Legacy effects of climate extremes on deep soil water storage and water use efficiency across different land-use systems

Legacy effects of climate extremes on deep soil water storage and water use efficiency across different land-use systems | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Legacy effects of climate extremes on deep soil water storage and water use efficiency across different land-use systems Mengqi Wu, Christiane Roscher, Martin Schädler, Mika Tarkka, Doris Vetterlein, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7062058/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Climate extremes, including multiyear droughts and extreme rainfall events, are projected to intensify, threatening the global water cycle and reducing agricultural productivity. Deep soil water storage plays a key role in buffering extremes, yet its influence on plant productivity and water use across land-use systems remains insufficiently understood. Here, we monitored soil moisture dynamics over three years and vegetation responses in a long-term field trial comprising five land-use types (two croplands: conventional & organic farming; three grasslands: intensive meadow, extensive meadow & pasture). The monitoring period captured both prolonged droughts and an extreme rainfall. We found strong legacy effects of past droughts on deep soil water storage (30–110 cm), which decoupled plant productivity from short-term climate fluctuations. Extensive grasslands exploited the deep soil water storage more efficiently than intensive grasslands and croplands, because of longer transpiration demand and higher interception caused by the perennial vegetation cover. In turn, water use efficiency increased with land-use intensity, driven by shorter growing periods in croplands and higher mowing frequency in intensive grasslands. Our findings highlight how land-use practices shape ecosystem responses to climate extremes and underscore the need to incorporate deep soil water dynamics into sustainable land-management strategies under future climate conditions. Earth and environmental sciences/Climate sciences/Hydrology Earth and environmental sciences/Ecology/Ecosystem services Scientific community and society/Agriculture Earth and environmental sciences/Environmental social sciences/Climate-change adaptation climate change cropland grassland water cycling root length density plant community composition yield Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 1. INTRODUCTION Extreme drought and precipitation events, projected to increase in frequency, duration and intensity with human-induced climate change, threaten the global water cycle and agricultural productivity (Maurel and Nacry, 2020 ; Wunsch et al., 2022 ; Chen et al., 2025 ). Soil water content, a critical mediator between meteorological conditions and biogeochemical processes (Li et al., 2024 ; Sun et al., 2025 ; Vicente-Serrano et al., 2025 ), governs root and shoot growth dynamics and thus ecosystem productivity (Zuo et al., 2006 ; Vereecken et al., 2022 ; Tissink et al., 2025 ). Yet, the supply of precipitation and the storage of water in the topsoil often fail to meet the water demands of plants, particularly during peak growth (Ali et al., 2021 ; Wang et al., 2024a ). Deep soil water storage, which is accessible to plant roots, may play an unaccounted role in buffering against summer droughts, sustaining aboveground growth and belowground ecosystem stability under extreme climatic conditions (Wang et al., 2024b ). However, the extent to which deep soil water storage stabilizes plant productivity and ecosystem functions under climate extremes across different land-use systems remains poorly understood. The spatiotemporal distribution of soil water is strongly influenced by land-use type and intensity through the differences in vegetation cover, root system activity, and the timing and duration of water use across the growing season (Spera et al., 2016 ; Zhao et al., 2017 ; Wang-Erlandsson et al., 2022 ; Yu et al., 2025 ). Understanding how ecosystems respond differently to climate extremes under various land-use types and intensities is critical to predict future ecosystem services. For example, annual cropland and perennial grassland differ markedly in plant community composition, root system phenology, and seasonal water demand (Fan et al., 2015 ; Yang et al., 2024 ; Wang et al., 2024b ). Grasslands, dominated by perennial plant species, typically maintain dense and continuous vegetation cover throughout the year, while croplands undergo seasonal cycles of sowing and harvest, leading to periods of low cover or even bare ground and increased evaporative water loss from the topsoil (Wang et al., 2012 ; Bagley et al., 2017 ). Moreover, croplands are characterized by shallow, fast-developing root systems that concentrate water uptake during a short peak growth stage, whereas grasslands support deeper and more persistent root networks that sustain water uptake and continuous transpiration throughout the growing season (Holmes and Rice, 1996 ; Swindon et al., 2019 ). These contrasting root and vegetation cover dynamics result in distinct seasonal patterns of soil water depletion and influence the ecosystem’s capacity to withstand climate extremes (Maestre et al., 2016 ; Scherzinger et al., 2024 ). Additionally, land-use intensity, characterized by fertilization dose, tillage intensity in croplands, or mowing frequency, the use of a few grass cultivars or a more diverse species composition in grasslands, can further modify root development and vegetation cover, this in turn affects water uptake efficiency and soil water distribution (Rose et al., 2011 ; Spera et al., 2016 ). The interaction between land-use type and intensity and a projected future climate scenario in shaping soil water dynamics, and the potential implications of this for plant productivity under climate extremes, remains unclear. In particular, the relationship between rooting density and water availability remains poorly understood, complicating efforts to interpret plant responses to climate extremes. This complexity arises from several reasons: (1) water uptake is not uniform across the entire root system but can be root type specific, is higher close to root tips and can be altered by the presence of root hairs (Segal et al., 2008 ; Ahmed et al., 2018 ); (2) spatial heterogeneity in soil water can decouple the water uptake patterns from root density distribution (Carminati et al., 2016 ); and (3) elevated root densities may reflect nutrient foraging strategies rather than water demand (Lynch, 2019 ). This mismatch between root distribution and water uptake poses a critical challenge for understanding the plant adjustment to extreme climate conditions. The structure and function of deep rooting systems and their contribution to deep soil water uptake remain poorly understood, particularly under field conditions (Draye et al., 2010 ; Pierret et al., 2016 ). These uncertainties hinder a comprehensive evaluation of the extent to which shifts in rooting patterns constitute an effective mechanism for coping with water stress under changing climate. Plants rely on different strategies that enhance water uptake and improve water use efficiency (WUE) to maintain productivity under climate extremes. In agricultural systems, management strategies include selecting or breeding for cultivars with inherently higher WUE, optimizing cropping system and soil tillage regimes in croplands or adjusting mowing schedules in grasslands (Condon et al., 2004 ; Rose et al., 2011 ; Sun et al., 2018 ). While such strategies may delay the onset of drought stress, they often have limited effectiveness in sustaining overall productivity under prolonged water deficits (Leakey et al., 2019 ). Drought resistance tends to decline with increasing land-use intensity as higher productivity is mostly associated with higher water demand (Van Sundert et al., 2021 ; Bazzichetto et al., 2024 ; Korell et al., 2024 ), potentially exacerbating ecosystem vulnerability under intensified management regimes. In grasslands, diverse plant communities contribute to ecosystem resistance under climate extremes by combining shallow- and deep-rooting species, which enhances partitioning of soil water resources and minimizes competition under water stress (Craine et al., 2012 ; Isbell et al., 2015 ; Weides et al., 2024 ). Increasing attention has been paid to the legacy effect of climate extremes: persistent alterations in soil water availability that extend beyond the period of stress itself (Bastos et al., 2020 ; Muller and Bahn, 2022 ; Sun et al., 2022 ; Liu et al., 2025 ). This legacy of soil water storage may reshape subsequent rooting patterns, alter plant community composition, and influence WUE, with significant implications for the sustainability of plant production systems under ongoing climate change. The extent to which deep soil water storage, driven by the legacy effect of climate extremes, can buffer productivity loss and affect WUE under future climate scenario remains poorly understood. Therefore, we made use of a unique experimental platform, Global Change Experimental Facility – GCEF (see Schädler et al. ( 2019 )), in which a realistic future climate scenario is compared against ambient climate across five land-use types (two croplands and three grasslands), each representing a gradient of land-use intensities. The objective of this study was to investigate how climate extremes influence spatiotemporal patterns of soil water storage under different land-use types and how these patterns are affected by a future climate scenario involving modified seasonal precipitation and higher temperature. We further aimed to investigate whether and how legacy effects of deep soil water storage influence plant productivity and WUE in response to climate extremes that we observed in a three-year monitoring campaign. We hypothesized that, (1) topsoil water content is primarily driven by atmospheric forcing, but deep soil water content is regulated by root water uptake and legacy effect of past soil moisture conditions. (2) On an annual basis, croplands maintain a higher soil water content than grasslands due to their sparser root systems and shorter growing season, whereas the impact of the future climate scenario on deep soil water storage is unclear and should be clarified by this study. Furthermore, we expected that (3) crops achieve higher yields than grasslands due to breeding for resource efficiency and rapid growth, whereas extensive grasslands, with greater plant diversity, are most resistant to climate extremes. Intensive grasslands, characterized by low diversity but permanent cover, are expected to be less resilient than extensive grasslands but more stable than croplands, due to perennial root systems and continuous cover. Finally, we hypothesized that (4) plant adjustment to drought conditions and limitations in deep soil water supply would mainly occur through higher WUE and less through changing rooting patterns or changes in vegetation composition. These hypotheses were tested through three years of monitoring soil water content profiles, yield, and WUE across all five land-use types and two climate scenarios. Interpretation of the observed growth and water uptake patterns was supported by data on plant community composition and root length density profiles collected during peak growth. Our analysis explicitly accounted for site-specific precipitation history and interannual climatic variability, providing a robust context for evaluating land-use and climate interactions on ecosystem water dynamics and productivity. 