Enhancing agricultural productivity requires a thorough assessment of fungal-induced crop diseases under varying climatic conditions. We implemented infection risk response functions into a coupled land surface-crop model (Noah-MP-Gecros) for both winter wheat and maize. For wheat, the infection risk module covers Septoria tritici blotch (Zymoseptoria tritici), brown rust (Puccinia triticina), yellow rust (Puccinia striiformis), and Fusarium head blight (Fusarium spp., mainly F. graminearum). For maize, the focus is on Fusarium ear and stalk rot (primarily Fusarium verticillioides) and Gibberella ear rot (Fusarium graminearum). We evaluated model performance by simulating infection risks for major wheat and maize diseases across three growing between 2020 and 2023, analyzing infection dynamics and the sensitivity of crop-specific parameters. Model outputs were partially evaluated against field observations. Simulated pathogen infection varied significantly from year to year within the same region. Rust diseases in wheat and Fusarium ear rot and stalk rot in maize displayed similar interannual trends, while Fusarium head blight in wheat, and Gibberella ear rot in maize exhibited divergent patterns across growing seasons. The maximum amount of water intercepted by the canopy and the land-surface parameter momentum roughness length were identified as the most sensitive crop parameters for infection risk. The model, which relies exclusively on meteorological variables, has shown potential for real-time disease risk prediction and guiding pesticide application. The model performance could be improved by integrating biological processes such as host-pathogen interactions and spore availability. On the crop side, pathogen impacts could potentially be mitigated by management and cultivar traits influencing canopy water interception and aerodynamic roughness.

Quantifying the spatial extent of the rhizosphere is essential for understanding root-soil interactions and assessing the impact of rhizosphere processes. Although methods for determining the spatial distribution of rhizosphere compounds (e.g., nutrients, carbon, microbes, enzymes) are increasingly available, the mathematical approaches used to delineate the boundary between the rhizosphere and bulk soil often introduce systematic differences and uncertainties, hindering cross-study comparability. We introduce the integral rhizosphere extent (IRE) method which derives the spatial extent of the rhizosphere from a defined fraction of the total integral over a radial concentration profile. We tested this method against two threshold-based methods that use bulk soil data to set the rhizosphere endpoint. To assess their applicability, we compared approach-specific estimates from imaging data of organic N, beta-glucosidase (BG) enzyme activity, as well as from modeling results on dissolved organic carbon and microbial biomass profiles. Our analysis revealed that the threshold-based methods produce highly variable or even no estimates of the rhizosphere extent, particularly in low-gradient concentration profiles. In contrast, the IRE method produced estimates that are robust to variability in bulk-soil concentrations used to define thresholds while remaining responsive to systematic differences among concentration profiles across datasets. A sensitivity analysis showed that the IRE method remains stable across a range of predefined fractions of the total integrated solute enrichment or depletion around the root. Its mathematical rigor and ease of application make it a strong candidate for harmonizing rhizosphere data evaluation and improving the comparability of root-soil interaction studies. Moreover, it can naturally accommodate three-dimensional spatial data by integrating solute enrichment within a defined soil volume, rather than relying on point-based estimations derived from local concentration thresholds.

A plant's development is strongly linked to the water and carbon (C) flows in the soil-plant-atmosphere continuum. Ongoing climate shifts will alter the water and C cycles and affect plant phenotypes. Comprehensive models that simulate mechanistically and dynamically the feedback loops between water and C fluxes in the soil-plant system are useful tools to evaluate the sustainability of genotype-environment-management combinations that do not yet exist. In this study, we present the equations and implementation of a rhizosphere-soil model within the CPlantBox framework, a functional-structural plant model that represents plant processes and plant-soil interactions. The multi-scale plant-rhizosphere-soil coupling scheme previously used for CPlantBox was likewise updated, among others to increase the accuracy and stability of the model outputs. The model was implemented to simulate the effect of dry spells occurring at different plant development stages, and for different soil kinetic parameterisations of microbial dynamics in soil. We could observe diverging results according to the date of occurrence of the dry spells and soil parameterisations. For instance, earlier dry spells (from 11th to the 18th day of growth) led to a lower cumulative plant C release, while later dry spells (from 18th to the 25th day of growth) led to higher C input to the soil. For more reactive microbial communities (higher maximum C uptake rate and (de)activation rates), this higher C input caused a strong increase in CO2 emissions. For the same weather scenario, we observed lower microbial CO2 emissions with less reactive communities. This model can be used to gain insight into C and water flows at the plant scale, and the influence of soil-plant interactions on C cycling in soils.

