adaptation and −7.5% [range: −46%, +13%] with adaptation. For maize and rice, irrigation method and cultivar choice were the adaptation types predicted to most prevent large yield losses, respectively. When adaptation practices are applied, some areas are predicted to experience yield gains, especially at northern high latitudes. These results reveal the critical importance of implementing adequate adaptation strategies to mitigate the impact of climate change on crop yields.

Effects of Long-Term Organic–Inorganic Nitrogen Application on Maize Yield and Nitrogen-Containing Gas Emission

A sustainable model of combined organic–inorganic fertilizer application for high maize yields and environmental health is important for food security. The short-term combined application of organic and inorganic fertilizers can improve crop yields; however, the effect of different proportions of organic and inorganic fertilizers on the maize yield and nitrogen gas emissions in a long time series has not been reported. In this study, field experiments and DeNitrification-DeComposition (DNDC) model simulations were used to study the long-term effects of substituting inorganic fertilizers with organic fertilizers on crop yields and nitrogen-containing gas emissions. Six treatments were included: no nitrogen (CK); urea (U1); and 25%, 50%, 75%, and 100% of the urea N substituted by organic fertilizers (U3O1, U1O1, U1O3, and O1, respectively). The DNDC model was calibrated using the field data from the U1 treatment from 2018 to 2020 and was validated for the other treatments. The results showed that this model could effectively simulate crop yields (e.g., nRMSE &lt; 5%), soil NH3 volatilization, and N2O emissions (nRMSE &lt; 25%). In addition, long-term (26 years) simulation studies found that the U1O1 treatment could considerably increase maize yields and ensure yield stability, which was 15.69–55.31% higher than that of the U1 treatment. The N2O, NH3, and NO emissions were in the descending order of U1 &gt; U3O1 &gt; O1 &gt; U1O3 &gt; U1O1, and the total nitrogen-containing gas emissions from the U1O1 treatment decreased by 53.72% compared with the U1 treatment (26 years). Overall, substituting 50% of inorganic nitrogen with organic nitrogen could maintain the high yield of maize and reduce emissions of nitrogen-containing gases, constituting a good mode for the combined application of organic–inorganic nitrogen in this area.

Satellite-based soil moisture enhances the reliability of agro-hydrological modeling in large transboundary river basins

Satellite-based observations of soil moisture, leaf area index, precipitation, and evapotranspiration facilitate agro-hydrological modeling thanks to the spatially distributed information. In this study, the Climate Change Initiative Soil Moisture dataset (CCI SM, a product of the European Space Agency (ESA)) adjusted based on Soil Water Index (SWI) was used as an additional (in relation to discharge) observed dataset in agro-hydrological modeling over a large-scale transboundary river basin (Odra River Basin) in the Baltic Sea region. This basin is located in Central Europe within Poland, Czech Republic, and Germany and drains into the Baltic Sea. The Soil and Water Assessment Tool+ (SWAT+) model was selected for agro-hydrological modeling, and measured data from 26 river discharge stations and soil moisture from CCI SM (for topsoil and entire soil profile) over 1476 sub-basins were used in model calibration for the period 1997–2019. Kling–Gupta efficiency (KGE) and SPAtial EFficiency (SPAEF) indices were chosen as objective functions for runoff and soil moisture calibration, respectively. Two calibration strategies were compared: one involving only river discharge data (single-objective - SO), and the second one involving river discharge and satellite-based soil moisture (multi-objective – MO). In the SO approach, the average KGE for discharge was above 0.60, whereas in the MO approach, it increased to 0.67. The SPAEF values showed that SWAT+ has acceptable accuracy in soil moisture simulations. Moreover, crop yield assessments showed that MO calibration also increases the crop yield simulation accuracy. The results show that in this transboundary river basin, adding satellite-based soil moisture into the calibration process could improve the accuracy and consistency of agro-hydrological modeling.

How can machine learning help in understanding the impact of climate change on crop yields?