2. MATERIAL AND METHODS 2.1 Site description Our study was conducted within the Global Change Experimental Facility (GCEF) of the Helmholtz-Centre for Environmental Research (UFZ) in Bad Lauchstädt, Germany (51° 23′ N, 11° 53′ E, 118 m a.s.l.). The area is characterized by a sub-continental climate, with a mean annual precipitation of 525 mm (1993–2013) and a mean annual temperature of 9.7°C (1993–2013) (Schädler et al., 2019 ). The soil type is a Haplic Chernozem, with on average 21% clay, 69% silt, and 10% sand (Altermann et al., 2005 ). The volumetric water content at field capacity (pF1.8 = -60 hPa) and permanent wilting point (pF4.2 = -15000 hPa) amounted to 40.7% and 14.9% in the topsoil, respectively (Fig. S1 ). The GCEF platform was established in 2013 and consists of a split-plot design with two climate scenarios as the main plot factor and five land-use types as the subplot factor. A more detailed description of this facility is given in Schädler et al. ( 2019 ). In brief, the GCEF includes 50 subplots which are arranged in ten main plots of 80 × 24 m size with five main plots subjected to ambient climate and to future climate each. The climate treatment is based on a projection of the climate of Central Germany for the years 2070–2100 based on different dynamic regional climate models (Jacob and Podzun, 1997 ; Döscher, 2002; Rockel et al., 2008 ). The use of automated roofs and side panels in the future climate to passively increase the night temperatures increased mean air temperature on average by 0.55°C. Precipitation was increased by 10% in spring and autumn and reduced by 20% in summer as compared to the ambient climate (Schädler et al., 2019 ; Yin et al., 2020 ). The land-use types include two croplands: conventional farming (CF) with a typical regional crop rotation consisting of winter barley, triticale, and winter wheat, the application of chemical weed control and mineral fertilizers, and organic farming (OF) characterized by legume (white clover) replacing the triticale in the crop rotation, mechanical weed control, non-stained seeds and a restricted use of pesticides. It also includes three grasslands: intensive meadow (IM) sown with a commercially used mixture of cultivars of forage grasses, and managed by moderate fertilization and frequent mowing (2–3 times per year), extensive meadow (EM) sown with a species-rich mixture of more than 50 regionally typical grassland species and managed by moderate mowing (1–2 times per year), and extensive pasture (EP) with the same sown species mixture as EM but managed by 2–3 times one-day grazing events per year with sheep. For more details we refer to Schädler et al. ( 2019 ). The precipitation and irrigation patterns of both climate scenarios and the management of each land-use type from 2022 to 2024 are summarized in Fig. 1 . The cumulative growing season rainfall (GSR), measured as total precipitation and irrigation from April to October, was almost twice as high in 2023 and 2024 as in 2022 (Fig. 1 a). 2.2 Meteorological water deficit and standardized precipitation-evapotranspiration index (SPEI) The meteorological water deficit and SPEI are widely used to identify and quantify wet and dry climate events in ecological studies. Meteorological water deficit is the difference between precipitation and potential evapotranspiration. SPEI is derived from the non-exceedance probability of the water deficit, fitted to a three-parameter log-logistic distribution to account for common negative values (Beguería et al., 2014 ). Values below − 1.28 and above 1.28 indicate extreme dry and wet conditions, respectively (Isbell et al., 2015 ). Precipitation is calculated as the sum of precipitation and irrigation, whereas potential evapotranspiration is estimated using the Penman-Monteith equation, with solar radiation derived from the diurnal temperature range following the Hargreaves method (Hargreaves and Allen, 2003 ). As no crop-specific parameters are included, this estimate reflects atmospheric demand rather than actual plant evapotranspiration. The daily precipitation and temperature data used for water deficit and SPEI calculations were obtained from the German Meteorological Service (DWD) weather station in Bad Lauchstädt ( https://opendata.dwd.de/climate_environment/CDC/observations_germany/climate/daily/kl/ ) and aggregated to monthly values. The water deficit was also aggregated to monthly values, whereas the SPEI of a given month represented the 3-month average of this month and the two antecedent months to account for seasonal hydrological consistency. Detailed descriptions of the water deficit and SPEI calculation can be found in Vicente-Serrano et al. ( 2010 ) and Schnabel et al. ( 2022 ). The “ SPEI ” package in R (Beguería et al., 2017 ) was used for analysis. 2.3 Soil water content and storage profiles Soil moisture profiles were monitored at each subplot using time domain reflectometry (TDR) probe at approximately weekly to bi-weekly intervals from April 2022 to December 2024. The device was a TRIME-PICO T3/IPH44 probe (IMKO Micromodultechnik GmbH, Ettlingen, Germany), which allows non-destructive determination of soil volumetric water content in a soil profile (Kano-Nakata et al., 2019 ). On each subplot of five land-use types, a soil auger was used to drill down to 110–120 cm and install the plastic access tubes. The soil volumetric water content of each layer was calculated as the average of two measurements and were made at depths of 10, 20, 30, 50, 70, 90 and 110 cm. The probe was rotated by 90° between two measurements to compensate for the elliptical sensitivity of the measuring field. Missing data on the croplands from July to October/November in three years were due to the removal of access tubes between harvest in summer and tillage in fall. Topsoil and deep soil water storage were calculated as the average soil water content during the growing season (the period from the first day when the mean daily temperature consistently exceeds 5°C (the minimum threshold for the growth of most grassland and crop species in Central Germany) until harvest in early July) in the 0–30 cm and 30–110 cm soil layers, respectively. The change in topsoil and deep soil water storage was assessed by comparing at the beginning and end of the growing season. To check the accuracy of the TDR, the monitoring data were calibrated by a comparison with four volumetric water content measurements ranging from saturated soil water content to the permanent wilting point using a balance and the TDR probe in an access tube at the center of a homogeneous soil column (34 cm diameter, 35 cm height). The soil column consisted of sieved topsoil material from the GCEF that was compacted to a representative bulk density for the site (1.44 g cm − ³). To facilitate a homogeneous moisture profile across the sensor depth, a new soil column was packed for each water content and the water was already added before packing for the lower target water contents and partly added from the top after packing for the higher target water content with sufficient time for equilibration. Volumetric water contents were derived from the known dry weight and bulk density of the soil column determined after oven-drying at 105°C for 48 hours. This calibration resulted in the regression equation Y = 1.63X – 1.63, R 2 = 0.997, p < 0.01, n = 4, where X is the soil moisture measured by TDR and Y is the calculated soil moisture. The calibration was determined once and used for the entire monitoring period. 2.4 Root sampling and analysis Cylindrical soil cores were taken in May 2022 and 2023 using a 5-cm-diameter soil auger. Two cores were taken per subplot and sectioned into three depth layers: 0–15 cm, 15–30 cm and 30–50 cm, respectively. In total, 200 soil cores were collected (2 climate scenarios × 5 land-use types × 5 treatment replicates × 2 sampling replicates per subplot × 2 sampling years), resulting in 600 depth-specific samples. All samples were stored at 4°C immediately to inhibit microbial activity and prevent root decay. All soil cores were washed within a week after sampling with tap water over 0.63 mm sieves to extract the roots from the soil, with each sample requiring about 40 minutes to process. Care was taken to minimize damage to the root system during washing. The collected roots were stored in a 50% ethanol solution (Rotisol) prior to analysis. Roots were scanned on a flatbed scanner at 400 dpi (EPSON Perfection V700). Then the obtained images were analyzed using WinRHIZO Pro™ (Version 2019a, Regent Instruments, Canada) to measure root characteristics, i.e., root length density and root length distribution in different root diameter range. In addition, root length density was adjusted using correction coefficients specific to land use and year to obtain correlated root length density, as fresh roots primarily contribute to root water uptake. Fresh roots cannot be distinguished from root residues, that were dead before sampling, during root washing and sorting and also not with image analysis with WinRhizo software. These correction coefficients were estimated independently as the relative proportion of fresh roots to total roots (Fig. S2), derived from X-ray CT images of intact soil cores (Phalempin et al., 2025 ), which were taken in the same soil depths shortly after root sampling. The proportion of fresh roots was 85.1–91.2% in croplands and 50.4–59.4% in grasslands. 2.5 Vegetation recording Shortly before root sampling, an area of 30 cm in diameter, centered on the sampling points for soil coring, was marked with a metal frame in the grassland subplots. All plant species rooting within this area were recorded and species-level cover as well as total vegetation cover were estimated to the nearest percentage. 2.6 Evapotranspiration and water use efficiency Evapotranspiration (ET) was estimated using the soil water balance equation (Zeleke and Wade, 2012 ) as follows: $$\:ET=P+I+\varDelta\:S-R-D+CR\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\left[mm\right]$$ where ET is the evapotranspiration (mm), P is precipitation (mm) (referring to the values measured under the automated proofs), I is irrigation application (mm), \(\:\varDelta\:\) S is soil water storage change (mm) (soil water content at the start minus soil water content at the end of a given period for the 0-110 cm depth), R is surface runoff (mm), D is drainage from the root zone and CR is capillary rise to the root zone. Due to the deep soil profile and the large water holding capacity, runoff is zero in the field. Due to the deep groundwater table (about 32 m below surface), capillary rise is negligible. Lysimeters with 1 m deep intact soil monoliths on-site featured no seepage at the lower boundary during three years. Therefore, R , D and CR were set to zero in this study (Sun et al., 2010 ). The net change in water content during peak growth used to link with root length density at 0–50 cm depth was calculated as the cumulative water input ( P + I ) and \(\:\varDelta\:\) S in the 0–50 cm depth over eight weeks spanning four weeks before to four weeks after root sampling. Water use efficiency (WUE) is defined as yield divided by ET . The cumulative ET was calculated from the first day when the mean temperature consistently exceeds 5°C until the day before harvest. Yield data was obtained from machine harvest records on all subplots, where the sum of grain and straw was taken for the croplands. Harvest dates for croplands and grasslands were synchronized through linear interpolation of the biomass production between two cuts in the meadows to ensure comparability between land-use types. Extensive pastures were excluded from WUE estimation because biomass removal by sheep grazing could not be accurately determined. 