Background The SuedLink was planned as a high-voltage direct-current underground cable route. The construction itself is a massive intervention on soil, and during operation, the power cables emit up to 32 W of heat per cable meter at maximum load.Aims The study aimed to investigate the impact of the construction and the heat emission of the SuedLink on soils typical of Southwest Germany.Methods Four test sites, each covering approximately 0.6 ha, were established following best practice along the planned SuedLink route in 2021. Each test site was set up in a fully randomized split-plot design. Heating pipes were installed at a depth of 1.5 m with a target heat emission of 32 W per meter of cable. Each test site was extensively instrumented with temperature, heat flux, and soil moisture sensors.Results Temperatures 5 cm above the heating pipes at a depth of 1.25 m were 14-16 degrees C higher than the control. In the heated plots, soil temperatures in the center of the plowing horizon at 0.15 m depth were 1-3 degrees C higher than in the control soil. The strip affected by the heat emission had a width of about 10 m. Negative impacts on bulk density, plant-available water, and air capacity were not observed.Conclusions During regular operation at a load <= 75%, the average increase in temperature in the plowing horizon at a depth of 0.15 m is expected to be less than 1.6 degrees C in the soils along the SuedLink route in Southwest Germany.

Understanding the feedback mechanisms between roots and soil, and their effects on microbial communities, is crucial for predicting carbon cycling processes in agroecosystems. Process-based modeling is a valuable tool for quantifying biogeochemical processes and identifying regulatory mechanisms in the rhizosphere. A novel one-dimensional axisymmetric rhizosphere model is used to simulate the spatially resolved dynamics of microorganisms and soil organic matter turnover around a single root segment. The model accounts for two functional groups with different life history strategies (copiotrophs and oligotrophs), reflecting trade-offs in functional microbial traits related to substrate utilization and microbial metabolism. It considers differences in the accessibility of soil organic matter by including the microbial utilization of low and high molecular weight organic carbon compounds (LMW-OC, HMW-OC). The model was conditioned using Bayesian inference with constraint-based parameter sampling, which enabled the identification of parameter sets resulting in plausible model predictions in agreement with experimental evidence.Mimicking the behavior of growing roots, the model assumed 15 days of rhizodeposition for LMW-OC. The simulations show a decreasing pattern of dissolved LMW-OC away from the root surface. We observed a dominance of copiotrophs close to the root surface (0-0.1 mm). Spatial patterns of functional microbial groups persisted after rhizodeposition ended, indicating a legacy effect of rhizodeposition on microbial communities, particularly on oligotrophic activity. Simulated microbial biomass exhibits a very rapid change within 0-0.2 mm away from the root surface, which points to the importance of resolving soil properties and states at sub-millimeter resolution. Microbial-explicit rhizosphere modeling thus facilitates elucidating spatiotemporal patterns of microorganisms and carbon turnover in the rhizosphere. The identified legacy effect of rhizodeposition on soil microorganisms might be leveraged for rhizosphere-based carbon stabilization strategies in agroecosystems.

To assess the impact of agricultural practices on water and carbon cycles within specific Genome-Environment-Management combinations, understanding the interactions across the Soil-Plant-Atmosphere continuum (SPAC) is crucial.Indeed, soil water conditions influence carbon concentration and transport, impacting soil carbon physical and biochemical reactions.The soil water and carbon status affect, in turn, the plant water and carbon dynamics directly via the plant-to-soil water or carbon gradient, and indirectly via plant water status, influencing its inner balance of water (uptake, transpiration and flow) and carbon (assimilation, usage for maintenance and growth, storage, respiration, rhizodeposition, and transport).Reciprocally, plant water and carbon balances affect the soil carbon cycle in the short term through root water uptake and rhizodeposition. Those rhizodeposits are, for the most part, made of exudates and mucilage. Root exudates are low molecular weight organic compounds that are mainly passively diffused, while mucilage is a fluid made of polymers with high molecular weight created from starch via an active process.Modelling plant and soil water and carbon processes, along with their interactions, can help to understand better and represent the effects of the underlying feedback loops. In this study, we therefore coupled the Functional Structural Plant Model (FSPM) CPlantBox with the rhizosphere model TraiRhizo, implemented using the porous medium flow and transport solver DuMux.The overall coupled model and multiscale framework includes a module of 3D plant architecture development, and modules to represent flow and transport within the plant, the soil and the perirhizal zone around each root segment. Flows between compartments are solved implicitly via fixed-point iteration, using parallel computation for both the 3D soil and rhizosphere models.We present a case study in which we simulated the growth of a C3 monocot and observed how changes in soil water content, due to root water uptake, influenced dissolved carbon concentration and (de)activation of the soil microbial communities during a dry spell.In the future, the model will be applied to assess the impact of small dry spells at various stages of plant development against a baseline scenario. In time, this model could support plant breeding efforts to find root traits that aim for more drought-resistant plants in specific pedoclimatic environments.

