Figure - available from: PLOS One
This content is subject to copyright.
Phenological phases with values according to the BBCH scale available from the PHASE model and growing phases (colored) for winter wheat along the vegetation period
Source publication
Crop yields are increasingly affected by climate change-induced weather extremes in Germany. However, there is still little knowledge of the specific crop-climate relations and respective heat and drought stress-induced yield losses. Therefore, we configure weather indices (WIs) that differ in the timing and intensity of heat and drought stress in...
Citations
... The inclusion of the duration of the four stages as independent variables in the RF3 model ( Fig. 1) aims to capture the influence of stress duration (water stress or high temperature) in each stage (Hu et al., 2023;Ge et al., 2012). The important aspects of water and/or temperature stress include intensity of stress, timing of stress, and duration of stress (Riedesel et al., 2023;Cavus et al., 2023). The intensity can be captured by rainfall data combined with radiation, temperature, and soil depth, all of which are included in the training of the RF3. ...
Context: Accurately projecting crop yields under climate change is essential for understanding potential impacts
and planning of agricultural adaptation in sub-Saharan Africa (SSA). Crop growth models and machine learning
(ML) are often used, but their effectiveness is limited by data availability, precision, and geographic coverage in
SSA.
Objective: This study aimed to integrate ML with a process-based crop model to produce geographically
continuous gridded crop yield projections while reducing uncertainties associated with standalone ML or crop
growth models. As a case study, we implemented it to project the climate change impact on water-limited po�tential yield of maize across SSA.
Methods: We developed an integrated system that combines ML with eco-physiological processes to estimate
sowing dates and thermal times, ensuring that crop phenology is accounted for, thus improving potential rainfed
yield simulations under varying environmental conditions. Random Forest and crop model-based algorithms are
integrated in three steps: (i) RF1, a Random Forest model integrated with a sowing algorithm, designed to es�timate the sowing window and sowing date; (ii) RF2, a Random Forest model combined with a crop model algorithm to estimate cumulative thermal time during the growing season, used to determine the timing of
phenological stages; and (iii) RF3, another Random Forest model, trained based on eco-physiological principles
applied in phases (i) and (ii), employed to simulate water-limited potential yield. The outcomes of the different
steps of the framework under historical conditions were tested against reported data across SSA.
Results and conclusions: For maize and historical climatic conditions, the framework delivers yields which differ
less than 20 % of those simulated with a crop model with high-quality inputs, in 95 % of the cases. Our approach
thus shows value for generating crop yield projections in data-scarce regions under historical climate, and under
future climatic conditions which already feature today somewhere in SSA and for which the framework has been
trained.
Significance: Our approach can also be applied to other major food crops in SSA, under both current and climate
change conditions. It allows testing the effect of adaptation of crop cultivars in terms of maturity group. Thus, it
can be used for different crops and with far less data requirements compared to process-based crop models. It has
the potential for significant applications in assessing climate change impacts, guiding adaptation strategies, and
supporting crop breeding programmes and policymaking efforts in SSA.
... The impacts of SM droughts vary significantly with duration, intensity and timing within the year (Riedesel et al., 2023;Sippel et al., 2018). Furthermore, impacts of SM droughts depend on the considered soil depth, leading to different effects on shallow-rooted vegetation compared to deep-rooted vegetation (grasslands and agricultural land vs. forests) (Fu et al., 2020). ...
Global warming is altering soil moisture (SM) droughts in Europe with a strong drying trend projected in the Mediterranean and wetting trends projected in Scandinavia. Central Europe, including Germany, lies in a transitional zone showing weaker and diverging change signals exposing the region to uncertainties. The recent extreme drought years in Germany, which resulted in multi‐sectoral impacts accounting to combined drought and heat damages of 35 billion Euros and large scale forest losses, underline the relevance of studying future changes in SM droughts. To analyze the projected SM drought changes and associated uncertainties in Germany, we utilize a large ensemble of 57 bias‐adjusted and spatially disaggregated regional climate model simulations to run the hydrologic model mHM at a high spatial resolution of approximately 1.2 km. We show that projections of future changes in soil moisture droughts over Germany depend on the emission scenario, the soil depth and the timing during the vegetation growing period. Most robust and widespread increases in soil moisture drought intensities are projected for upper soil layers in the late growing season (July–September) under the high emission scenario. There are greater uncertainties in the changes in soil moisture droughts in the early vegetation growing period (April–June). We find stronger imprints of changes in meteorological drivers controlling the spatial disparities of SM droughts than regional diversity in physio‐geographic landscape properties. Our study provides nuanced insights into SM drought changes for an important climatic transition zone and is therefore relevant for regions with similar transitions.