2.7 Statistical analysis All statistical analyses were conducted using R 4.1.3. Linear mixed-effects models (LMMs) were fitted using the “ lmer” function in the “ lme4” package (Bates, 2005 ), incorporating random intercepts to account for the split-plot design and repeated measures. These models were applied to analyze soil water content and storage at different depths, root length density at different depths, total vegetation cover, and water use efficiency. To account for the compositional nature of functional group data, we calculated log-ratios using grass as the reference group. Specifically, we computed the log-ratio of forb to grass (log(forb/grass)) and legume to grass (log(legume/grass)). These log-ratios were then analyzed jointly in a linear mixed-effects model to capture functional group dependence. Land use, climate, year, and their interactions were fixed factors, while year (sampling time), and subplot nested within main plot served as random effects. For models involving root length density, sampling depth, as well as its interaction with land use and year, were additionally tested. Models were fitted using maximum likelihood (ML) estimation, and likelihood ratio tests (χ 2 ratio) were used to assess the significance of fixed effects through stepwise model comparison. We used the “emmeans” package (Lenth et al., 2018 ) to further analyze significant interactions, by running post-hoc pairwise comparisons of estimated marginal means between treatment levels if LMMs yielding significant effects of land-use types and interactions between land use and climate. Linear regression analysis was performed to assess relationships between net change in water content and correlated root length density. We used the “lme” function in the “piecewiseSEM” package (Lefcheck et al., 2015 ) to develop structure equation modeling (SEM) to further uncover the direct and indirect contributions of land use, climate scenario, and climate extremes (SPEI) to yield and water use efficiency (Fig. S3). Prior to scaling, the categorical levels of land use were ordered as CF, IM, OF, and EM to reflect the gradient of land-use intensity. SPEI, topsoil water storage, deep soil water storage, yield, and water use efficiency were the measured values during the growing seasons until crop harvest. The dataset under extensive pasture was excluded because the yield and water use efficiency were missing. If the initial model was identified as saturated based on Shipley’s test of d -separation, indicating by a non-significant Fisher’s C statistic ( Chi-square distributed), non-significant paths were subsequently removed to gain a reduced/best-fitted model (Shipley, 2009 ). Figures were produced with the ggplot2 package (Wickham et al., 2016 ). 3. RESULTS 3.1 SPEI and soil water content spatiotemporal dynamics for three years The SPEI values indicate mild- to extreme drought conditions (-2.39 < SPEI < -0.09) throughout 2022, followed by wetter conditions in mid- and late- 2023 and moderate drought to extreme wet conditions (-1.20 < SPEI < 2.08) in 2024 (Fig. 2 a). Notably, a heavy rainfall event in August 2023 resulted in a water surplus of 142 mm, causing the SPEI to shift abruptly from − 2.17 in July to 0.77 in August. The consequences of interannual changes in atmospheric forcing on soil water profiles are demonstrated exemplarily for extensive meadows under ambient climate scenario (Fig. 2 b). The soil moisture at 0–20 cm depth was closely related with the SPEI (looking three months backward in time, R 2 = 0.57, p < 0.001) and to a lesser degree also with monthly water deficit (R 2 = 0.23, p = 0.005) (Fig. 2 , S4). Summer drought conditions persisted in deeper soil layers, with water content approaching the permanent wilting point, which for our site is at 14.9% volumetric water content. However, precipitation alleviated these conditions in topsoil, resulting in drier deep soil than topsoil throughout most of the monitoring period. The legacy effect of summer drought was manifested by the long duration for infiltration fronts to reach the depth of 100 cm (five months in 2022, two months in 2023). In 2024, increased precipitation led to higher soil water content well above the permanent wilting point throughout the profile and the entire time, making it the wettest year in the monitoring period (Fig. 2 b). The heavy rain event in August 2023 (on average 5.8 mm per day, with a maximum daily rainfall of 59.3 mm and a cumulative total of 180.4 mm) caused short-term water ponding, but eventually the entire volume of precipitation was taken up by the soil profile without raising the water content to field capacity (at 40.7% soil water content) due to the enormous soil water deficit at the time (305 mm available capacity) (Fig. 1 a, 2 b). 3.1.1 Comparison of soil water content between cropland and grassland Cropland exhibited consistently higher soil water content than grassland across the entire soil profile, except for 0–10 cm depth in which cropland was drier (Fig. 3 a). The discrepancy in average soil water content between cropland and grassland increased with depth. Specifically, when averaged across the period for which data is available in both croplands and grasslands, soil water content in cropland was 7.6–7.8% lower than in grassland at 0–10 cm depth, but 4.0-15.8% greater at 20–50 cm depth, and 17.8–31.9% higher at 50–110 cm depth ( p < 0.05, Fig. 3 a, Table S1 ). The future climate scenario did not affect average soil water content compared to ambient climate irrespective of soil depth (Table S1 ). In the topsoil (0–30 cm), soil moisture exhibited strong seasonal fluctuations, characterized by pronounced increases during winter and declines in summer (Fig. 3 b). In contrast, the deep soil layer (30–110 cm) displayed a more gradual, delayed response, with cropland maintaining higher soil moisture than grassland during winter 2022, most of 2023, and spring 2024. Differences between climate scenarios were evident, but they were episodic and depended on land-use types. From January to June 2023, the water content of deep cropland soil under the future climate scenario was higher than under the ambient climate, reflecting the percolation of added irrigation in autumn of 2022 and spring of 2023. In contrast, the water content of deep grassland soil remained consistently lower after the heavy rainfall event in August 2023, as 20% of this event was shielded in the future climate scenario. 3.1.2 Comparison of soil water content between conventional and organic farming No significant differences were observed in 0–20 cm depth, but organic farming consistently exhibited higher soil water content than conventional farming in deeper soil layers ( p < 0.05, Fig. 4 a, Table S1 ). The differences between the two farming systems were more pronounced under the ambient climate scenario, with organic farming maintaining 6.4–17.9% higher water content levels ( p < 0.05, Fig. 4 a, Table S1 ). Higher soil moisture in the deeper soil under organic farming was most evident during the dry years of 2022 and 2023, but disappeared during the wetter year of 2024 (Fig. 4 b). Under the future climate scenario, soil water content in conventional farming was significantly higher from January to June 2023 in the deep soil layers as a delayed response to irrigation, but declined below ambient climate levels after August 2023 because of precipitation shielding. 3.1.3 Comparison of soil water content between intensive and extensive grasslands Compared to extensive meadows and pastures, intensive meadows had drier topsoil but moister deeper soil layers (Fig. 5 a, Table S1 ). On average, soil water content in intensive meadows across the entire profile was 21.9–41.8% and 13.1–18.6% higher under ambient and future climate scenarios, respectively (Fig. 5 a). The gap between deep soil water contents of intensive and extensive grasslands opened when the storage was declining in late summer and became smaller, when the storage was refilled in spring (Fig. 5 b). Under the future climate scenario, deep soil water content significantly declined after September 2023 compared to ambient climate, when the 20% deficit of the August rain event started to reach the subsoil. This discrepancy was strongest in intensive meadows, as the antecedent soil moisture was the same and persisted until the end of the monitoring period in December 2024 ( p < 0.05). However, in extensive pastures, soil water content under the future climate scenario was initially higher, a result of the additional irrigation in autumn and spring, but dropped below ambient climate levels after September 2023 (Fig. 5 b). 3.2 Root length density (RLD), root length distribution, and vegetation cover RLD in grasslands was 3.6–5.5 times higher than in croplands, with the highest values observed in extensive meadows and lowest values in organic farming (Fig. 6 a). From 0–15 cm to 15-30cm and 30–50 cm soil depth, RLD in grasslands were 3.0–12.0 times, 2.7–7.4 times and 1.6–3.4 times as high as in croplands (Fig. 6 b, Table S2-3). Across all land-use types, RLD decreased with increasing soil depth at similar proportions. A significant interaction of land use and year suggested interannual variability in RLD, particularly in croplands at 0–30 cm depths (Fig. 6 b). Compared to 2022, the RLD in 2023 increased by 69.9% in conventional farming at 0–15 cm depth, 102.9% and 166.2% in organic farming at 0–15 cm and 15–30 cm depth, but decreased by 32.3% in intensive meadow at 0–15 cm depth ( p < 0.01, Fig. 6 b, Table S2-3). RLD was greater in grasslands than in croplands for all root diameter classes, but especially for fine roots (< 0.2 mm) (Fig. 6 c, S5). Fine roots (< 0.1 mm) were the most abundant diameter class in intensive meadows, whereas other land-use types had as many or more roots in the 0.1–0.2 mm diameter range. Notably, the length density of roots with 0.1–0.3 mm diameter range were significantly higher in extensive meadows and pastures than in intensive meadows (Fig. 6 c). Across all three grassland types, the total vegetation cover of the soil surface was higher in 2023 than in 2022 (Fig. 6 d, Table S2-3). The relative abundance of plant functional groups (grass, forb, legume) and functional group dependence varied both across land-use types and between years (Fig. 6 e, Table S2-3). The higher total vegetation cover for extensive meadows and pastures in 2023 as compared to 2022, was mainly caused by a higher relative abundance (Fig. 6 e) and summed cover per functional group (Fig. S6) of legumes. Furthermore, the future climate scenario had no significant effect on overall RLD (0–50 cm), RLD at individual depth intervals, root length distribution, total vegetation cover, and plant functional group dependence (Table S2). 3.3 Soil water depletion and root water uptake patterns Changes in both topsoil and deep soil water storage during the growing season significantly varied across land-use types and years (Fig. S7, Table S2, S4). In both the topsoil (0–30 cm) and deep soil layer (30–110 cm), croplands generally experienced greater water depletion than grasslands, particularly in the drought years of 2022 and 2023. The highest topsoil water depletion occurred in 2023, while deep soil water storage depletion peaked in 2022, suggesting pronounced deep water uptake during the extreme drought in 2022. In contrast, soil water storage in 2024 showed minimal change or slight increases across all land-use types. The relationship between corrected RLD and net change in water content ( P + ΔS) in the same depth profiles (0–50 cm) during the period of peak growth (eight weeks centered around root sampling in May) differed between croplands and grasslands in both years (Fig. 7 ). In croplands, a strong negative correlation (R 2 = 0.53, P < 0.001) emerged in 2022 (dry period), but disappeared in 2023 (wet period), indicating that the annual crops relied on the limited shallow soil water storage in 2022, but decoupled the water uptake from the actual RLD under sufficient water supply in 2023. In contrast, perennial grasslands with an extensive root system showed no correlation between shallow RLD and net change in water content at 0–50 cm depth in the dry period of 2022, perhaps because of a shift toward deep root water uptake. A negative correlation (R 2 = 0.33, P < 0.001) for grasslands emerged in 2023, when the initial soil water contents in the topsoil were higher (close to field capacity, Fig. 5 b) and shallow soil water storage change was greater (Fig. 7 ). 