Abstract. Carbon dioxide emissions from burning fossil fuels play a major role in driving global climate change. Reducing these emissions through innovative technologies is critical to achieve climate change mitigation goals. Methane pyrolysis, including catalytic and "plasmalytic" approaches, has attracted attention for its ability to produce so-called turquoise hydrogen alongside solid carbon as a by-product that could be reused as soil amendment. This study investigated the potential of solid carbon materials from catalytic pyrolysis and plasmalysis, alongside reference materials (biochar and graphite), to improve soil hydraulic properties and to reduce heavy metal mobility and whether ecotoxicological reactions affect soil organisms. In experiment 1, two arable soils of contrasting textures (sand and silty loam) were amended with these carbon materials at an application rate of 40 t ha-¹, followed by assessments of soil physical properties, soil respiration rate, microbial biomass, extractable organic carbon, nitrogen mineralization, the activity of soil macro- (earthworms) and mesofauna (springtails, Folsomia candida). In Experiment 2, we evaluated heavy metal mobility and availability in metal-contaminated soils. In uncontaminated soil, solid carbon from plasmalysis (SCplas) increased water retention of the silty loam, particularly in the range of pF 1.8–3.0, but had no effect in the sandy soil, likely due to its hydrophobic properties, which limited moisture retention. In the silty loam, SCplas reduced microbial activity and the abundance of springtails. In the sandy soil, it had a negative effect on soil macrofauna (earthworms). The solid carbon from catalytic pyrolysis (SCcat) had almost no effect on the biological properties studied. In soils contaminated with heavy metals, SCplas showed strong immobilisation of heavy metals particularly for Cd and Cu, across several sites, outperforming the reference materials. However, SCcat increased Cd mobility at some sites, indicating little or even adverse effects on heavy metal mobility. Our results highlight the promise of SCplas for site-specific soil improvement, while cautioning against its hydrophobic effects in sandy soils. In contrast, SCcat has more negative effects, especially ecotoxicological than positive ones depending on soil.

The planthopper Pentastiridius leporinus, commonly called reed glass-winged cicada, transmits the pathogens “Candidatus Arsenophonus phytopathogenicus” and “Candidatus Phytoplasma solani”, which are infesting sugar beet and, most recently, also potato in the Upper Rhine valley area of Germany. They cause the “Syndrome Basses Richesses” associated with reduced yield and sugar content in sugar beet, leading to substantial monetary losses to farmers in the region. No effective solutions exist currently. This study uses statistical models to understand to what extent the abundance of cicadas depends on climate regions during the vegetation period (April–October). We further investigated what influence temperature and precipitation have on the abundance of the cicadas in sugar beet fields. Furthermore, we investigated the possible impacts of future climate on cicada abundance. Also, 22 °C and 8 mm/day were found to be the optimal temperature and precipitation conditions for peak male cicada flight activity, while 28 °C and 8 mm/day were the optimum for females. By the end of the 21st century, daily male cicada abundance is projected to increase significantly under the worst-case high greenhouse gas emission scenario RCP8.5 (RCP-Representative Concentration Pathways), with confidence intervals suggesting a possible 5–15-fold increase compared to current levels. In contrast, under the low-emission scenario RCP2.6, male cicada populations are projected to be 60–70% lower than RCP8.5. An understanding of the influence of changing temperature and precipitation conditions is crucial for predicting the spread of this pest to different regions of Germany and other European countries.