Ordinary least squares linear regression (LR) has long been a popular choice among researchers interested in using historical data for estimating crop yield response to climate change. Today, the rapidly growing field of machine learning (ML) offers a wide range of advanced statistical tools that are increasingly being used for more accurate estimates of this relationship. This study compares LR to a popular ML technique called boosted regression trees (BRTs). We find that BRTs provide a significantly better prediction accuracy compared to various LR specifications, including those fitting quadratic and piece-wise linear functions. BRTs are also able to identify break points where the relationship between climate and yield undergoes significant shifts (for example, increasing yields with precipitation followed by a plateauing of the relationship beyond a certain point). Tests we performed with synthetically simulated climate and crop yield data showed that BRTs can automatically account for not only spatial variation in climate–yield relationships, but also interactions between different variables that affect crop yields. We then used both statistical techniques to estimate the influence of historical climate change on rice, wheat, and pearl millet in India. BRTs predicted a considerably smaller negative impact compared to LR. This may be an artifact of BRTs conflating time and climate variables, signaling a potential weakness of models with excessively flexible functional forms for inferring climate impacts on agriculture. Our findings thus suggest caution while interpreting the results from single-model analyses, especially in regions with highly varied climate and agricultural practices.

Comparison of process-based and statistical approaches for simulation and projections of rainfed crop yields

Accurate and comprehensive modelling aimed at investigating the impact of climate change on rainfed crop yields is of great importance due to the interconnected issues of water scarcity and food security. Because the process-based and statistical approaches to simulating crop yields are different in nature, a comparison between them is needed. This study investigates the accuracy of crop yield simulations in the historical period as well as future projections using two modelling approaches: 1) a process-based approach employing the Soil and Water Assessment Tool+ (SWAT+) model, and 2) a statistical approach employing a data-driven model, Feed Forward Back Propagation Neural Network (FFBPNN) over a medium-sized catchment in north-western Poland. The application of two potential evapotranspiration methods (Penman-Monteith and Hargreaves) in SWAT+ permitted calibration (2004–2011) and validation (2012–2019) of runoff and yields of winter wheat and spring barley. Different combinations of climatic parameters with a drought index based on Joint Deficit Index were applied to simulate and project rainfed crop yields (winter wheat, barley, potato, rye, rapeseed, sugar beets, cereals, maize for grain, maize for green forage, pulses) with FFBPNN. The results reveal that adding the new drought index helped increase the FFBPNN performance. This approach showed that future yields of the studied crops would slightly increase under RCP8.5 by 2060. Winter wheat and spring barley projections from SWAT+ showed very small changes using both the Penman-Monteith and Hargreaves method. Policy-wise, the results should be of interest to climate change adaptation practitioners and food security experts. Future studies should aim at more thorough investigation of the role of the downscaling technique and extreme events, as well as the effect of elevated CO2 on future crop yields.

Effect of nitrogen management on the environmental and economic sustainability of wheat production in different climates

Increasing chemical inputs to improve food security has resulted in worrying environmental consequences. Recently, significant efforts have been made to reduce the environmental impacts of the food supply chain. The current study examined the effects of different levels of urea fertilizer on the technical, economic, and environmental aspects of wheat production in arid and humid climates. A questionnaire was used to collect data on wheat cultivation in the 2019 and 2020 crop seasons. The calibrated and validated AquaCrop model was applied to simulate crop yield under different urea fertilizer levels of 80, 160, and 220 kg ha-1 in arid climate and 100, 180, and 250 kg ha-1 in humid climate. Economic productivity, water productivity, and nitrogen use efficiency indicators was used to assess the economic and technical sustainability of different cropping systems. The environmental impacts was investigated by using the life cycle assessment (LCA) approach involving the ReCiPe 2016 method, covering 18 midpoints and three endpoint indicators, respectively. The most important midpoint categories from foreground–background analysis were global warming, fine particulate matter formation, and terrestrial acidification. Additionally, 75% of the total environmental damage was related to the background system. On average, considering different fertilizer treatments, wheat cultivation in the arid climate had a 9% higher environmental damage potential than the humid climate. Nitrogen played the most significant role in midpoint categories, followed by diesel. The most appropriate urea fertilizer level in the arid climate, compared to the corresponding level in the humid climate, increased human health, ecosystem, and resources by 14.3%, 12.5%, and 3.2%, respectively, and decreased water and economic productivity by 375% and 8%, respectively. Based on the results, combining LCA with technical and economic indicators provides an appropriate method for prioritizing the cultivation of different crops in each county.