... The higher volatility of combined crop yields in North-Eastern and Eastern Germany compared with other regions agrees with other studies (Albers et al 2017, Lüttger andFeike 2018). It has been related to comparatively high levels of drought and heat stress and their interaction with regional soil characteristics, i.e. the dominance of light sandy soils with low water holding capacity (Webber et al 2020, Schmitt et al 2022, Riedesel et al 2023. The relatively high area share of rapeseed (supplementary material figure A3(C)), which had a higher yield variance compared to wheat or barley (supplementary material figure A1(C)), likely contributed to the higher yield volatilities (Rondanini et al 2012). ...
Recent evidence suggests a stabilizing effect of crop diversity on agricultural production. However, different methods are used for assessing these effects and there is little systematic quantitative evidence on diversification benefits. The aim of this study was to assess the relationship between volatility of combined crop yields (denoted as standard deviation) and diversity (denoted as Shannon’s Evenness Index SEI) for standardized yield data of major crop species grown in Germany between 1977 and 2018 (winter wheat, winter barley, silage maize and winter rapeseed) at the county level. Portfolio theory was used to estimate the optimal crop area share for minimizing yield volatility. On average, results indicated a weak negative relationship between volatility and the SEI during the past decades for the case of Germany. Optimizing crop area shares for minimizing volatility reduced yield variance on average by 24% but was associated with a decrease in SEI for most counties. This was related to the finding that the stability of individual species, i.e., barley and wheat, was more effective in reducing the volatility of combined yields than the asynchronous variation in annual yields among crops. Future studies might include an increased number of crop species and consider temporal diversification effects for a more realistic assessment of the relation between yield volatility and crop diversity and test the relationship in other regions and production conditions.
... Due to its high importance, vegetation phenology has been the focus of interest in various research domains. This concerns crop-related applications in particular, where accurate estimates of crop physiological growth phases or phenophases are crucial for many agronomic practices, such as irrigation and fertilization scheduling, agricultural weather or biodiversity index derivation, soil erosion, crop condition monitoring, and yield estimation or integrated pest management [2][3][4][5][6][7][8][9]. Crop phenophases were traditionally recorded by manual field observations and surveys, e.g., [10]. ...
Operational crop monitoring applications, including crop type mapping, condition monitoring, and yield estimation, would benefit from the ability to robustly detect and map crop phenology measures related to the crop calendar and management activities like emergence, stem elongation, and harvest timing. However, this has proven to be challenging due to two main issues: first, the lack of optimised approaches for accurate crop phenology retrievals, and second, the cloud cover during the crop growth period, which hampers the use of optical data. Hence, in the current study, we outline a novel calibration procedure that optimises the settings to produce high-quality NDVI time series as well as the thresholds for retrieving the start of the season (SOS) and end of the season (EOS) of different crops, making them more comparable and related to ground crop phenological measures. As a first step, we introduce a new method, termed UE-WS, to reconstruct high-quality NDVI time series data by integrating a robust upper envelope detection technique with the Whittaker smoothing filter. The experimental results demonstrate that the new method can achieve satisfactory performance in reducing noise in the original NDVI time series and producing high-quality NDVI profiles. As a second step, a threshold optimisation approach was carried out for each phenophase of three crops (winter wheat, corn, and sugarbeet) using an optimisation framework, primarily leveraging the state-of-the-art hyperparameter optimization method (Optuna) by first narrowing down the search space for the threshold parameter and then applying a grid search to pinpoint the optimal value within this refined range. This process focused on minimising the error between the satellite-derived and observed days of the year (DOY) based on data from the German Meteorological Service (DWD) covering two years (2019–2020) and three federal states in Germany. The results of the calculation of the median of the temporal difference between the DOY observations of DWD phenology held out from a separate year (2021) and those derived from satellite data reveal that it typically ranged within ±10 days for almost all phenological phases. The validation results of the detection of dates of phenological phases against separate field-based phenological observations resulted in an RMSE of less than 10 days and an R-squared value of approximately 0.9 or greater. The findings demonstrate how optimising the thresholds required for deriving crop-specific phenophases using high-quality NDVI time series data could produce timely and spatially explicit phenological information at the field and crop levels.
