Within-season crop yield prediction by a multi-model ensemble with integrated data assimilation

Field Crops Research, Journal Year: 2024, Volume and Issue: 308, P. 109293 - 109293

Published: Feb. 6, 2024

Improving crop yield prediction accuracy is crucial for sustainable agriculture. One approach to use data assimilation (DA) techniques based on satellite remote sensing, which can help improve predictions at the regional national scale. However, interaction between uncertain model inputs and DA, as well impact of structure DA results, have received little attention date. In this work, we assimilated leaf area index (LAI) into three single models (CERES, GECROS, SPASS) their multi-model ensemble (MME) using a particle filtering (PF) algorithm. Mimicking common lack information large scale, considered nitrogen fertilization, sowing date, soil hydraulic parameters, weather sources uncertainties. case study, applied setup six winter wheat site years in southwestern Germany. Before applying all were calibrated validated in-situ measured from multi-site, multi-year independent set. The performance calibration was used assign weights MME. Results show that parameters had highest predictions. substantially improved precision LAI simulation models. Moreover, enhanced grain by SPASS, ensemble, but no considerable effect CERES. Specifically, bias decreased 25% 15% 26% 19% 7% contrast, even without error CERES below 5%. correlation errors key factor indicating how be effective specific model. When analysis unavailable, promising assimilation. Further investigations calibration, input uncertainty, MME size, weighting scheme are necessary applications.

Language: Английский

---

Do rotations with cover crops increase yield and soil organic carbon?—A modeling study in southwest Germany

Agriculture Ecosystems & Environment, Journal Year: 2024, Volume and Issue: 375, P. 109167 - 109167

Published: July 29, 2024

Conservation agriculture practices of crop rotation with permanent soil cover have been widely promoted for improving long-term agroecosystem resilience in the face changing climate. However, there has no comprehensive evaluation site-specific services health and yield response to improved rotations without crops (CCs) on field spatial scales. We calibrated applied a process-based agroecosystems model determine effects cropping organic N content mineralization rate, carbon (SOC) change CO2 efflux, yields. A 10-year systems dataset from six sites southwest Germany was used calibrate evaluate DSSAT provide typical management conventional farming system region as business-as-usual (BAU) scenario application. 4-year then designed inclusion commonly grown non-legume legume CCs three cycles at research surrounding region. Crop treatments provided no-CC scenario, therefore effect CC could be tested. Relative BAU no-CC, annual resulted 12% 3% higher 6% 8% SOC respectively. Additional advantage C more pronounced by while were efficient reducing leaching. Combined positive rotational observed winter wheat oilseed rape yields sites. we variability these results regional scale, suggesting environment interactions that should considered recommendations. significantly increased water productivity cereal crops, but did not produce spring barley or silage maize compared unless only certain areas are vulnerable losses. Our findings highlight sequestration potential emphasizing need agronomically environmentally sound systems.

Language: Английский

---

Improving winter wheat yield prediction by accounting for weather and model parameter uncertainty while assimilating LAI and updating weather data within a crop model

European Journal of Agronomy, Journal Year: 2024, Volume and Issue: 156, P. 127149 - 127149

Published: March 11, 2024

Accurate crop yield predictions play a crucial role in enabling informed policy-making to ensure food security. Beyond using advanced methods such as remote sensing and data assimilation (DA), it is essential comprehend the influence of various sources uncertainty on overall prediction uncertainty. This study presents novel approach for enhancing accuracy by assimilating remotely-sensed Leaf Area Index (LAI) updating weather ensemble into model (SPASS) while accounting calibration In addition, we investigated effect prior DA four type scenarios. These scenarios involve calibrating different combinations yield, phenology, LAI, ranging from minimum (yield only) maximum (yield, LAI) availability. To address uncertainty, derived forecasts downscaled climate models utilizing MarkSim generator. Our results demonstrate that LAI significantly reduces predictions. Notably, associated with ensembles has more substantial compared resulting calibration. finding highlights significance variations discrepancies when assessing Additionally, given set SPASS parameters used winter wheat calibration, additional field-based does not improve quality.

