Yield‐Dependent Allocation Functions Overestimate Root‐Derived Carbon Inputs in Wheat and Maize

Multi-model ensemble improved the prediction of trends in soil organic carbon stocks in German croplands

Soil organic carbon (SOC) constitutes the largest terrestrial biological carbon pool globally. SOC in croplands has declined by approximately 50% since the intensification of agriculture. In light of climate change due to rising greenhouse gas concentrations in the atmosphere, the 4p1000 initiative was launched, suggesting that anthropogenic CO2 emissions could be offset by increasing SOC stocks in arable land by 0.4% per year by implementing more sustainable agronomic measures. In order to estimate the potential effect of different measures on SOC at the national scale, modelling approaches are required. In the last decades, a wide array of SOC models have been developed and validated for different soils, climate conditions and land uses across the globe. These models all have their own advantages, disadvantages, and sources of uncertainty. Carbon inputs into soil, a major driver of SOC dynamics, are an estimated quantity in all modelling procedures and represent an additional, large source of uncertainty. To reduce uncertainties, multi-model ensembles are suggested to outperform single model runs. The objective of this study is to determine the optimal SOC model ensemble to reduce estimation errors in future studies.Therefore, a combination of four carbon turnover models (RothC, Yasso07, ICBM, and C-TOOL) and five published carbon input estimation methods was evaluated by comparing simulations to experimental data from six long-term experiments with 56 treatments on arable land in Austria, with durations from 10 to 32 years to obtain a possible optimal combination for future SOC modelling studies in Austria. Evaluation of model prediction was performed by calculating the absolute mean error (AME), Root Square Mean Error (RMSE) and coefficient of determination on yearly SOC changes to eliminate the effect of different experimental durations on model evaluation.We show that obtained models strongly differ in their stock estimates, and our selected ensemble strongly improved the estimations of SOC against single model runs with significantly lower absolute mean errors and root mean square error. This is in accordance with literature results and presents a way forward towards a more accurate modelling. We thus argue that multi-model ensembles to estimate SOC stocks in arable soils in Austria should be preferred over single-model approaches due to improved accuracy.

Hou et al. (2025) conducted a meta-analysis of field studies from 2019 to 2023 to evaluate the effects of noncontinuous flooding (NCF) practices on net carbon sequestration (NCS) in rice fields. The authors concluded that compared with continuous flooding (CF), NCF substantially enhances ecosystem-scale NCS, primarily by reducing methane (CH4) emissions while increasing photosynthetic carbon sequestration (PCS), despite a decrease in soil organic carbon (SOC) sequestration. This conclusion, however, warrants further scrutiny regarding methodological assumptions and data representativeness, potentially leading to a systematic overestimation of the carbon sequestration potential of NCF in rice systems. A primary concern lies in the calculation framework applied to rice systems, where PCS was directly added as a carbon sink component in the NCS budget. PCS represents short-term carbon input via plant photosynthesis, most of which is harvested and rapidly returned to the atmosphere through consumption, combustion, or decomposition. In agricultural systems, only carbon stored long term in soils or stable organic pools is considered true sequestration (Smith et al. 2020). Treating PCS as ecosystem-level carbon sequestration conflates “carbon input” with “carbon retention,” thereby overstating the mitigation potential of carbon sequestration pathways in croplands. This approach also diverges from IPCC guidelines, which emphasize long-term carbon storage as the accounting basis (Ogle et al. 2019). Additionally, since SOC sequestration was estimated by comparing SOC content before and after a single growing season, part of the PCS might have already been reflected in the SOC increment. Adding both PCS and SOC sequestration therefore raises the possibility of double counting in Hou et al. (2025). If the SOC change estimation method proposed by Guan et al. (2023) is adopted, NCS should be calculated based on both the net ecosystem exchange during the rice growing season and non-CO2 greenhouse gas emissions. SOC sequestration estimates themselves are also highly uncertain in Hou et al. (2025). Single-season SOC comparisons are vulnerable to sampling error, spatial heterogeneity, and seasonal variability. Numerous studies have shown that detectable SOC changes require multi-year monitoring due to the slow turnover and delayed stabilization of carbon inputs (Smith et al. 2020). Due to the fact that most studies did not incorporate straw return or organic amendments, the reported SOC increase under CF, which exceeds 2800 kg CO2-eq ha−1 in one season (see figure 7h, Hou et al. 2025), is unusually high. The values of SOC sequestration reported by Hou et al. (2025) substantially exceed estimates from comparable field conditions without external carbon inputs (Lessmann et al. 2022; Liu et al. 2024). Hou et al. also attributed the reduction in SOC sequestration under NCF to increased microbial biomass carbon, implying greater microbial activity and enhanced SOC decomposition. However, increased microbial biomass merely indicates a larger microbial biomass pool and does not necessarily imply higher microbial activity or increased SOC mineralization. Assessing microbial activity requires dynamic, process-based indicators such as respiration rates, enzyme activities, or mineralization fluxes, which were not evaluated in their analysis (Blagodatskaya and Kuzyakov 2013). Dataset constraints further weaken the conclusions. Only 23 observations from seven publications report complete NCS metrics, including yield, SOC, and non-CO2 greenhouse gas emissions, with limited geographic coverage and small sample sizes (Figure 1a). Unfortunately, studies published before 2019 were excluded by Hou et al. (2025), despite representing a comparable volume of relevant research (Nikolaisen et al. 2023). Incorporating these earlier data significantly reduces the estimated CH4 mitigation effect of NCF and may negate or even reverse its yield advantage (Figure 1b). While these factors are beyond the scope of this Letter, it is important to acknowledge that variations in water management practices and the use of organic amendments in rice systems can significantly affect emissions, yield, and SOC dynamics, and should not be overlooked in future assessments. Jinyang Wang: writing – original draft. Jianwen Zou: writing – review and editing. The authors declare no conflicts of interest. This article is a Letter to the Editor regarding Hou et al., https://doi.org/10.1111/gcb.70283. See also the Response to the Letter by Hou et al., https://doi.org/10.1111/gcb.70464. No new data were generated in this study. The data sources used for Figure 1 are documented in Hou et al. (2025) and Nikolaisen et al. (2023).