... For example, Bönecke et al. (2020) and Mäkinen et al. (2018) used linear mixed methods to quantify the contribution of agro-climatic extremes to wheat yield in Europe. They found that the contribution of heat and drought stresses varied across phenological phases and regions, further supported by findings from Riedesel et al. (2023) and Schmitt et al. (2022). Additionally, Lobell et al. (2015) used a modified APSIM (Agricultural Production Systems Simulator) model to quantify the influence of drought and heat stresses on crop production in northeast Australia. ...
Drought is projected to intensify under warming climate and will continuously threaten global food security. Assessing the risk of yield loss due to drought is key to developing effective agronomic options for farmers and policymakers. However, little has been known about determining the likelihood of reduced crop yield under different drought conditions and defining thresholds that trigger yield loss at the regional scale in Australia. Here, we estimated the dependence of yield variation on drought conditions and identified drought thresholds for 12 Australia’s key wheat producing regions with historical yield data by developing bivariate models based on copula functions. These identified drought thresholds were used to investigate drought statistics under climate change with an ensemble of 36 climate models from Coupled Model Intercomparison Project Phase 6 (CMIP6). We found that drought-induced yield loss was region-specific. The drought thresholds leading to the same magnitude of wheat yield reduction were smaller in regions of southern Queensland and larger in Western Australia mainly due to different climate and soil conditions. Drought will be more frequent and affect larger areas under future warming climates. Based on our results, we advocate for more effective crop management options, particularly in regions where wheat yield is vulnerable to drought in Australia. This will mitigate potential drought impacts on crop production and safeguard global food security.
... Despite existing efforts on crop damage assessment, the potential of RS to assist in timely attribution of crop damage to weather extremes remains under-explored. Previous studies on attribution analysis primarily utilize data-driven approaches employing phenologically and spatially dynamic extreme weather indicators (Lüttger and Feike, 2018;Riedesel et al., 2023;Schmitt et al., 2022). These approaches explain yield losses by establishing statistical relationships between extreme weather indicators and end-of-season yield losses. ...
... Hence, covariate time series based on S2 ARD imagery were temporally downsampled to regular time series by computing mean values over 7-day intervals with a total length of 22 7-day intervals, consistently covering a period, ranging from 1st of March to 31st of July for each included year. By focusing on this period, RS-based covariates capture vegetation signals of targeted winter crops during their transition from vegetative to reproductive and maturity phases of phenological development, where several factors, such as drought and heat stress, impact yield formation (Hlaváčová et al. 2018;Riedesel et al. 2023). Remaining gaps in covariate time series were imputed, applying a linear interpolation approach, using the Python package "pandas" (The pandas development team 2023). ...
... Spatially explicit time series datasets based on interpolated observations of phenological events are available for Germany. These datasets have been used to characterize phenological windows (Möller, Boutarfa, and Strassemeyer 2020), to calculate phenological and phase-specific weather indices (Möller et al. 2019), and to disentangle extreme weather effects on crop yields (Bucheli, Dalhaus, and Finger 2022;Riedesel et al. 2023Riedesel et al. , 2024. ...
Detailed and accurate statistics on crop productivity are key to inform decision-making related to sustainable food production and supply ensuring global food security. However, annual and high-resolution crop yield data provided by official agricultural statistics are generally lacking. Earth observation (EO) imagery, geodata on meteorological and soil conditions, as well as advances in machine learning (ML) provide huge opportunities for model-based crop yield estimation in terms of covering large spatial scales with unprecedented granularity. This study proposes a novel yield estimation approach that is bottom-up scalable from parcel to administrative levels by leveraging ML-ensembles, comprising of six regression estimators (base estimators), and multi-source geodata, including EO imagery. To ensure the approach’s robustness, two ensemble learning techniques are investigated, namely meta-learning through model stacking and majority voting. ML-ensembles were evaluated multi-annually and crop-specifically for three major winter crops, namely winter wheat (WW), winter barley (WB), and winter rapeseed (WR) in two German federal states, covering 140,000 to 155,000 parcels per year. ML-ensembles were evaluated at the parcel and district level for two German federal states against official yield reports, ranging from 2019 to 2022, based on metrics such as coefficient of determination (RSQ) and normalized root mean square error (nRMSE). Overall, the most robustly performing ensemble learning technique was majority voting yielding RSQ and nRMSE values of 0.74, 13.4% for WW, 0.68, 16.9% for WB, and 0.66, 14.1% for WR, respectively, through cross-validation at parcel level. At the district level, majority voting reached RSQ and nRMSE ranges of 0.79–0.89, 7.2–8.1% for WW, 0.80–0.84, 6.0–9.9% for WB, and 0.60–0.78, 6.1–10.4% for WR, respectively. Capitalizing on ensemble learning-based majority voting, examples of unprecedented high-resolution crop yield maps at spatial resolution are presented. Implementing a scalable yield estimation approach, as proposed in this study, into crop yield reporting frameworks of public authorities mandated to provide official agricultural statistics would increase the spatial resolution of annually reported yields, eventually covering the entire cropland available. Such unprecedented data products delivered through map services may improve decision-making support for a variety of stakeholders across different spatial scales, ranging from parcel to higher administrative levels.