3.4 Yield, evapotranspiration and water use efficiency Annual yield across all land-use types increased consistently with water supply from 2022 to 2024 (Fig. S8). Using the normal precipitation regime in 2024 as a reference point, the yield drop in the severe drought year of 2022 was found to be 27.8%, 30.0%, 32.6% and 34.5% for conventional farming, organic farming, intensive meadow and extensive meadow, respectively. Despite interannual variability, the yield differences among land-use types remained stable: conventional farming > intensive meadow > organic farming > extensive meadow. Conventional farming produced 55.40-136.50% higher yields than organic farming, while intensive meadow yielded 61.69-100.61% more plant biomass than extensive meadow. Yield increased with increasing evapotranspiration (ET) during the growing season across all land-use types and both climate scenarios (Fig. 8 a), reflecting interannual changes in drought intensity (Fig. 2 a). Interannual variability in ET was evident. In the absence of drought in 2024, ET was highest on average, regardless of land use. It was 9.5% lower under the future climate scenario than under the ambient climate and this gap was consistent across land-use types, indicating that it is mainly caused by the precipitation shielding in summer, when ET demand is highest. During the drought years of 2022 and 2023 (yield in dry periods of 2023 before heavy rainfall in August), ET was lower on average. It was higher under conventional (and organic) farming compared to grasslands. Consistent trends imposed by the future climate scenario vanished. Water use efficiency (WUE) varied significantly across land-use types and years, with 17.6–54.7% higher WUEs in croplands than in grasslands (Fig. 8 b, Table S2, S4). Conventional farming consistently exhibited the highest WUE (4.48–6.60 kg m − 3 ), whereas extensive meadows had the lowest (1.89–2.78 kg m − 3 ), suggesting that agricultural intensification increases WUE. WUE in intensive meadows was 65.3–87.4% higher than in extensive meadows ( p < 0.05, Table S2, S4). The drought years of 2022 and 2023 caused an increased water use efficiency compared to the normal year of 2024 in all land uses except for the white clover in the organic farming crop rotation of 2023. The future climate scenario only affected WUE under conventional farming, increasing it by 23.6% in 2023 and decreasing it by 19.6% in 2024 compared to ambient climate ( p < 0.05, Table S2, S4). 3.5 Relationship between soil water storage and plant productivity The SEM revealed that SPEI during the growing season significantly influenced both topsoil and deep soil water storage, although its effect was comparatively weaker on deep soil water storage because of legacy effects (Fig. 9 ). In contrast, land use and the future climate scenario exhibited smaller our even negligible effects on water storage terms, respectively (paths removed in the final model). WUE was directly and strongly affected by land use and SPEI, with small indirect effects through water storage terms. Yield was closely associated with WUE, directly and strongly affected by land use, but the strong SPEI effect on yield was clearly mediated by both water storage terms. These combined observations indicated that yield was mainly driven by water availability and the trade-off between growth and maintenance. The model explained a substantial proportion of the variance in WUE (R² = 0.76), yield (R² = 0.81) and topsoil water storage during the growing season (R² = 0.81), all of which exceeded the explained variance in deep soil water storage (R² = 0.45) suggesting again the impact of legacy effects on deep water cycling. 4. DISCUSSION 4.1 Effect of climate extremes on topsoil and deep soil water storage Our results supported the first hypothesis that topsoil water dynamics were primarily controlled by atmospheric forcing in terms of radiation, temperature, and precipitation, and were closely coupled to short-term climatic variability. Shallow soil water content at 0–20 cm depth was tightly linked to the SPEI throughout a three-year monitoring period with two extended droughts (2022, 2023), a heavy summer rain event (2023) and a normal year (2024), indicating that the shallow soil water storage closely tracked precipitation inputs and evapotranspiration (ET) losses, with a rapid turnover and limited buffering capacity under changing climatic conditions (Good et al., 2015 ; Warter et al., 2021 ). In contrast, deep soil water exhibited a stronger “memory” effect on climate variation, particularly during extreme drought events. This finding further supported our hypothesis that deep soil water storage is governed by the legacy effect of past soil moisture conditions. Prolonged periods of negative SPEI led to persistent depletion of deep soil water storage, resulting in an inversion of the soil moisture profiles with depth (soil water content under equilibrium would normally increase with depth due to gravity) that reflected the legacy effects of droughts, while the shallow soil water fluctuated around higher levels following moderate rain events that failed to induce deep infiltration. Notably, infiltration fronts required approximately five months in 2022 and two months in 2023 to reach 100 cm depth after drought, illustrating the delayed replenishment of deep soil moisture. This delay is not only governed by the low unsaturated hydraulic conductivity of dry soils but is further exacerbated by concurrent plant water uptake during infiltration, which reduces the amount of water available for percolation (Bens et al., 2006 ; Rickard et al., 2025 ). The persistence of drought in deep soil layers restricts plant water access, delaying phenological development, and influencing ecosystem productivity (Wu et al., 2018 ; Bastos et al., 2020 ; Sun et al., 2022 ). Furthermore, extreme droughts pushed deep soil water content toward the plant wilting point for extended periods, while topsoil water storage recovered more quickly following precipitation (Fig. 2 b, 3 b). These patterns aligned with a global modeling study demonstrating increased drying of deep soil layers (> 20 cm) during the growing season under a future climate scenario, although not specific to our study site (Schlaepfer et al., 2017 ). As drought extremes intensify, understanding how rooting strategies influence soil water use becomes increasingly critical. Deep rooting can confer drought resistance by accessing subsoil water (Lynch, 2013 ; Fan et al., 2017 ), but its effectiveness remains debated due to high metabolic costs and limited deep soil water storage (Palta and Turner, 2018 ; Rasmussen et al., 2019 ; Figueroa-Bustos et al., 2020 ). The effectiveness of the rooting strategy is highly context-dependent, influenced by field conditions and land-use systems (Lynch, 2019 ; van der Bom et al., 2020 ; Li et al., 2022 ). If the unsaturated hydraulic conductivity of the bulk soil is high, like for the silt loam at this site, then water flow towards the roots sustains efficient root water uptake, reducing the need for extensive root systems (Schlüter et al., 2013 ; Jorda et al., 2022 ). Additionally, a heavy rain event in August 2023 substantially replenished deep soil water (Fig. 2 b), offering a long-lasting storage for subsequent dry spells (Feldman et al., 2024 ). However, this recharge occurred post-harvest for the croplands (outside the main growing season), limiting its immediate benefit for crop productivity. In grasslands, although harvest timing is more flexible and depends on the biomass accumulation (typically one or two cuts in extensive meadow, and up to three in intensive meadow), late-summer rainfall was also unlikely to substantially improve productivity. This is probably because the peak growth of grasses generally occurs in late spring during flowering, and later in the season, declining photoperiod and photosynthesis limit further growth (Brookshire and Weaver, 2015 ). Thus, deep water storage remains a vital buffer in upland soils without groundwater access, mitigating mismatches between water supply and plant demand. 4.2 Effect of land use and future climate scenario on deep water storage Croplands significantly reduced both the extent and duration of water shortage in deep soil compared to grasslands, primarily due to the 3–4 months bare soil phase after harvest in croplands, reducing water loss by transpiration, as compared to a longer period of ET losses in grasslands through the continuous transpiration of perennial plant species (Spera et al., 2016 ; Wu et al., 2021 ). Our results showed that croplands had 4.0-31.9% higher average soil water content than grasslands across a 10–110 cm depth profile. This finding primarily supported the second hypothesis that croplands maintain a higher soil water content than grasslands on an annual basis. These results are in accordance with previous studies showing that the conversion of cropland to grassland primarily lead to reduced soil moisture (Gao et al., 2011 ; Wang et al., 2024b ). Another reason for higher soil water content in croplands may be reduced rainfall interception due to sparser and seasonal variable canopies, which allow for more efficient precipitation infiltration into the topsoil, especially during early growth stages (Lin et al., 2020 ; Lian et al., 2022 ). The third reason may be the lower root water uptake on the annual basis in croplands than grasslands due to the lower root length density (Bayala and Prieto, 2019 ), which might be more critical in coarse textured soil with less efficient water flow through soil due to low unsaturated hydraulic conductivity. Notably, the differences in soil water content among different land-use types became increasingly pronounced with depth and were especially evident during dry years, further supporting the legacy effects of deep soil water storage discussed earlier and suggesting that soil water “legacy effects” following droughts are influenced by both vegetation phenology and water use patterns (Zeng et al., 2021 ). Furthermore, land-use intensity significantly influences the deep soil water storage, particularly during the climate extremes. The organic farming without mineral fertilization consistently exhibited higher soil water content in deep soil layer compared to conventional farming (highest land-use intensity), particularly during the dry years of 2022 and 2023, likely because nutrient limitations constrained plant growth and thus reduced transpiration and overall ET demand (Barton et al., 2009 ). In grassland systems, intensive meadow with fertilization, only four forage grass species, and frequent mowing showed lower soil water levels in the topsoil but significantly and consistently higher soil water content in the deeper layers than unfertilized extensive meadows and pastures (lowest land-use intensity). This contrast may be explained by a combination of factors: reduced rooting depth due to early stage of root system development, rapid shoot re-growth after frequent mowing events, and regular fertilization that promote aboveground biomass production and increase water uptake from shallow layers at initially low transpiration rates (Rose