Language: Английский

---

Enhancing Hectare‐Scale Groundwater Recharge Estimation by Integrating Data From Cosmic‐Ray Neutron Sensing Into Soil Hydrological Modeling

Water Resources Research, Journal Year: 2025, Volume and Issue: 61(1)

Published: Jan. 1, 2025

Abstract Vadose zone models, calibrated with state variables, may offer a robust approach for deriving groundwater recharge. Cosmic‐ray neutron sensing (CRNS) provides soil moisture over large support volume (horizontal extent of hectares) and offers the opportunity to estimate water fluxes at this scale. However, horizontal vertical sensitivity method results in an inherently weighted content, which poses challenge its application hydrologic modeling. We systematically assess calibrating hydraulic model HYDRUS 1D cropped field site. Calibration was performed using different field‐scale time series ability represent root derive recharge assessed. As our benchmark, we used distributed point sensor network from within footprint CRNS. Models on CRNS data or combinations deeper measurements resulted cumulative comparable benchmark. While models based exclusively do not dynamics adequately, combining profile overcomes limitation. also perform well timing downward flux compared independent tension measurements. latter quantitative estimates spanning wide range values, including unrealistic highs exceeding local annual precipitation. Conversely, modeled ranging between 30% 40%

Language: Английский

---

Facilitating Effective Reuse of Soil Research Data: The BonaRes Repository

European Journal of Soil Science, Journal Year: 2025, Volume and Issue: 76(2)

Published: March 1, 2025

ABSTRACT Soil plays a paramount role in addressing complex challenges related to climate change, the agri‐food system, and ecosystem services. This importance makes soil research data highly relevant for meta‐analysis, synthesis, modelling, assessment. As data‐intensive techniques proliferate studying global change impacts on agricultural systems, effective management reuse are essential. Repositories that adhere FAIR (Findability, Accessibility, Interoperability, Reusability) principles crucial maximizing value efficiency of data. While publishing an Open Access repository is necessary reusability, it alone not sufficient. Specialized repositories enhance potential by discipline‐specific needs through targeted metadata technical frameworks. The BonaRes Repository was developed guided principles, with focus reusability. Here, we introduce repository's infrastructures services, including specialized tools quality assurance profile as well long‐term field experiment We emphasize ability these services promote publication specifically sciences. review examples reuse, highlighting their scientific contributions understanding systems. Finally, discuss remaining achieving open From 2018 date, has facilitated 815 publications; 62 papers have reused published Reuse applications range widely—from extracting study site or environmental covariates reanalysing (meta)data light new questions, developing scenarios conducting model calibration evaluation. A key insight from our researchers frequently apply advance method development. Initiatives such reciprocal harvesting integration into larger national international infrastructure will further expand scope data, broader agrosystems science.

Language: Английский

---

The land–atmosphere feedback observatory: a new observational approach for characterizing land–atmosphere feedback

Geoscientific instrumentation, methods and data systems, Journal Year: 2023, Volume and Issue: 12(1), P. 25 - 44

Published: Jan. 25, 2023

Abstract. Important topics in land–atmosphere (L–A) feedback research are water and energy balances heterogeneities of fluxes at the land surface atmospheric boundary layer (ABL). To target these questions, Land–Atmosphere Feedback Observatory (LAFO) has been installed southwestern Germany. The instrumentation allows comprehensive high-resolution measurements from bedrock to lower free troposphere. Grouped into three components, atmosphere, soil surface, vegetation, LAFO observation strategy aims for simultaneous all compartments. For this purpose sensor synergy contains lidar systems measure key variables humidity, temperature wind. At eddy covariance stations operated record distribution radiation, sensible, latent ground heat fluxes. Together with a network, content monitored agricultural investigation area. As crop height, leaf area index phenological growth stage values registered. observations organized operational intensive periods (IOPs). Operational aim long time series datasets investigate statistics, we present as an example correlation between mixing height potential IOPs is demonstrated 24 h case study using dynamic thermodynamic profiles that uses scanning differential absorption relate humidity patterns structures. Both long-term will provide new insight exchange processes their statistics improving representation L–A feedbacks climate numerical weather prediction models. component particular support coupling atmosphere.

Language: Английский

---

Harvest residues: A relevant term in the carbon balance of croplands?

Agricultural and Forest Meteorology, Journal Year: 2024, Volume and Issue: 349, P. 109935 - 109935

Published: March 2, 2024

Over the past two decades, major efforts have been made to quantify extent which and under what conditions croplands are sources or sinks for carbon. For this purpose, net carbon stock change of study site is typically quantified based on CO2 fluxes monitored with an eddy covariance chamber system, measured C import by organic fertilizer export harvest. While in cropland studies balance usually referred as biome productivity (NBP), we prefer use term ecosystem (NECB) here. NECB basically sum plant (ΔPC) soil (ΔSOC). In standard approach, assumption that at annual bulk biomass removed harvest, ΔPC can therefore be neglected. case, ΔSOC equals NECB. paper show problematic, particularly if crop rotation systems budget determined over a single cropping period. The present contribution extends concept include harvest residues (HR) applies it case maize - winter wheat southwest Germany. all three periods, sign was opposite ΔSOC. Accordingly, neglecting HR led incorrect result concerning question whether sink source Our findings demonstrate must included obtain accurate meaningful balance.