... Исследование устойчивости к неблагоприятным факторам среды, вызывающим нарушение водного режима, -одна из центральных проблем современных наук о растениях. Как известно, засуха относится к наиболее распространенным стрессам, приводящим к нарушению всех звеньев метаболизма и существенной потере урожая у культурных растений, в том числе у пшеницы (1)(2)(3). Наиболее критичная фаза в развитии пшеницы -ранний этап онтогенеза. В этот период растения особенно чувствительны к действию неблагоприятных факторов и дефицит влаги вызывает снижение всхожести семян, ингибирование их прорастания и роста всходов, а также отставание культуры в развитии, что, в свою очередь, приводит к значительному снижению продуктивности (4,5). ...
... Исследование устойчивости к неблагоприятным факторам среды, вызывающим нарушение водного режима, -одна из центральных проблем современных наук о растениях. Как известно, засуха относится к наиболее распространенным стрессам, приводящим к нарушению всех звеньев метаболизма и существенной потере урожая у культурных растений, в том числе у пшеницы (1)(2)(3). Наиболее критичная фаза в развитии пшеницы -ранний этап онтогенеза. В этот период растения особенно чувствительны к действию неблагоприятных факторов и дефицит влаги вызывает снижение всхожести семян, ингибирование их прорастания и роста всходов, а также отставание культуры в развитии, что, в свою очередь, приводит к значительному снижению продуктивности (4,5). ...
... Crop yields and the respective yield formation processes are complex and influenced by a combination of genetic (G) and management (M) factors, as well as the local environmental conditions (E) in which the crops are grown [1,2]. As global warming continues, heat and drought stress are increasingly affecting yields and their variability globally [3][4][5][6], in Europe [7][8][9], and in Germany [10][11][12]. ...
... While their genetic differences and their plant physiological reaction to heat and drought stress have been well studied for wheat and rye in greenhouse experiments [19,21], there is still high uncertainty regarding the site-specific effects of adverse weather on wheat and rye yields [22][23][24][25]. Different studies found diverging yield responses to heat and drought stress in different regions [7,11]. ...
... Therefore, high-resolution G × E × M data are essential to disentangle influencing factors and analyze site-specific weather effects on crop yields [11]. In addition, crop-specific phenology data are a fundamental prerequisite to comprehend these confounding factors and achieve a thorough understanding of local weather effects on crops [26][27][28]. ...
Heat and drought are major abiotic stressors threatening cereal yields, but little is known regarding the spatio-temporal development of their yield-effects. In this study, we assess genotype (G) × environment (E) × management (M) specific weather-yield relations utilizing spatially explicit weather indices (WIs) and variety trial yield data of winter wheat (Triticum aestivum) and winter rye (Secale cereale) for all German cereal growing regions and the period 1993-2021. The objectives of this study are to determine the explanatory power of different heat and drought WIs in wheat and rye, to quantify their site-specific yield effects, and to examine the development of stress tolerance from old to new varieties. We use mixed linear models with G × E × M specific covariates as fixed and random factors. We find for both crops that combined heat and drought WIs have the strongest explanatory power during the reproductive phase. Furthermore, our results strongly emphasize the importance of site conditions regarding climate resilience, where poor sites reveal two to three times higher yield losses than sites with high soil quality and high annual precipitation in both crops. Finally, our analysis reveals significantly higher stress-induced absolute yield losses in modern vs. older varieties for both crops, while relative losses also significantly increased in wheat but did not change in rye. Our findings highlight the importance of site conditions and the value of high-yielding locations for global food security. They further underscore the need to integrate site-specific considerations more effectively into agricultural strategies and breeding programs.