et al., 2011 ; Prechsl et al., 2015 ; Fan et al., 2017 ). Additionally, limited canopy cover after mowing reduces interception surface shading, while low plant diversity constrains vertical ET buffering (He and Richards, 2015 ). This pattern likely also reflects the relatively recent establishment of the intensive meadow, which was ploughed and re-sown in 2020, and may not yet have developed fully functional and deeper rooting networks characteristic of older, undisturbed grassland (Korell et al., 2024 ). Notably, the differences in the deep soil water storage between intensive meadow and extensive grasslands were amplified following the heavy rainfall event in August 2023, with the discrepancy becoming even stronger under the wettest conditions in 2024. The capacity of land-use systems to retain or deplete deep soil water following recharge events depend on vegetation community composition and root distribution, which in turn affect infiltration pathways and soil hydraulic properties (Fischer et al., 2014 ; Fischer et al., 2014b ; Cui et al., 2019 ; Shi et al., 2021 ). Thus, our findings suggest that both land-use types and intensities modulated the capacity to access and retain deep water, with croplands and intensive grasslands retaining more water than extensive grasslands due to lower transpiration and shallower rooting. Our results indicate that the future climate scenario further modulated deep soil water storage through interactions with land-use types, especially during the extreme events, by the altered precipitation regimes and increased temperature. Under conventional farming, the future climate scenario exhibited higher deep soil water content than ambient climate when the legacy effect of drought was most dominant (January to June 2023), likely due to the added irrigation during fall and spring, which allowed deeper infiltration and water retention when atmospheric demand was low (Peterson and Westfall, 2005 ). In contrast, intensive meadow under the future scenario exhibited lower deep soil water content following heavy rain events (August 2023 and June 2024), as 20% of summer rain events were shielded and large events cause large absolute differences. The future climate scenario effect might be greater in intensive meadow than in extensive meadow, probably due to the dominance of shallow-rooted grass species in these systems, which more rapidly depleted surface moisture but were less effective at utilizing or retaining water at depth. Additionally, lower plant diversity and frequent mowing could have influenced water use patterns, leading to less effective recharge in deeper layer under high-intensity rainfall (Rodriguez-Iturbe et al., 2001 ). These patterns highlight the importance of both seasonal deep soil water dynamics and system-specific water use strategies in mediating land-use responses to altered precipitation regimes. Deep soil water storage acts as a critical buffer during climatic extremes, but its effectiveness depends heavily on land-use practices, rooting depth distributions and vegetation composition. 4.3 Responses of plant productivity on climate extreme and land-use-based adjustment strategies Our results supported the third hypothesis: croplands generally produced 47.0-63.5% higher yields than grasslands, reflecting crop breeding for high resource use efficiency and rapid aboveground biomass production within a short growth period (Leakey et al., 2019 ). In contrast, grasslands comprising perennial and diverse species, exhibited extended growth periods and lower yield optimization. Productivity was further influenced by land-use intensity, with the lowest yields observed in extensive meadows across years, likely due to lower initial water content and the absence of fertilization. Furthermore, yield was modulated by interannual climate variability and its interaction with land-use systems. Highest yields occurred in the normal year 2024, while significant declines were observed in the drought year 2022 and to a lesser extent in 2023. Notably, the decrease of yields in 2022 relative to 2024 was similar (~ 30%) across croplands and grasslands, indicating that the extensive grasslands did not exhibit higher resistance to climate extremes. This contradicts our hypothesis and previous findings from the same field site (Isbell et al., 2015 ; Korell et al., 2024 ). That is probably due to the differences in the study period (2015–2022), the intensive meadow system being established in 2020, and the fact that the yield analyzed in the current study was based on machine harvesting, as is usual in agriculture. In contrast, Korell et al. ( 2024 ) analyzed manually harvested biomass cut close to the ground level, which also included the pastures. Deep soil water storage partially buffered yield losses against drought extremes. In 2022, the pronounced decline in deep soil water storage indicated its use to maintain productivity during extreme drought. In 2023, despite a summer drought, yields were higher, likely due to uptake of topsoil water and sustained by the infiltration of winter and spring precipitation, with the wetting front reaching the deep soil by March, and supplemented by late-season rainfall recharge. In contrast, the normal conditions in 2024 eliminated the need to access deeper water reservoirs. Our SEM revealed that extreme climates exerted indirect effect on yield through changes in both topsoil and deep soil water storage during the growing season. These findings suggested that deep soil water provides critical buffering, but is insufficient to fully compensate for severe topsoil moisture deficits, particularly under prolonged drought. Additional coping mechanisms, such as shifts in root patterns, plant community composition, and improved WUE are essential under extreme drought (Zwicke et al., 2015 ; Qi et al., 2018 ). We observed positive correlation between RLD and net change in water content during peak growth, emerging under grasslands in 2023 and croplands in 2022, indicating that root water uptake shifted between topsoil and deep layer depending on drought severity (Zwicke et al., 2015 ; Bristiel et al., 2018 ). In grasslands, deep-rooted perennial species, especially the forbs, likely facilitate deep soil water uptake during extreme drought through their dense and persistent root system networks (Barkaoui et al., 2016 ; Hanslin et al., 2019 ; Künzi et al., 2025 ). However, deeper rooting may not be the dominant adjustment mechanism, as efficient water uptake can occur with sparse roots, and high RLD is often more critical for nutrient acquisition (Jorda et al., 2022 ). In croplands, RLD varied more across years, with higher values in the topsoil in 2023 linked to better moisture availability. Yet, yield variation likely reflects both climate-driven root responses and crop-specific traits, such as finer roots under organic management (Wong et al., 2023 ). We observed that the total vegetation cover and relative abundance of legumes slightly increased in 2023, suggesting the ability of grasslands with high species richness to undergo year-to-year changes in species composition depending on environmental conditions and species interactions. However, this interpretation is limited by several factors. RLD data only extend to a depth of 50 cm and may be influenced by the effects of crop rotation. Furthermore, vegetation community data are only available for grasslands and do not include observations from 2024. This restricts our ability to fully assess root and community-level responses. Croplands exhibited significantly higher ET than grasslands in 2022 and 2023, likely due to higher transpiration from fast-growing annual crops with greater leaf area (Rajan et al., 2014 ; Li et al., 2015 ). In 2024, ET rates became comparable across all land-use types, suggesting that increased transpiration potential in grasslands (Spera et al., 2016 ). Croplands consistently had a higher WUE than grasslands, and the WUE showed higher correlation coefficient with yield than RLD, supporting our fourth hypothesis that adaptations to drought and deep soil water limitations would rely more on WUE than on shifts in rooting patterns or plant community. This likely reflects breeding strategies that enhance biomass production, drought tolerance, and stomatal regulation (Turner, 2004 ; Sun et al., 2019 ; Haworth et al., 2021 ; Zhang et al., 2023 ). In contrast, deep-rooted and drought-resilient grassland perennials often show higher ET losses and lower yield responsiveness, as well as the water loss during maintenance without growth, may contribute to lower WUE (Ponton et al., 2006 ; Poppe Terán et al., 2023 ). Furthermore, extensive grasslands include a high proportion of perennial species capable of clonal regeneration through rhizomes or other belowground tissues, which can sustain transpiration without proportional aboveground biomass gain (Volaire, 2018 ). Notably, WUE was 65.3–87.4% higher in intensive meadow than extensive meadow, probably due to the high frequency of mowing, and the application of fertilizer promoting aboveground growth during the peak productivity (Rose et al., 2011 ). The future climate scenario showed limited effect on yield but altered water cycling. In 2024, yields were unaffected by the reduced in precipitation, likely due to other limiting factors such as nutrients regulating plant growth during this normal year (Basso and Ritchie, 2018 ; Fu et al., 2022 ). During the drought years, legacy effects of additional irrigation in autumn and spring became more relevant, causing greater deep soil water storage, which elevated the ET in the growing season to levels comparable to ambient climate, rendering its effect on yield again negligible. However, root biomass and plant community composition remained unchanged, possibly due to shallow sampling, moderate climatic changes, and ecosystem buffering capacity (Kühn et al., 2022 ). Overall, while deep soil water storage provided partial drought buffering, WUE optimization emerges as the dominant adjustment strategy under climate extremes. High-intensity systems, such as conventional farming, maintain productivity through efficient water use and flexible responses to future scenarios. These findings highlight the need for adaptive management to sustain productivity under intensifying climate variability. 