This study investigates the operational efficiency of the lab-scale oxidation ditch (OD) functioning in simultaneous nitrification and denitrification modes, focusing on forecasting biochemical oxygen demand (BOD5) concentrations over a five-day horizon. This forecasting capability aims to optimize the operational regime of aeration tanks by adjusting the specific load on organic pollutants through active sludge dosage modulation. A comprehensive statistical analysis was conducted to identify trends and seasonality alongside significant correlations between the forecasted values and various time lags. A total of 20 time lags and the “month” feature were selected as significant predictors. These models employed include Multi-head Attention Gated Recurrent Unit (MAGRU), long short-term memory (LSTM), Autoregressive Integrated Moving Average–Long Short-Term Memory (ARIMA–LSTM), and Prophet and gradient boosting models: CatBoost and XGBoost. Evaluation metrics (Mean Squared Error (MSE), Mean Absolute Error (MAE), Symmetric Mean Absolute Percentage Error (SMAPE), and Coefficient of Determination (R2)) indicated similar performance across models, with ARIMA–LSTM yielding the best results. This architecture effectively captures short-term trends associated with the variability of incoming wastewater. The SMAPE score of 1.052% on test data demonstrates the model’s accuracy and highlights the potential of integrating artificial neural networks (ANN) and machine learning (ML) with mechanistic models for optimizing wastewater treatment processes. However, residual analysis revealed systematic overestimation, necessitating further exploration of significant predictors across various datasets to enhance forecasting quality.

BackgroundThe aims of this study were to create a method for estimation of dental age in Saudi children and adolescents based on the Willems model developed using the Belgian Caucasian (BC) reference data and to compare the ability of the two models to predict age in Saudi children.MethodsDevelopment of the seven lower left permanent mandibular teeth was staged in 1146 panoramic radiographs from healthy Saudi children (605 male, 541 female) without missing permanent teeth and without all permanent teeth fully developed (except third molars). The data were used to validate the Willems BC model and to construct a Saudi Arabian-specific (Willems SA) model. The mean error, mean absolute error, and root mean square error obtained from both validations were compared to quantify the variance in errors in the sample.ResultsThe overall mean error for the Willems SA method was 0.023 years (standard deviation, ± 0.55), indicating no systematic underestimation or overestimation of age. For girls, the error using the Willems SA method was significantly lower but still negligible at 0.06 years. A small but statistically significant difference in total mean absolute error (11 days) was found between the Willems BC and Willems SA models when the data were compared independent of sex. The overall mean absolute error for girls was slightly lower for the Willems BC method than for the Willems SA method (1.33 years vs. 1.37 years).ConclusionsThe difference in ability to predict dental age between the Willems BC and Willems SA methods is very small, indicating that the data from the BC population can be used as a reference in the Saudi population.