The importance of evapotranspiration (ET) fluxes for the terrestrial water cycle is demonstrated by an overwhelming body of literature. Unfortunately, errors in their measurement contribute significantly to (model) uncertainties in quantifying and understanding ecohydrological systems. Measurements of surface-atmosphere fluxes of water at the ecosystem scale, the eddy covariance method can be considered a powerful technique and considered an important tool to validate ET models. Spatially averaged fluxes of several hundred square meters may be obtained. While the eddy-covariance technique has become a routine method to estimate the turbulent energy fluxes at the soil-atmosphere boundary, it remains not error free. Some of the inherent errors are quantifiable and may be partitioned into systematic and stochastic errors. For model-data comparison, the nature of the measurement error needs to be known to derive knowledge about model adequacy. To this end, we compare several assumptions found in the literature to describe the statistical properties of the error with newly derived descriptions, in this study. We are able to show, how sensitive the assumptions about the error are on the model selection process. We demonstrate this by comparing daily agro-ecosystem ET fluxes simulated with the detailed agro-hydrological model Expert-N to data gathered using the eddy-covariance technique.

Understanding the turnover and stabilization of soil organic matter (SOM) is the key to highly productive agriculture and to climate change mitigation. Understanding and predicting biogeochemical matter and energy flows in soil systems is challenging due to persisting knowledge gaps regarding biological and energetic controls of SOM turnover and limited knowledge integration of informative data with process-based models. We present modeling concepts of microbially explicit soil organic matter models and shed light on integrating trait-based ecological frameworks in process-based models and the use of advanced data-model fusion approaches. We highlight some insights from applying process-based models and challenges we need to address to acquire robust predictions.

Studies of land-atmosphere (L-A) feedbacks are essential for understanding the Earth system. These feedbacks are the result of an interaction of processes related to exchanges of momentum, energy, and mass in the soil-vegetation-surface layer (SL)-atmospheric boundary layer (ABL) continuum. Quantification of feedbacks are often made using L-A feedback metrics. Inaccurate representation/parameterization of feedbacks are a weakness of current weather models, and their improvement will thus contribute to better simulations over all spatiotemporal scales. Improving feedback representation requires simultaneous measurements in all L-A compartments using a synergy of in-situ and active remote sensing instruments. To that end, a new Land-Atmosphere Feedback Observatory (LAFO) was established at the University of Hohenheim, Stuttgart, Germany funded by the Carl Zeiss Foundation. It was developed as a prototype for a future network of GEWEX LAFOs (GLAFOs), proposed by the Global Energy and Water Exchanges (GEWEX) program and GEWEX Global Land/Atmosphere System Study (GLASS) panel (Wulfmeyer et al. 2020). The main goals are to:1) investigate the diurnal cycle and statistics of ABL temperature, humidity and wind profiles,2) characterize L-A feedback by suitable metrics.3) improve parameterizations of vegetation, surface and ABL fluxes,4) verify mesoscale and turbulence permitting models,LAFO brings together a sensor synergy with fine spatiotemporal resolution. An extended set of soil physical, plant dynamic as well as meteorological variables throughout the ABL are measured, focusing on evapotranspiration and other exchanges over agricultural landscapes. The LAFO observations with current instruments are continuously archived, according to FAIR data principles (Findable, Accessible, Interoperable, Reusable) and are complemented by additional field campaign measurements.The first key component of the current LAFO sensor synergy consists of four 3D scanning lidar systems: A scanning water vapor Differential Absorption Lidar (DIAL, Muppa et al. 2016, Späth et al. 2016) and the Atmospheric Rotational-Raman Temperature and Humidity Sounder (ARTHUS, Lange et al. 2019), both developed at the Institute of Physics and Meteorology. Both these systems are unique and provide water vapor and temperature profiles from the surface layer to the free troposphere with fine resolution down to turbulence scales (Behrendt et al. 2015, Wulfmeyer et al. 2015). These lidars are complemented by a scanning Doppler cloud radar and two Doppler lidars for measuring horizontal and vertical wind profiles and turbulent fluctuations. This combination allows determination of sensible and latent heat flux profiles. The second key component is a soil moisture and temperature sensor network distributed over agricultural land and two 10-m towers, measuring turbulent fluxes at two heights.LAFO will soon form part of a new Research Unit, funded by the German Research Foundation (DFG), called the Land-Atmosphere-Feedback-Initiative (LAFI) which begins in 2024, and incorporates novel crop, hydrology and atmospheric instruments, operated by several research partners within Germany. Here, we present measurement examples from the LAFO and show how these can be used to reach our research goals. ReferencesWulfmeyer et al. 2020, GEWEX Quarterly Vol. 30, No. 1.Behrendt et al. 2015, doi:10.5194/acp-15-5485-2015Wulfmeyer et al. 2015, doi:10.1002/2014RG000476Muppa et al. 2016, doi:10.1007/s10546-015-0078-9Späth et al. 2016, doi:10.5194/amt-9-1701-2016Lange et al. 2019, doi:10.1029/2019GL085774