Language: Английский

---

CLIM4OMICS: a geospatially comprehensive climate and multi-OMICS database for maize phenotype predictability in the United States and Canada

Earth system science data, Journal Year: 2023, Volume and Issue: 15(9), P. 3963 - 3990

Published: Sept. 6, 2023

Abstract. The performance of numerical, statistical, and data-driven diagnostic predictive crop production modeling relies heavily on data quality for input calibration or validation processes. This study presents a comprehensive database the analytics used to consolidate it as homogeneous, consistent, multidimensional genotype, phenotypic, environmental maize phenotype modeling, diagnostics, prediction. are obtained from Genomes Fields (G2F) initiative, which provides multiyear genomic (G), (E), phenotypic (P) datasets that can be train test growth models understand genotype by environment (GxE) interaction phenomenon. A particular advantage G2F is its diverse set DNA sequences (G2F-G), measurements (G2F-P), station-based time series (mainly climatic data) observations collected during maize-growing season (G2F-E), metadata each field trial (G2F-M) across United States (US), province Ontario in Canada, state Lower Saxony Germany. construction this climate incorporates control (QC) consistency (CC) digital representation geospatially distributed required GxE interaction. two-phase QC–CC preprocessing algorithm also includes module estimate uncertainties. Generally, pipeline collects raw files, checks their formats, corrects structures, identifies cures imputes missing data. uses machine-learning techniques fill gaps, quantifies uncertainty introduced using other sources gap imputation G2F-E, discards values G2F-P, removes rare variants G2F-G. Finally, an integrated enhanced was generated. improving improved called Climate OMICS (CLIM4OMICS) follow findability, accessibility, interoperability, reusability (FAIR) principles, all codes available at https://doi.org/10.5281/zenodo.8002909 (Aslam et al., 2023a) https://doi.org/10.5281/zenodo.8161662 2023b), respectively.

Language: Английский

---

A Bayesian sequential updating approach to predict phenology of silage maize

Biogeosciences, Journal Year: 2022, Volume and Issue: 19(8), P. 2187 - 2209

Published: April 22, 2022

Abstract. Crop models are tools used for predicting year-to-year crop development on field to regional scales. However, robust predictions hampered by uncertainty in model parameters and the data calibration. Bayesian calibration allows estimation of quantification uncertainties, with consideration prior information. In this study, we a sequential updating (BSU) approach progressively incorporate additional at yearly time-step order calibrate phenology (SPASS) while analysing changes parameter prediction quality. We measurements silage maize grown between 2010 2016 regions Kraichgau Swabian Alb southwestern Germany. Parameter errors were expected be reduced final, irreducible value. was as updates. For two sequences using synthetic data, one which able accurately simulate observations, other single cultivar under same environmental conditions, error mostly reduced. true that followed actual chronological cultivation farmers regions, increased when not representative validation data. This could explained differences ripening group temperature conditions during vegetative growth. With implications manual automatic streams updating, our study highlights success methods depends comprehensive understanding inherent structure observation limitations.

Language: Английский

---

The Land-Atmosphere Feedback Observatory: A New Observational Approach for Characterizing Land-Atmosphere Feedback

Published: July 14, 2022

Abstract. Important topics in Land-Atmosphere (L-A) feedback research are water and energy balances heterogeneities of fluxes at the land-surface atmospheric booundary layer. To target these questions, Feedback Observatory (LAFO) has been installed Southwest Germany. The instrumentation allows comprehensive high-resolution measurements from bedrock to lower free troposphere. Grouped three components: atmosphere, soil vegetation, LAFO observation strategy aims for simultaneous all compartments. For that sensor synergy contains lidar systems measure key variables humidity, temperature wind. At eddy covariance stations operated record distribution radiation, sensible, latent ground heat fluxes. With a network is monitored agricultural investigation area. observations organized operational intensive periods (IOPs). Operational aim long timeseries dataset investigate statistics as we present example correlation between mixing layer height surface potential IOPs demonstrated with 24 hour case study dynamic thermodynamic profiles well scanning differential absorption relate humidity patterns structures. Both long-term important improving representation L-A feedbacks climate numerical weather prediction models.

Language: Английский

---