WRF-Solar is a numerical weather prediction model specifically designed to meet the increasing demand for accurate solar irradiance forecasting. The model provides flexibility in the representation of the aerosol–cloud–radiation processes. This flexibility can be argued to make it more difficult to improve the model’s performance because of the necessity of inspecting different configurations. To alleviate this situation, WRF-Solar has a reference configuration to use as a benchmark in sensitivity experiments. However, the scarcity of high-quality ground observations is a handicap to accurately quantify the model performance. An alternative to ground observations are satellite irradiance retrievals. Herein we analyze the adequacy of the National Solar Radiation Database (NSRDB) to validate the WRF-Solar performance using high-quality global horizontal irradiance (GHI) observations across the contiguous United States (CONUS). Based on the sufficient performance of NSRDB, we further analyze the WRF-Solar forecast errors across the CONUS, the growth of the forecasting errors as a function of the lead time, and sensitivities to the grid spacing and the representation of the radiative effects of unresolved clouds. Our results based on WRF-Solar forecasts spanning 2018 reveal a 7% median degradation of the mean absolute error (MAE) from the first to the second daytime period. Reducing the grid spacing from 9 to 3 km leads to a 4% improvement in the MAE, whereas activating the radiative effects of unresolved clouds is desirable over most of the CONUS even at 3 km of grid spacing. A systematic overestimation of the GHI is found. These results illustrate the potential of GHI retrievals to contribute to increasing the WRF-Solar performance.

Large-scale hydrological models simulate the water cycle for regions, countries, and continents. The choice of input data directly impacts the accuracy of these models' final output and the spatial and temporal pattern, as well as the quality of data (air temperature and precipitation), influences the quality and pattern of water availability estimates. It is essential to acknowledge that the accuracy of these estimates depends on the input quality of the data used.In this regard, precipitation and air temperature gridded European Meteorological Observations (EMO1) datasets specifically used for hydrological modeling inputs (CWATM) are evaluated over the Danube River Basin (DRB). The observation data (air temperature and precipitation) from 9 different countries within the DRB are used for the EMO1’s evaluation.  The performance of the datasets was evaluated at daily, monthly, and annual scales, using Pearson Correlation (r), root mean square errors (RMSE), mean absolute errors (MAE), Nash Sutcliffe Efficiency (NSE), Percent Bias (PBIAS), and Kling Gupta Efficiency (KGE) criterion.The results showed the range of temperature differences varies between approximately -3°C and +2°C. This reflects both underestimations and overestimations by EMO1 compared to observations. The median differences are close to 0 for most months, indicating the EMO1 model is generally unbiased or well-calibrated overall. Larger variability and more outliers occur in warmer months (e.g., May–August), suggesting the model may struggle with accurately capturing summer temperature dynamics. For precipitation, the median is slightly positive, suggesting a systematic overestimation of precipitation during the summer months. This could be due to the model overestimating convective rainfall. By identifying the periods where the EMO-1 deviates most from observations, researchers can target specific processes for calibration or refinement, which is especially important for hydrological  applicationsAcknowledgment This work was supported as part of DANUBE WATER BALANCE, an Interreg Danube Region Programme project co-funded by the European Union. 

Accurate dietary assessment is vital for preventing malnutrition in aging populations, particularly in home-care settings. Although Large Multimodal Models (LMMs) for nutrient estimation are evolving, their nutrient-specific accuracy requires rigorous validation. Fifteen standardized hospital meals were photographed under controlled conditions (90-degree angle, 500 lux). Ground truth values were determined by direct weighing. Estimates for energy and macronutrients were performed by 10 registered dietitians (RDs) and 10 AI models (including ChatGPT-4o and Gemini 1.5 Pro). Accuracy was assessed using Pearson's correlation, Mean Absolute Error (MAE), and Bland-Altman analysis to quantify systematic bias. For energy and carbohydrates, RDs and top-performing AI models (notably ChatGPT-4o and Gemini 1.5 Pro) demonstrated practical accuracy (r > 0.8, frequently within ±10% range). However, accuracy for protein and lipids was significantly lower across all AI models. Specifically, all AI models exhibited a substantial systematic overestimation of lipids (Mean Bias > +20%, p < 0.01), highlighting a critical "invisible nutrient" bias. Current AI tools show potential for caloric and carbohydrate monitoring but struggle with lipid and protein density. These findings emphasize the need for human-AI collaboration ("human-in-the-loop") and the integration of cooking metadata to improve clinical utility in geriatric nutrition.

