Affordable and accurate weather monitoring systems are essential in low-income and developing countries and, more recently, are needed in small-scale research such as precision agriculture and urban climate studies. A variety of low-cost solutions are available on the market, but the use of non-standard technologies raises concerns for data quality. Research-grade all-in-one weather stations could present a reliable, cost effective solution while being robust and easy to use. This study evaluates the performance of the commercially available ATMOS41 all-in-one weather station. Three stations were deployed next to a high-performance reference station over a three-month period. The ATMOS41 stations showed good performance compared to the reference, and close agreement among the three stations for most standard weather variables. However, measured atmospheric pressure showed uncertainties >0.6 hPa and solar radiation was underestimated by 3%, which could be corrected with a locally obtained linear regression function. Furthermore, precipitation measurements showed considerable variability, with observed differences of ±7.5% compared to the reference gauge, which suggests relatively high susceptibility to wind-induced errors. Overall, the station is well suited for private user applications such as farming, while the use in research should consider the limitations of the station, especially regarding precise precipitation measurements.

Due to their unique characteristics, cosmic-ray neutron sensors (CRNSs) have potential in monitoring and informing irrigation management, and thus optimising the use of water resources in agriculture. However, practical methods to monitor small, irrigated fields with CRNSs are currently not available and the challenges of targeting areas smaller than the CRNS sensing volume are mostly unaddressed. In this study, CRNSs are used to continuously monitor soil moisture (SM) dynamics in two irrigated apple orchards (Agia, Greece) of ~1.2 ha. The CRNS-derived SM was compared to a reference SM obtained by weighting a dense sensor network. In the 2021 irrigation period, CRNSs could only capture the timing of irrigation events, and an ad hoc calibration resulted in improvements only in the hours before irrigation (RMSE between 0.020 and 0.035). In 2022, a correction based on neutron transport simulations, and on SM measurements from a non-irrigated location, was tested. In the nearby irrigated field, the proposed correction improved the CRNS-derived SM (from 0.052 to 0.031 RMSE) and, most importantly, allowed for monitoring the magnitude of SM dynamics that are due to irrigation. The results are a step forward in using CRNSs as a decision support system in irrigation management.

The inclusion of perennial, woody crops in land surface models (LSMs) is crucial for addressing their role in carbon (C) sequestration, food production, and water requirements under climate change. To help quantify the biogeochemical and biogeophysical processes associated with these agroecosystems, we developed and tested a new sub-model, CLM5-FruitTree, for deciduous fruit orchards within the framework of the Community Land Model version 5 (CLM5). The model development included (1) a new perennial crop phenology description, (2) an adapted C and nitrogen allocation scheme, considering both storage and photosynthetic growth of annual and perennial plant organs, (3) typical management practices associated with fruit orchards, and (4) the parameterization of an apple plant functional type. CLM5-FruitTree was tested using extensive field measurements from an apple orchard in South Tyrol, Italy. Growth and partitioning of biomass to the individual plant components were well represented by CLM5-FruitTree, and average yield was predicted within 2.3 % of the observed values despite low simulated inter-annual variability compared to observations. The simulated seasonal course of C, energy, and water fluxes was in good agreement with the eddy covariance (EC) measurements owing to the accurate representation of the prolonged growing season and typical leaf area development of the orchard. We found that gross primary production, net radiation, and latent heat flux were highly correlated (r>0.94) with EC measurements and showed little bias (<±5 %). Simulated respiration components, sensible heat, and soil heat flux were less consistent with observations. This was attributed to simplifications in the orchard structure and to the presence of additional management practices that are not yet represented in CLM5-FruitTree. Finally, the results suggested that the representation of microbial and autotrophic respiration and energy partitioning in complex, discontinuous canopies in CLM5 requires further attention. The new CLM5-FruitTree sub-model improved the representation of agricultural systems in CLM5 and can be used to study land surface processes in fruit orchards at the local, regional, or larger scale.

Irrigation strongly influences land‐atmosphere processes from regional to global scale. Therefore, an accurate representation of irrigation is crucial to understand these interactions and address water resources issues. While irrigation schemes are increasingly integrated into land surface models, their evaluation and further development remains challenging due to data limitations. This study assessed the representation of field‐scale irrigation using the Community Land Model version 5 (CLM5) through comparison of observed and simulated soil moisture, transpiration and crop yield. Irrigation was simulated by (a) adjusting the current irrigation routine and (b) by implementing a novel irrigation data stream that allows to directly use observed irrigation amounts and schedules. In a following step, the effect of different irrigation scenarios at the regional scale was simulated by using this novel data stream. At the plot scale, the novel irrigation data stream performed better in representing observed SM dynamics compared to the current irrigation routine. Nonetheless, simplifications in crop and irrigation representation and uncertainty in the relation between water stress and yield currently limit the ability of CLM5 for field‐scale irrigation scheduling. Still, the simulations revealed valuable insights into model performance that can inform and improve the modeling beyond the field scale. At regional scale, the simulations identified irrigation priorities and potential water savings. Furthermore, application of LSMs such as CLM5 can help to study the effects of irrigation beyond water availability, for example, on energy fluxes and climate, thus providing a powerful tool to assess the broader implications of irrigation at larger scale.

Land surface modeling as a tool to explore sustainable irrigation practices in Mediterranean fruit orchardsO. Dombrowski1, C. Brogi1, H.-J. Hendricks Franssen1, V. Pisinaras2, A. Panagopoulos2, S. Swenson3, H. Bogena11Agrosphere (IBG-3), Forschungszentrum Jülich GmbH, Jülich 52425, Germany2Soil & Water Resources Institute, Hellenic Agricultural Organization, Gorgopotamou, Sindos, Thessaloniki 57400, Greece3Climate and Global Dynamics Laboratory, National Center for Atmospheric Research, Boulder 80305, Colorado, United States of AmericaCorresponding author: Olga Dombrowski (o.dombrowski@fz-juelich.de)

Proper agricultural soil carbon management is important since soil organic matter plays a crucial role in many soil functions. Due to the variety of field-specific soil carbon management options, decision support for farmers is beneficial. We designed and established a field-scale digital twin for soil carbon management by combining an agroecosystem process model with inversion algorithms and sensor data. A prototype of the digital twin was implemented for an Integrated Carbon Observation System (ICOS) field-site near Cologne, Germany. Soil water fluxes, soil thermal fluxes, crop-specific evapotranspiration and photosynthesis processes were calibrated using sensed soil water content, soil temperature, latent heat flux, net ecosystem exchange (NEE) or yield data. The digital twin calibrated using measured on-site NEE observations resulted in a model efficiency (ME) of 0.86, whereas calibration with yield data resulted in a similar ME of 0.84 for NEE. This suggests that a digital twin calibrated using plot-specific yield observations is sufficiently accurate to characterize soil carbon turnover. In a next step, the digital twin was used to evaluate carbon management scenarios for decision support. The results of this scenario analysis suggest that retaining cereal harvest residues on the field offers significant potential for carbon accrual at this site. The results also demonstrate the effectiveness of established management practices, such as growing cover crops (+2.7 Mg C ha-1 over 13 years) and using organic fertilization (+0.55 Mg C ha-1 over 13 years) on soil organic carbon (SOC) stocks. We conclude that a digital twin calibrated on yield data is a helpful decision-support tool for soil carbon management.