5. CONCLUSION Our study highlights the role of deep soil water storage in buffering plant productivity against climate extremes. Unlike shallow soil moisture, the storage of water in deeper soil responds more slowly to climatic fluctuations. This can help to sustain the supply of water to plants and their productivity during prolonged droughts. However, this buffering capacity is not unlimited. Extreme droughts can thoroughly exploit deep storage, particularly when topsoil moisture is severely depleted. This drives deep soil water content toward the plant wilting point, resulting in a situation that persists until the subsequent year. Conventional farming can enhance deep soil water uptake during drought extremes by proper root systems and increased fertilization, whereas intensive meadow management was beneficial to deep soil water retention through higher mowing frequency. These practices enhance water uptake efficiency and help maintain productivity under extreme conditions. The future climate scenarios had weaker impact on the water cycle than climate extremes. Management strategies such as optimizing water use efficiency and promoting species or cultivars with appropriate growth spans and rooting systems that align with seasonal water availability can help stabilize productivity. These findings underscore the importance of deep soil water storage and context-specific management practices for maintaining ecosystem resistance and sustainability in an increasingly variable climate. Declarations DECLARATION OF COMPETING INTEREST The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ACKNOWLEDGEMENTS We appreciate the Helmholtz Association, the Federal Ministry of Education and Research, the State Ministry of Science and Economy of Saxony-Anhalt and the State Ministry for Higher Education, Research and the Arts Saxony to fund the Global Change Experimental Facility (GCEF) project. We thank the staff of the Bad Lauchstädt Experimental Research Station (especially Ines Merbach and Konrad Kirsch) for their work in in maintaining the plots and infrastructures of the GCEF, and Harald Auge, François Buscot, and Stefan Klotz for their role in setting up the GCEF. We thank Ralf Gruendling, Max Koehne, and Eric Braatz from the UFZ, Tobias Klauder at the Technical University of Dresden, Vanessa Schwanengel, Jakob Apelt and Elin Briese at the Martin Luther University Halle-Wittenberg for assistance during field work and laboratory analyses. AUTHORSHUIP CONTRIBUTION STATAMENT Mengqi Wu : Writing – original draft, Visualization, Validation, Investigation, Formal analysis, Data curation, Methodology. Christiane Roscher : Writing – review & editing, Investigation, Data curation, Validation, Methodology. Martin Schädler : Writing – review & editing, Investigation, Data curation, Validation. Mika Tarkka : Writing – review & editing, Methodology, Conceptualization. Doris Vetterlein : Writing – review & editing, Methodology, Conceptualization. Steffen Schlüter : Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. References Ahmed, M.A., Zarebanadkouki, M., Meunier, F., Javaux, M., Kaestner, A., Carminati, A., 2018. Root type matters: measurement of water uptake by seminal, crown, and lateral roots in maize. Journal of Experimental Botany 69, 1199–1206. Ali, G., Wang, Z., Li, X., Jin, N., Chu, H., Jing, L., 2021. Deep soil water deficit and recovery in alfalfa fields of the Loess Plateau of China. Field Crops Research 260, 107990. Altermann, M., Rinklebe, J., Merbach, I., Körschens, M., Langer, U., Hofmann, B., 2005. Chernozem—Soil of the Year 2005. Journal of Plant Nutrition and Soil Science 168, 725–740. Bagley, J.E., Kueppers, L.M., Billesbach, D.P., Williams, I.N., Biraud, S.C., Torn, M.S., 2017. The influence of land cover on surface energy partitioning and evaporative fraction regimes in the U.S. Southern Great Plains. Journal of Geophysical Research: Atmospheres 122, 5793–5807. Barkaoui, K., Roumet, C., Volaire, F., 2016. Mean root trait more than root trait diversity determines drought resilience in native and cultivated Mediterranean grass mixtures. Agriculture, Ecosystems & Environment 231, 122–132. Barton, L., Wan, G.G.Y., Buck, R.P., Colmer, T.D., 2009. Nitrogen Increases Evapotranspiration and Growth of a Warm-Season Turfgrass. Agronomy Journal 101, 17–24. Basso, B., Ritchie, J.T., 2018. Evapotranspiration in High-Yielding Maize and under Increased Vapor Pressure Deficit in the US Midwest. Agricultural & Environmental Letters 3, 170039. Bastos, A., Ciais, P., Friedlingstein, P., Sitch, S., Pongratz, J., Fan, L., Wigneron, J.P., Weber, U., Reichstein, M., Fu, Z., Anthoni, P., Arneth, A., Haverd, V., Jain, A.K., Joetzjer, E., Knauer, J., Lienert, S., Loughran, T., McGuire, P.C., Tian, H., Viovy, N., Zaehle, S., 2020. Direct and seasonal legacy effects of the 2018 heat wave and drought on European ecosystem productivity. Science Advances 6, eaba2724. Bates, D., 2005. Fitting linear mixed models in R. R News 5, 27–30. Bayala, J., Prieto, I., 2019. Water acquisition, sharing and redistribution by roots: applications to agroforestry systems. Plant and Soil 453, 17–28. Bazzichetto, M., Sperandii, M.G., Penone, C., Keil, P., Allan, E., Lepš, J., Prati, D., Fischer, M., Bolliger, R., Gossner, M.M., de Bello, F., 2024. Biodiversity promotes resistance but dominant species shape recovery of grasslands under extreme drought. Journal of Ecology 112, 1087–1100. Beguería, S., Vicente-Serrano, S.M., Maintainer Santiago, B., 2017. Package ‘spei’. Calculation of the Standardised Precipitation-Evapotranspiration Index. Beguería, S., Vicente-Serrano, S.M., Reig, F., Latorre, B., 2014. Standardized precipitation evapotranspiration index (SPEI) revisited: parameter fitting, evapotranspiration models, tools, datasets and drought monitoring. International Journal of Climatology 34, 3001–3023. Bens, O., Wahl, N.A., Fischer, H., Hüttl, R.F., 2006. Water infiltration and hydraulic conductivity in sandy cambisols: impacts of forest transformation on soil hydrological properties. European Journal of Forest Research 126, 101–109. Bristiel, P., Roumet, C., Violle, C., Volaire, F., 2018. Coping with drought: root trait variability within the perennial grass Dactylis glomerata captures a trade-off between dehydration avoidance and dehydration tolerance. Plant and Soil 434, 327–342. Brookshire, E.N., Weaver, T., 2015. Long-term decline in grassland productivity driven by increasing dryness. Nature Communications 6, 7148. Carminati, A., Zarebanadkouki, M., Kroener, E., Ahmed, M.A., Holz, M., 2016. Biophysical rhizosphere processes affecting root water uptake. Annals of Botany 118, 561–571. Chen, L., Brun, P., Buri, P., Fatichi, S., Gessler, A., McCarthy, M.J., Pellicciotti, F., Stocker, B., Karger, D.N., 2025. Global increase in the occurrence and impact of multiyear droughts. Science 387, 278–284. Condon, A.G., Richards, R.A., Rebetzke, G.J., Farquhar, G.D., 2004. Breeding for high water-use efficiency. Journal of Experimental Botany 55, 2447–2460. Craine, J.M., Ocheltree, T.W., Nippert, J.B., Towne, E.G., Skibbe, A.M., Kembel, S.W., Fargione, J.E., 2012. Global diversity of drought tolerance and grassland climate-change resilience. Nature Climate Change 3, 63–67. Cui, Z., Wu, G.-L., Huang, Z., Liu, Y., 2019. Fine roots determine soil infiltration potential than soil water content in semi-arid grassland soils. Journal of Hydrology 578, 124023. Döscher, R.W.N., U.; Jones, C.; Rutgersson, A.; Meier, H. E. M.; Hansson, U., 2002. The Development of The Coupled Ocean-atmosphere Model Rcao For The Northern European Area. EGS XXVII General Assembly, Nice. Draye, X., Kim, Y., Lobet, G., Javaux, M., 2010. Model-assisted integration of physiological and environmental constraints affecting the dynamic and spatial patterns of root water uptake from soils. Journal of Experimental Botany 61, 2145–2155. Fan, J., Wang, Q., Jones, S.B., Shao, M., 2015. Soil water depletion and recharge under different land cover in China's Loess Plateau. Ecohydrology 9, 396–406. Fan, Y., Miguez-Macho, G., Jobbagy, E.G., Jackson, R.B., Otero-Casal, C., 2017. Hydrologic regulation of plant rooting depth. Proceedings of the National Academy of Sciences 114, 10572–10577. Feldman, A.F., Feng, X., Felton, A.J., Konings, A.G., Knapp, A.K., Biederman, J.A., Poulter, B., 2024. Plant responses to changing rainfall frequency and intensity. Nature Reviews Earth & Environment 5, 276–294. Figueroa-Bustos, V., Palta, J.A., Chen, Y., Stefanova, K., Siddique, K.H.M., 2020. Wheat Cultivars With Contrasting Root System Size Responded Differently to Terminal Drought. Frontiers in Plant Science 11, 1285. Fischer, C., Roscher, C., Jensen, B., Eisenhauer, N., Baade, J., Attinger, S., Scheu, S., Weisser, W.W., Schumacher, J., Hildebrandt, A., 2014. How do earthworms, soil texture and plant composition affect infiltration along an experimental plant diversity gradient in grassland? PLoS One 9, e98987. Fischer, C., Tischer, J., Roscher, C., Eisenhauer, N., Ravenek, J., Gleixner, G., Attinger, S., Jensen, B., de Kroon, H., Mommer, L., Scheu, S., Hildebrandt, A., 2014b. Plant species diversity affects infiltration capacity in an experimental grassland through changes in soil properties. Plant and Soil 397, 1–16. Fu, Z., Ciais, P., Feldman, A.F., Gentine, P., Makowski, D., Prentice, I.C., Stoy, P.C., Bastos, A., Wigneron, J.P., 2022. Critical soil moisture thresholds of plant water stress in terrestrial ecosystems. Science Advances 8, eabq7827. Gao, X., Wu, P., Zhao, X., Wang, J., Shi, Y., 2011. Effects of Land Use on Soil Moisture Variations in a Semi-Arid Catchment: Implications for Land and Agricultural Water Management. Land Degradation & Development 25, 163–172. Good, S.P., Noone, D., Bowen, G., 2015. WATER RESOURCES. Hydrologic connectivity constrains partitioning of global terrestrial water fluxes. Science 349, 175–177. Hanslin, H.M., Bischoff, A., Hovstad, K.A., 2019. Root growth plasticity to drought in seedlings of perennial grasses. Plant and Soil 440, 551–568. Hargreaves, G.H., Allen, R.G., 2003. History and evaluation of Hargreaves evapotranspiration equation. Journal of Irrigation and Drainage Engineering 129, 53–63. Haworth, M., Marino, G., Loreto, F., Centritto, M., 2021. Integrating stomatal physiology and morphology: evolution of stomatal control and development of future crops. Oecologia 197, 867–883. He, S., Richards, K., 2015. Impact of Meadow Degradation on Soil Water Status and Pasture Management—A Case Study in Tibet. Land Degradation & Development 26, 468–479. Holmes, T.H., Rice, K.J., 1996. Patterns of growth and soil-water utilization in some exotic annuals and native perennial bunchgrasses of California. Annals of Botany 78, 233–243. Isbell, F., Craven, D., Connolly, J., Loreau, M., Schmid, B., Beierkuhnlein, C., Bezemer, T.M., Bonin, C., Bruelheide, H., de Luca, E., Ebeling, A., Griffin, J.N., Guo, Q., Hautier, Y., Hector, A., Jentsch, A., Kreyling, J., Lanta, V., Manning, P., Meyer, S.T., Mori, A.S., Naeem, S., Niklaus, P.A., Polley, H.W., Reich, P.B., Roscher, C., Seabloom, E.W., Smith, M.D., Thakur, M.P., Tilman, D., Tracy, B.F., van der Putten, W.H., van Ruijven, J., Weigelt, A., Weisser, W.W., Wilsey, B., Eisenhauer, N., 2015. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574–577. Jacob, D., Podzun, R., 1997. Sensitivity studies with the regional climate model REMO. Meteorology and Atmospheric Physics 63, 119–129. Jorda, H., Ahmed, M.A., Javaux, M., Carminati, A., Duddek, P., Vetterlein, D., Vanderborght, J., 2022. Field scale plant water relation of maize (Zea mays) under drought – impact of root hairs and soil texture. Plant and Soil 478, 59–84. Kano-Nakata, M., Nakamura, T., Mitsuya, S., Yamauchi, A., 2019. Plasticity in root system architecture of rice genotypes exhibited under different soil water distributions in soil profile. Plant Production Science 22, 501–509. Korell, L., Andrzejak, M., Berger, S., Durka, W., Haider, S., Hensen, I., Herion, Y., Hofner, J., Kindermann, L., Klotz, S., Knight, T.M., Linstadter, A., Madaj, A.M., Merbach, I., Michalski, S., Plos, C., Roscher, C., Schadler, M., Welk, E., Auge, H., 2024. Land use modulates resistance of grasslands against future climate and inter-annual climate variability in a large field experiment. Global Change Biology 30, e17418. Kühn, N., Spiegel, M.P., Tovar, C., Willis, K.J., Macias-Fauria, M., 2022. Seeing roots from space: aboveground fingerprints of root depth in vegetation sensitivity to climate in dry biomes. Environmental Research Letters 17, 114062. Künzi, Y., Zeiter, M., Fischer, M., Stampfli, A., 2025. Rooting depth and specific leaf area modify the impact of experimental drought duration on temperate grassland species. Journal of Ecology 113, 445–458. Leakey, A.D.B., Ferguson, J.N., Pignon, C.P., Wu, A., Jin, Z., Hammer, G.L., Lobell, D.B., 2019. Water Use Efficiency as a Constraint and Target for Improving the Resilience and Productivity of C(3) and C(4) Crops. Annual Review of Plant Biology 70, 781–808. Lefcheck, J., Byrnes, J., Grace, J., 2015. piecewiseSEM: Piecewise Structural Equation Modeling. R package version 1.1. Lenth, R., Henrik, S., Jonathon, L., Paul, B., Maxime, H., 2018. Emmeans: Estimated marginal means. AKA least-squares means 