Abstract. A plant's development is strongly linked to the water and carbon (C) flows in the soil-plant-atmosphere continuum. Ongoing climate shifts will alter the water and C cycles and affect plant phenotypes. Comprehensive models that simulate mechanistically and dynamically the feedback loops between water and C fluxes in the soil-plant system are useful tools to evaluate the sustainability of genotype-environment-management combinations that do not yet exist. In this study, we present the equations and implementation of a rhizosphere-soil model within the CPlantBox framework, a functional-structural plant model that represents plant processes and plant-soil interactions. The multi-scale plant-rhizosphere-soil coupling scheme previously used for CPlantBox was likewise updated, among others to include an implicit time-stepping. The model was implemented to simulate the effect of dry spells occurring at different plant development stages, and for different soil biokinetic parametrisations of microbial dynamics in soil. We could observe diverging results according to the date of occurrence of the dry spells and soil parametrisations. For instance, an earlier dry spell led to a lower cumulative plant C release, while later dry spells led to higher C input to the soil. For more reactive microbial communities, this higher C input caused a strong increase in CO2 emissions, while, for the same weather scenario, we observed a lasting stabilisation of soil C with less reactive communities. This model can be used to gain insight into C and water flows at the plant scale, and the influence of soil-plant interactions on C cycling in soil.

Climate change and other dynamic changes, such as demographic change, pose challenges for public water supply in Germany. This study contributes to the identification of hot spot regions that could experience increased water shortages in the future through nationwide, regionalized forecasts of water demand in domestic, industry and agriculture sectors and their balancing with projections of groundwater recharge. Multi-sectoral water demand forecasts for the periods 2021-2050, 2036-2065 and 2069-2098 were prepared using a top-down approach at NUTS-3 level (cities and districts). The total water demand in the domestic sector in Germany averages approx. 3.7 billion m³/a in the reference period 1998-2019. In the lower scenario, it decreases to approx. 2.2 billion m³/a by the end of the century. In the upper scenario, total water demand in the domestic sector in Germany increases to around 4.1 billion m³/a. Industrial water demand could fall to around half (10.9 billion m³/a) as early as 2030 compared to the reference period (approx. 21.6 billion m³/a) due to a sharp decline in cooling water demand. From the middle of the 21st century onwards, it is expected to stagnate at around 6.1 billion m³/a. Depending on the scenario, the irrigated agricultural area in Germany will almost double (RCP 2.6) or almost triple (RCP 8.5) by the end of the century, resulting in a near tripling (RCP 8.5) of irrigation volumes. Overall, the total water demand in Germany decreases significantly in both scenarios. In the lower scenario, water demand falls from around 26 billion m³/a to around 9 billion m³/a by the end of the century. In the upper scenario, it is reduced to around 12 billion m³/a by the end of the century. These enormous decreases in total water demand are due to reductions in water demand in the energy sector, which overlay increases in domestic and agricultural water demands. mHM-simulations of groundwater recharge based on climate projections show constant or increasing groundwater recharge rates in large parts of Germany in the ensemble median for the 2021-2050, 2036-2065 and 2069-2098 time slices, assuming both RCP 2.6 (21 RCMs) and RCP 8.5 (49 RCMs). However, declining groundwater recharge rates may also occur in certain regions, particularly in south-western Germany. In the 25th percentile of the model ensemble, falling groundwater recharge rates occur under RCP 2.6 in southern and western Germany. Towards the end of the century, groundwater recharge rates increase in eastern Germany. Under RCP 8.5, the 25th percentile of the model ensemble shows mostly constant or increasing groundwater recharge rates. However, south-western Germany is characterized by a significant decline in groundwater recharge. The risk index water balance RIWB was defined as an indicator to evaluate the regional water supply in relation to the balance between water demand and groundwater recharge. The RIWB shows that the ratio of water demand to groundwater recharge can be expected to remain the same or improve in the most regions, while, depending on the scenario, 4% or 11% of districts/cities, particularly in northern Germany, must prepare for a deterioration in this ratio.