Mapping the COPD Assessment Test onto EQ-5D

Borges, A, Teodósio, C, Matos, P, Mil-Homens, P, Pezarat-Correia, P, Fahs, C, and Mendonca, GV. Sexual dimorphism in the estimation of upper-limb blood flow restriction in the seated position. J Strength Cond Res 32(7): 2096-2102, 2018-Arterial occlusion pressure (AOP) is typically used to normalize blood flow restriction (BFR) during low-intensity BFR exercise. Despite strong evidence for sexual dimorphism in muscle blood flow, sex-related differences in AOP estimation remain a controversial topic. We aimed at determining whether the relationship of upper-limb AOP with arm circumference and systolic blood pressure (BP) differs between men and women resting in the seated position. Sixty-two healthy young participants (31 men: 21.7 ± 2.3; 31 women: 22.0 ± 2.0 years) were included in this study. Arm circumference, resting BP, and AOP were taken in the seated position. Multiple linear regression analysis was used to determine whether the relationship of AOP with arm circumference and resting BP differed between sexes. Prediction accuracy was assessed with the mean absolute percent error and Bland-Altman plots. Men had higher systolic BP and larger arm circumference than women (p < 0.05). Nevertheless, AOP was similar between sexes. Arm circumference, systolic BP, and sex were all significant predictors of AOP (p < 0.05), explaining 42% of its variance. The absolute percent error was similar in both sexes (men: -0.55 ± 7.12; women: -0.39 ± 6.31%, p > 0.05). Bland-Altman plots showed that the mean difference between actual and estimated AOP was nearly zero in both groups, with no systematic overestimation or underestimation. In conclusion, arm circumference, systolic BP, and sex are all significant predictors of upper-limb-seated AOP. Their measurement allows for the indirect estimation of BFR pressure within the context of exercise training.

Background: Wrist-worn consumer activity trackers are widely used to promote physical activity (PA) and reduce sedentary behavior (SB). However, evidence regarding their validity for measuring PA and SB in free-living school-aged children remains limited. This study evaluated the concurrent validity and wear compliance of a wrist-worn consumer activity tracker in school-aged children under free-living conditions with protocol-defined wear requirements. Methods: A total of 102 children (mean age: 10.2 years; 44.1% girls) wore a wrist-worn device (Fitbit Ace) and a waist-worn accelerometer (Omron Active Style Pro HJA-750c, ASP-750c). Of the 1122 person-days collected over 11 days, 135 person-days meeting inclusion criteria for both devices were included (≥10 h/day wear time and an inter-device wear time difference of ≤60 min). Step count and time in SB, light (LPA), moderate (MPA), vigorous (VPA), and moderate-to-vigorous PA (MVPA) were assessed. Correlations, mean absolute percentage error (MAPE), agreement, and wear compliance between the two devices were examined. Results: Correlations were strong for step count (r = 0.86), SB (r = 0.72), and LPA (r = 0.71); however, agreement was poor, with systematic overestimation of step count, SB, VPA, and MVPA and underestimation of LPA and MPA by the Fitbit Ace, and MAPE exceeding 20% for all PA variables. Wear compliance (≥10 h/day on ≥4 days) was higher for the Fitbit Ace (97.0%) than for the ASP-750c (62.2%). Conclusions: Although the Fitbit Ace may be useful for characterizing general patterns of LPA and SB in school-aged children, caution is warranted for accurate individual-level PA assessment.

A system for ensemble streamflow prediction, ESP, has been operational at SMHI since July 2004, based on 50 meteorological ensemble forecasts from ECMWF. Hydrological ensemble forecasts are produced daily for 51 basins in Sweden. All ensemble members, as well as statistics (minimum, 25% quartile, median, 75% quartile and maximum), are stored in a database. This paper presents an evaluation of the first 18 months of ESP median forecasts from this system, and in particular their performance in comparison with today's categorical forecast. The evaluation was made in terms of three statistical measures: bias B, root mean square error RMSE and absolute peak flow error PE. For ESP forecasts the bias ranged between -20% and 80% with a systematic overestimation for Sweden as a whole. A comparison between bias in input precipitation and ESP output, respectively, revealed only a weak relationship, but streamflow overestimation is likely related mainly to model properties. The results from the streamflow forecast comparison showed that the ESP median in deterministic terms performs overall as well as the presently used categorical forecast. Further, ESP has the advantage of providing at least a qualitative measure of the uncertainty in the forecasts, with probability forecasts being the ultimate goal.