1. Li, B., Zhang, X., Morita, S., Sekiya, N., Araki, H., Gu, H., Han, J., Lu, Y., Liu, X., 2022. Are crop deep roots always beneficial for combating drought: A review of root structure and function, regulation and phenotyping. Agricultural Water Management 271, 107781. Li, Y., Fan, J., Hu, Z., Shao, Q., Zhang, L., Yu, H., 2015. Influence of Land Use Patterns on Evapotranspiration and Its Components in a Temperate Grassland Ecosystem. Advances in Meteorology 2015, 1–12. Li, Y., Li, K., Zhou, Q., Zhao, Y., Cai, L., Yang, Z., 2024. Spatiotemporal dynamics and similarity in soil moisture in shallow soils on karst slopes. Journal of Hydrology 639, 131655. Lian, X., Zhao, W., Gentine, P., 2022. Recent global decline in rainfall interception loss due to altered rainfall regimes. Nature Communications 13, 7642. Lin, M., Sadeghi, S.M.M., Van Stan, J.T., 2020. Partitioning of Rainfall and Sprinkler-Irrigation by Crop Canopies: A Global Review and Evaluation of Available Research. Hydrology 7, 76–89. Liu, Y., Zhang, Y., Peñuelas, J., Kannenberg, S.A., Gong, H., Yuan, W., Wu, C., Zhou, S., Piao, S., 2025. Drought legacies delay spring green-up in northern ecosystems. Nature Climate Change 15, 444–451. Lynch, J.P., 2013. Steep, cheap and deep: an ideotype to optimize water and N acquisition by maize root systems. Annals of Botany 112, 347–357. Lynch, J.P., 2019. Root phenotypes for improved nutrient capture: an underexploited opportunity for global agriculture. New Phytologist 223, 548–564. Maestre, F.T., Eldridge, D.J., Soliveres, S., Kefi, S., Delgado-Baquerizo, M., Bowker, M.A., Garcia-Palacios, P., Gaitan, J., Gallardo, A., Lazaro, R., Berdugo, M., 2016. Structure and functioning of dryland ecosystems in a changing world. Annual Review of Ecology, Evolution, and Systematics 47, 215–237. Maurel, C., Nacry, P., 2020. Root architecture and hydraulics converge for acclimation to changing water availability. Nature Plants 6, 744–749. Muller, L.M., Bahn, M., 2022. Drought legacies and ecosystem responses to subsequent drought. Global Change Biology 28, 5086–5103. Palta, J.A., Turner, N.C., 2018. Crop root system traits cannot be seen as a silver bullet delivering drought resistance. Plant and Soil 439, 31–43. Peterson, G.A., Westfall, D.G., 2005. Managing precipitation use in sustainable dryland agroecosystems. Annals of Applied Biology 144, 127–138. Phalempin, M., Jentzsch, N., Köhne, J.M., Schreiter, S., Gründling, R., Vetterlein, D., Schlüter, S., 2025. Soil structure development in a five-year chronosequence of maize cropping on two contrasting soil textures. Soil and Tillage Research 251, 106561. Pierret, A., Maeght, J.L., Clement, C., Montoroi, J.P., Hartmann, C., Gonkhamdee, S., 2016. Understanding deep roots and their functions in ecosystems: an advocacy for more unconventional research. Annals of Botany 118, 621–635. Ponton, S., Flanagan, L.B., Alstad, K.P., Johnson, B.G., Morgenstern, K.A.I., Kljun, N., Black, T.A., Barr, A.G., 2006. Comparison of ecosystem water-use efficiency among Douglas‐fir forest, aspen forest and grassland using eddy covariance and carbon isotope techniques. Global Change Biology 12, 294–310. Poppe Terán, C., Naz, B.S., Graf, A., Qu, Y., Hendricks Franssen, H.-J., Baatz, R., Ciais, P., Vereecken, H., 2023. Rising water-use efficiency in European grasslands is driven by increased primary production. Communications Earth & Environment 4, 95. Prechsl, U.E., Burri, S., Gilgen, A.K., Kahmen, A., Buchmann, N., 2015. No shift to a deeper water uptake depth in response to summer drought of two lowland and sub-alpine C(3)-grasslands in Switzerland. Oecologia 177, 97–111. Qi, J., Markewitz, D., Radcliffe, D., 2018. Modelling the effect of changing precipitation inputs on deep soil water utilization. Hydrological Processes 32, 672–686. Rajan, N., Maas, S.J., Cui, S., 2014. Extreme drought effects on summer evapotranspiration and energy balance of a grassland in the Southern Great Plains. Ecohydrology 8, 1194–1204. Rasmussen, C.R., Thorup-Kristensen, K., Dresbøll, D.B., 2019. Uptake of subsoil water below 2 m fails to alleviate drought response in deep-rooted Chicory (Cichorium intybus L.). Plant and Soil 446, 275–290. Rickard, W., Hossain, I., Zhang, X., Cooper, H.V., Mooney, S.J., Hawkesford, M.J., Whalley, W.R., 2025. Field plants strategically regulate water uptake from different soil depths by spatiotemporally adjusting their radial root hydraulic conductivity. New Phytologist 247, 546–561. Rockel, B., Will, A., Hense, A., 2008. The Regional Climate Model COSMO-CLM (CCLM). Meteorologische Zeitschrift 17, 347–348. Rodriguez-Iturbe, I., Porporato, A., Laio, F., Ridolfi, L., 2001. Intensive or extensive use of soil moisture: Plant strategies to cope with stochastic water availability. Geophysical Research Letters 28, 4495–4497. Rose, L., Coners, H., Leuschner, C., 2011. Effects of fertilization and cutting frequency on the water balance of a temperate grassland. Ecohydrology 5, 64–72. Schädler, M., Buscot, F., Klotz, S., Reitz, T., Durka, W., Bumberger, J., Merbach, I., Michalski, S.G., Kirsch, K., Remmler, P., Schulz, E., Auge, H., 2019. Investigating the consequences of climate change under different land-use regimes: a novel experimental infrastructure. Ecosphere 10, e02635. Scherzinger, F., Schadler, M., Reitz, T., Yin, R., Auge, H., Merbach, I., Roscher, C., Harpole, W.S., Blagodatskaya, E., Siebert, J., Ciobanu, M., Marder, F., Eisenhauer, N., Quaas, M., 2024. Sustainable land management enhances ecological and economic multifunctionality under ambient and future climate. Nature Communications 15, 4930. Schlaepfer, D.R., Bradford, J.B., Lauenroth, W.K., Munson, S.M., Tietjen, B., Hall, S.A., Wilson, S.D., Duniway, M.C., Jia, G., Pyke, D.A., Lkhagva, A., Jamiyansharav, K., 2017. Climate change reduces extent of temperate drylands and intensifies drought in deep soils. Nature Communications 8, 14196. Schlüter, S., Vogel, H.-J., Ippisch, O., Vanderborght, J., 2013. Combined Impact of Soil Heterogeneity and Vegetation Type on the Annual Water Balance at the Field Scale. Vadose Zone Journal 12, 1–17. Schnabel, F., Purrucker, S., Schmitt, L., Engelmann, R.A., Kahl, A., Richter, R., Seele-Dilbat, C., Skiadaresis, G., Wirth, C., 2022. Cumulative growth and stress responses to the 2018–2019 drought in a European floodplain forest. Global Change Biology 28, 1870–1883. Segal, E., Kushnir, T., Mualem, Y., Shani, U., 2008. Water Uptake and Hydraulics of the Root Hair Rhizosphere. Vadose Zone Journal 7, 1027–1034. Shi, X., Qin, T., Yan, D., Tian, F., Wang, H., 2021. A meta-analysis on effects of root development on soil hydraulic properties. Geoderma 403. Shipley, B., 2009. Confirmatory path analysis in a generalized multilevel context. Ecology 90, 363–368. Spera, S.A., Galford, G.L., Coe, M.T., Macedo, M.N., Mustard, J.F., 2016. Land-use change affects water recycling in Brazil's last agricultural frontier. Global Change Biology 22, 3405–3413. Sun, H., Shen, Y., Yu, Q., Flerchinger, G.N., Zhang, Y., Liu, C., Zhang, X., 2010. Effect of precipitation change on water balance and WUE of the winter wheat–summer maize rotation in the North China Plain. Agricultural Water Management 97, 1139–1145. Sun, J., Liu, W., Pan, Q., Zhang, B., Lv, Y., Huang, J., Han, X., 2022. Positive legacies of severe droughts in the Inner Mongolia grassland. Science Advances 8, eadd6249. Sun, L., Wang, R., Li, J., Wang, Q., Lyu, W., Wang, X., Cheng, K., Mao, H., Zhang, X., 2019. Reasonable fertilization improves the conservation tillage benefit for soil water use and yield of rain-fed winter wheat: A case study from the Loess Plateau, China. Field Crops Research 242, 107589. Sun, L., Wang, S., Zhang, Y., Li, J., Wang, X., Wang, R., Lyu, W., Chen, N., Wang, Q., 2018. Conservation agriculture based on crop rotation and tillage in the semi-arid Loess Plateau, China: Effects on crop yield and soil water use. Agriculture, Ecosystems & Environment 251, 67–77. Sun, W., Zhou, S., Yu, B., Zhang, Y., Keenan, T., Fu, B., 2025. Soil moisture-atmosphere interactions drive terrestrial carbon-water trade-offs. Communications Earth & Environment 6, 169. Swindon, J.G., Burke, I.C., Lauenroth, W.K., 2019. Seasonal Patterns of Root Production with Water and Nitrogen Additions Across Three Dryland Ecosystems. Ecosystems 22, 1664–1675. Tissink, M., Radolinski, J., Reinthaler, D., Venier, S., Potsch, E.M., Schaumberger, A., Bahn, M., 2025. Individual Versus Combined Effects of Warming, Elevated CO(2) and Drought on Grassland Water Uptake and Fine Root Traits. Plant Cell & Environment 48, 2083–2098. Turner, N.C., 2004. Agronomic options for improving rainfall-use efficiency of crops in dryland farming systems. Journal of Experimental Botany 55, 2413–2425. van der Bom, F.J.T., Williams, A., Bell, M.J., 2020. Root architecture for improved resource capture: trade-offs in complex environments. Journal of Experimental Botany 71, 5752–5763. Van Sundert, K., Arfin Khan, M.A.S., Bharath, S., Buckley, Y.M., Caldeira, M.C., Donohue, I., Dubbert, M., Ebeling, A., Eisenhauer, N., Eskelinen, A., Finn, A., Gebauer, T., Haider, S., Hansart, A., Jentsch, A., Kubert, A., Nijs, I., Nock, C.A., Nogueira, C., Porath-Krause, A.J., Radujkovic, D., Raynaud, X., Risch, A.C., Roscher, C., Scherer-Lorenzen, M., Schuchardt, M.A., Schutz, M., Siebert, J., Sitters, J., Spohn, M., Virtanen, R., Werner, C., Wilfahrt, P., Vicca, S., 2021. Fertilized graminoids intensify negative drought effects on grassland productivity. Global Change Biology 27, 2441–2457. Vereecken, H., Amelung, W., Bauke, S.L., Bogena, H., Brüggemann, N., Montzka, C., Vanderborght, J., Bechtold, M., Blöschl, G., Carminati, A., Javaux, M., Konings, A.G., Kusche, J., Neuweiler, I., Or, D., Steele-Dunne, S., Verhoef, A., Young, M., Zhang, Y., 2022. Soil hydrology in the Earth system. Nature Reviews Earth & Environment 3, 573–587. Vicente-Serrano, S.M., Beguería, S., López-Moreno, J.I., 2010. A Multiscalar Drought Index Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index. Journal of Climate 23, 1696–1718. Vicente-Serrano, S.M., Domínguez-Castro, F., Beguería, S., El Kenawy, A., Gimeno-Sotelo, L., Franquesa, M., Azorin-Molina, C., Andres-Martin, M., Halifa-Marín, A., 2025. Atmospheric drought indices in future projections. Nature Water 3, 374–387. Volaire, F., 2018. A unified framework of plant adaptive strategies to drought: Crossing scales and disciplines. Global Change Biology 24, 2929–2938. Wang-Erlandsson, L., Tobian, A., van der Ent, R.J., Fetzer, I., te Wierik, S., Porkka, M., Staal, A., Jaramillo, F., Dahlmann, H., Singh, C., Greve, P., Gerten, D., Keys, P.W., Gleeson, T., Cornell, S.E., Steffen, W., Bai, X., Rockström, J., 2022. A planetary boundary for green water. Nature Reviews Earth & Environment 3, 380–392. Wang, S., Fu, B.J., Gao, G.Y., Yao, X.L., Zhou, J., 2012. Soil moisture and evapotranspiration of different land cover types in the Loess Plateau, China. Hydrology and Earth System Sciences 16, 2883–2892. Wang, Y., Hu, W., Sun, H., Zhao, Y., Zhang, P., Li, Z., Zhou, Z., Tong, Y., Liu, S., Zhou, J., Huang, M., Jia, X., Clothier, B., Shao, M., Zhou, W., An, Z., 2024b. Soil moisture decline in China's monsoon loess critical zone: More a result of land-use conversion than climate change. Proceedings of the National Academy of Sciences 121, e2322127121. Wang, Z., Ma, L., Liu, X., Xue, Z., Wu, J., Wen, X., 2024a. Grass ( Poa annua L. ) cover for eight years as an effective strategy for recovering soil moisture. Geoderma 449, 117010. Warter, M.M., Singer, M.B., Cuthbert, M.O., Roberts, D., Caylor, K.K., Sabathier, R., Stella, J., 2021. Drought onset and propagation into soil moisture and grassland vegetation responses during the 2012–2019 major drought in Southern California. Hydrology and Earth System Sciences 25, 