In Weber et al. (2019), https://doi.org/10.1029/2018wr024584 (hereafter W19), modeling soil hydraulic properties (SHP) was systematically framed and presented in a didactic, carefully thought‐out approach. In doing so, the authors coined the term SHP model system/model framework, acknowledging the decade old research on effective modeling of the SHP. At the heart of the model an integral was formulated that links non‐capillary saturation to capillary saturation, based on any given saturation function for the capillary part. This approach sparked the interest by Peters and Iden (2021), https://doi.org/10.1029/2020wr028397 who wrote a comment. In this reply, we providing a detailed perspective on the comment and scrutinize the opinions by Peters and Iden (2021), https://doi.org/10.1029/2020wr028397 . Additionally, we use the opportunity to clarify terminology with respect to the distinction between “non‐capillary” and “capillary” pore spaces. Further, the Brunswick model system presented in W19 has been implemented in the HYDRUS software suite (Diamantopoulos et al., 2024, https://doi.org/10.1002/vzj2.20326 ).

Climatic variability and recurrent drought can strongly affect the variability of crop yield and are therefore frequently considered a risk to food security in Ethiopia. A better understanding of how crop yields vary in space and time, and their relationship to climatic and other driving factors, can assist in enhancing agricultural production and adapting to and mitigating the impacts of climate change. We applied a multiple linear regression model to examine the spatiotemporal climatic signal (air temperature, precipitation, and solar radiation) in the yields of the most important crops (maize, sorghum, tef, and wheat) over the period 1995–2018. An analysis of the climatic data indicated that growing season temperature increased significantly in most regions, but the trends in precipitation were not significant. The yields of maize, sorghum, tef, and wheat tended to increase across most crop-growing areas, particularly in the west, but was highly variable. The results highlight large spatial differences in the contribution of climatic trends to crop-yield variability across Ethiopian regions. The trends in climatic variability did not significantly affect crop yields in some areas, whereas in the main crop-growing areas, up to − 39.2% of yield variability could be attributed to the climatic trends. Specifically, the climatic trends negatively affected maize yields but positively affected sorghum, tef, and wheat yields. Nationally, the average impacts of climatic trends on crop yields was relatively small, ranging from a 3.2% decrease for maize to a 0.7% increase for wheat. In contrast, technological advancements contributed substantially more to yield gains, with annual increases ranging from 4.3% for wheat to 5.1% for sorghum. These findings highlight the dominant role of non-climatic drivers, particularly improved agricultural technology, in shaping crop yield trends. Our findings underscore the spatial heterogeneity of climate impacts on agriculture and highlight the critical importance of technological progress in enhancing crop productivity. They also provide actionable insights for designing crop- and location-specific adaptation strategies, and stress the need for integrated, climate-resilient development pathways in the region.

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

à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers,

Background: The respiratory quotient, that is, the CO2-O-2-exchange ratio during aerobic heterotrophic soil respiration, (RQ(AHR)) is one of the key variables of the barometric process separation (BaPS) technique. For reliable application, the BaPS method requires the adjustment of RQ(AHR) to soil-specific substrate conditions. Aim: The objective of the present study was to determine RQ(AHR) in five arable soils varying between pH 4 and 8 and carbon/nitrogen (C/N) ratios between 8.4 and 11.9 using the BaPS-N-1(5) method. Further, we aimed to elucidate whether RQ(AHR) can be estimated from basic soil properties. Results: Determined RQ(AHR)s were soil specific and ranged between 0.74 and 1.53. RQ(AHR) particularly affected the calculation of gross nitrification rates but not soil respiration rates. Based on a multiple linear regression analysis, soil pH and the C/N ratio of the soil organic matter pool were identified to explain well (R-2 = 0.984) the measured RQ(AHR)s. Applying the thus derived RQ(MLR) to the calculation procedure led to good agreement between gross nitrification rates derived by the N-15 pool dilution technique. Conclusions: Future studies are needed to test whether this empirical relationship between RQ(AHR), pH, and C/N ratio is generalizable. If this would be the case, the BaPS method could be applied without the use of a stable isotope and thus, being less resource demanding than the N-15 pool dilution technique.

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