Extensive benchmarking calculations are presented to assess the accuracy of the standard approximate coupled cluster singles and doubles method (CC2) in studying ππ* excited states properties of model protein chains containing a phenylalanine residue, namely capped peptides, whose ground state conformers adopt the prototypical secondary structural features of proteins. First, the dependence with the basis set of the CC2 excitation energies, CC2 geometry optimizations, and amide A region frequencies of the lowest ππ* excited state in a reference system, the N-acetylphenylalaninylamide, are investigated, and the results are compared with experimental data. Second, at the best level of theory determined, the CC2/aug(N,O,π)-cc-pVDZ//CC2/cc-pVDZ level, a series of capped peptides of increasing size and containing residues of different nature are investigated. Along the series, compared to the experimental values, a mean absolute error of 0.10 eV is achieved for the 0-0 transition energies with a systematic overestimation. In addition, mode-dependent linear scaling functions for the calculated frequencies of the amide A region have been determined from the set of 95 experimental frequencies available; they lead to a quantitative simulation of the observed shifts of the amide A region frequencies upon ππ* excitation (root-mean-square deviation of 5 cm-1). These results confirm the reliability of the CC2 method to characterize the lowest ππ* excited state of such medium-sized systems, emphasizing this class of theoretical approaches as a relevant spectroscopic tool, including for tasks as difficult as conformational assignment.

The three-temperatures (3T) method is a robust approach to estimating evapotranspiration (ET), requiring relatively few measurable, physical parameters and an imitation surface, making it potentially suited for estimating ET from sustainable drainage systems (SuDS) and green infrastructure (GI) in urban environments. However, limited 3T-ET data from SuDS and/or GI makes it difficult to assess the conditions that affect its accuracy. The purpose of this study was to determine whether reasonable ET estimates could be achieved using the 3T method with a plastic imitation surface for a small, homogenous vegetated surface. The 3T-ET estimates were produced at an hourly timestep and compared to reference ET (ETo) derived using the Penman–Monteith equation. The 3T-ET estimates were consistently higher than ETo (mean absolute error of 0.05 to 0.15 mm·h−1), which may indicate systematic overestimation of ET or that the actual ET was greater than ETo. Unrealistic 3T-ET estimates are produced when the air temperature and the imitation surface temperature converge, limiting the method’s application to between mid-morning and late afternoon. Further work to validate and refine the 3T method is required before it can be recommended for deployment in the field for spot-sampling ET rates from urban SuDS/GI.

Dados de evapotranspiração de referência (ETo) são requeridos em diversas análises relacionadas ao uso de recursos hídricos, como no manejo de irrigação, balanço hídrico de sequeiro, manejo de bacias hidrográficas, etc. O método FAO Penman-Monteith (FAO-PM) é considerado padrão para cálculo da ETo. Contudo, a aplicação deste método pode ser restringida pela falta de dados meteorológicos. Um dos métodos alternativos com menor requisição de dados é o Hargreaves-Samani (HS), o qual comumente necessita calibração local para obtenção de maior exatidão na aplicação. O objetivo deste trabalho foi avaliar o desempenho do método Hargreaves-Samani, na sua forma original e calibrado em base anual, semestral, trimestral e mensal, frente ao método padrão FAO-PM, com base em extensa série histórica de dados meteorológicos diários (86 anos), para a localidade de Sete Lagoas, MG. Calibraram-se os coeficientes e o expoente da equação de HS pela minimização do erro absoluto médio (EAM). Para avaliar a adequabilidade de aplicação de testes comparativos e para determinar o ajuste das distribuições de probabilidade de ETo, foram aplicados os testes de Kolmogorov-Smirnov e Mann-Whitney, respectivamente. A calibração do método HS anulou a tendência de superestimativa sistemática dos valores de ETo diária, comparativamente ao FAO-PM, reduzindo-se o EAM. A calibração em diferentes bases temporais acarretou a aproximação das distribuições de probabilidade de ETo calculada por meio do método HS, que não diferiram significativamente entre si, à obtida com o método FAO-PM.