3713–3729. Weides, S.E., Hajek, T., Liancourt, P., Herberich, M.M., Kramp, R.E., Tomiolo, S., Pacheco-Riano, L.C., Tielborger, K., Majekova, M., 2024. Belowground niche partitioning is maintained under extreme drought. Ecology 105, e4198. Wickham, H., Chang, W., Wickham, M.H., 2016. Package ‘ggplot2’: An implementation of the Grammar of Graphics. Journal of Statistical Software 1, 189. Wong, K.M., Griffiths, M., Moran, A., Johnston, A., Liu, A.E., Sellers, M.A., Topp, C.N., 2023. Cover crop cultivars and species differ in root traits potentially impacting their selection for ecosystem services. Plant and Soil 500, 279–296. Wu, X., Liu, H., Li, X., Ciais, P., Babst, F., Guo, W., Zhang, C., Magliulo, V., Pavelka, M., Liu, S., Huang, Y., Wang, P., Shi, C., Ma, Y., 2018. Differentiating drought legacy effects on vegetation growth over the temperate Northern Hemisphere. Global Change Biology 24, 504–516. Wu, Z., Behzad, H.M., He, Q., Wu, C., Bai, Y., Jiang, Y., 2021. Seasonal transpiration dynamics of evergreen Ligustrum lucidum linked with water source and water-use strategy in a limestone karst area, southwest China. Journal of Hydrology 597, 126199. Wunsch, A., Liesch, T., Broda, S., 2022. Deep learning shows declining groundwater levels in Germany until 2100 due to climate change. Nature Communications 13, 1221. Yang, Q., Fan, J., Luo, Z., 2024. Response of soil moisture and vegetation growth to precipitation under different land uses in the Northern Loess Plateau, China. Catena 236, 107728. Yin, R., Kardol, P., Thakur, M.P., Gruss, I., Wu, G.L., Eisenhauer, N., Schadler, M., 2020. Soil functional biodiversity and biological quality under threat: intensive land use outweighs climate change. Soil Biology & Biochemistry 147, 107847. Yu, L., Leng, G., Tang, Q., Lu, C., Gao, Y., Yao, L., Fan, S., Dong, J., Tu, H., Ai, Z., Liao, X., 2025. Land-atmosphere interactions mitigate the direct surface-drying effects of land use and cover change. One Earth 8, 101210. Zeleke, K.T., Wade, L.J., 2012. Evapotranspiration Estimation Using Soil Water Balance, Weather and Crop Data, In: Irmak, A. (Ed.), Evapotranspiration – Remote Sensing and Modeling. InTech, Rijeka, Croatia. Zeng, Z., Wu, W., Ge, Q., Li, Z., Wang, X., Zhou, Y., Zhang, Z., Li, Y., Huang, H., Liu, G., Peñuelas, J., 2021. Legacy effects of spring phenology on vegetation growth under preseason meteorological drought in the Northern Hemisphere. Agricultural and Forest Meteorology 310, 108630. Zhang, G., Zhang, Y., Zhao, D., Liu, S., Wen, X., Han, J., Liao, Y., 2023. Quantifying the impacts of agricultural management practices on the water use efficiency for sustainable production in the Loess Plateau region: A meta-analysis. Field Crops Research 291, 108787. Zhao, C., Jia, X., Zhu, Y., Shao, M.a., 2017. Long-term temporal variations of soil water content under different vegetation types in the Loess Plateau, China. Catena 158, 55–62. Zuo, Q., Shi, J., Li, Y., Zhang, R., 2006. Root length density and water uptake distributions of winter wheat under sub-irrigation. Plant and Soil 285, 45–55. Zwicke, M., Picon-Cochard, C., Morvan-Bertrand, A., Prud'homme, M.P., Volaire, F., 2015. What functional strategies drive drought survival and recovery of perennial species from upland grassland? Annals of Botany 116, 1001–1015. Additional Declarations There is NO Competing Interest. Supplementary Files 20250704supplementary.docx Supplementary Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7062058","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":486696103,"identity":"97f01c77-7cdb-4186-a86e-ce6fa40acfc9","order_by":0,"name":"Mengqi Wu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYNACAyBmb2D4wMDATIoWngOMM0jQAgISCURq0W3vPfziTQFD4oabbwwbfzBYyxHUYnbmXJrlHAOglts5hg0SDOnGhLXcyDEz5jFgyAVqMX9gwHA4sYGglvtvoFpunjFsSGA4XE9Yyw0e48dgLTd4DBsOMBxOIMIvOWaMcwwk6meeSStsbDBINyRsy/Ezxh/e/LEx5jt+eGPjjwpreYK2AAGbBA+DBJRtQIwGYOx94CFO4SgYBaNgFIxUAAB4sz3trtDr4AAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-6538-619X","institution":"Helmholtz-Centre for Environmental Research - UFZ, Halle, Germany","correspondingAuthor":true,"prefix":"","firstName":"Mengqi","middleName":"","lastName":"Wu","suffix":""},{"id":486696104,"identity":"1b3092d1-cddb-4eaa-b040-504fa9240469","order_by":1,"name":"Christiane Roscher","email":"","orcid":"","institution":"German Centre of Integrative Biodiversity Research (iDiv), Germany","correspondingAuthor":false,"prefix":"","firstName":"Christiane","middleName":"","lastName":"Roscher","suffix":""},{"id":486696105,"identity":"327b3194-965f-4db0-a408-42bb8c54e6d9","order_by":2,"name":"Martin Schädler","email":"","orcid":"https://orcid.org/0000-0001-9700-0311","institution":"German Centre for Integrative Biodiversity Research (iDiv)","correspondingAuthor":false,"prefix":"","firstName":"Martin","middleName":"","lastName":"Schädler","suffix":""},{"id":486696106,"identity":"6b9fc61a-1b8e-4343-a7b4-61a37574d438","order_by":3,"name":"Mika Tarkka","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Mika","middleName":"","lastName":"Tarkka","suffix":""},{"id":486696107,"identity":"d4141036-0bc6-4b07-835b-d3bc5bf8a418","order_by":4,"name":"Doris Vetterlein","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Doris","middleName":"","lastName":"Vetterlein","suffix":""},{"id":486696108,"identity":"ddc4e561-15d4-48e7-97c7-d0c3e67cc69a","order_by":5,"name":"Steffen Schlueter","email":"","orcid":"https://orcid.org/0000-0002-3140-9058","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Steffen","middleName":"","lastName":"Schlueter","suffix":""}],"badges":[],"createdAt":"2025-07-07 06:50:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7062058/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7062058/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87226742,"identity":"bfad6dac-a06e-42db-a92b-aa7d22e241c4","added_by":"auto","created_at":"2025-07-21 17:33:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":183463,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions21.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/feda036d80b9dd472c8d36d7.png"},{"id":87228297,"identity":"77fc93e4-d999-4184-8e3e-e3ab39e18a40","added_by":"auto","created_at":"2025-07-21 17:57:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":118470,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions22.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/dc001401427ed890e65595c9.png"},{"id":87227554,"identity":"7a8a9487-04c1-43c6-a8ef-7738afef22eb","added_by":"auto","created_at":"2025-07-21 17:41:34","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":268423,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions23.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/c37f122551776fa4732d47be.png"},{"id":87227550,"identity":"9cb79973-c940-4348-94a5-916786933c7a","added_by":"auto","created_at":"2025-07-21 17:41:34","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":254800,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions24.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/4ae9b0b46d7e9f733b17a241.png"},{"id":87227799,"identity":"50b318a8-7fe5-426b-bf3e-03507dc87d51","added_by":"auto","created_at":"2025-07-21 17:49:34","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":321607,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions25.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/575c8c7f0a80760163dc07d5.png"},{"id":87226745,"identity":"50d9cb2f-61d4-464d-addf-fc84c4917f22","added_by":"auto","created_at":"2025-07-21 17:33:34","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":476486,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions26.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/98df7b7669557a2ceffabba1.png"},{"id":87227557,"identity":"1c28a023-e3cb-48fe-9d82-94d3d128365a","added_by":"auto","created_at":"2025-07-21 17:41:34","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":463580,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions27.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/baf19c7b27759a3fc8e150a3.png"},{"id":87227795,"identity":"dc7e1173-6604-44e7-8ad7-8e81f4ed588f","added_by":"auto","created_at":"2025-07-21 17:49:34","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":129686,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions28.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/31f4524c43406e29434a9a48.png"},{"id":87226746,"identity":"9932ffcc-8f47-4b1b-93c0-99922bac741c","added_by":"auto","created_at":"2025-07-21 17:33:34","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":146922,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"20250704figurecaptions29.png","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/8453cf718826a4083787a6cc.png"},{"id":94211280,"identity":"3b36bb4c-a661-410b-8d70-8025e6f3f327","added_by":"auto","created_at":"2025-10-23 15:45:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3298036,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/c66a7470-d585-4b02-8fe9-78b20a71c86e.pdf"},{"id":87228298,"identity":"c0dcd538-4b61-40e6-900b-88f68dbd047a","added_by":"auto","created_at":"2025-07-21 17:57:34","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":917985,"visible":true,"origin":"","legend":"Supplementary","description":"","filename":"20250704supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-7062058/v1/cdccc2802390fee1fff81f17.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Legacy effects of climate extremes on deep soil water storage and water use efficiency across different land-use systems","fulltext":[{"header":"1. INTRODUCTION","content":"\u003cp\u003eExtreme drought and precipitation events, projected to increase in frequency, duration and intensity with human-induced climate change, threaten the global water cycle and agricultural productivity (Maurel and Nacry, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Wunsch et al., \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Chen et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Soil water content, a critical mediator between meteorological conditions and biogeochemical processes (Li et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Vicente-Serrano et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), governs root and shoot growth dynamics and thus ecosystem productivity (Zuo et al., \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Vereecken et al., \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Tissink et al., \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Yet, the supply of precipitation and the storage of water in the topsoil often fail to meet the water demands of plants, particularly during peak growth (Ali et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2024a\u003c/span\u003e). Deep soil water storage, which is accessible to plant roots, may play an unaccounted role in buffering against summer droughts, sustaining aboveground growth and belowground ecosystem stability under extreme climatic conditions (Wang et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e). However, the extent to which deep soil water storage stabilizes plant productivity and ecosystem functions under climate extremes across different land-use systems remains poorly understood.\u003c/p\u003e\u003cp\u003eThe spatiotemporal distribution of soil water is strongly influenced by land-use type and intensity through the differences in vegetation cover, root system activity, and the timing and duration of water use across the growing season (Spera et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Wang-Erlandsson et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Yu et al., \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Understanding how ecosystems respond differently to climate extremes under various land-use types and intensities is critical to predict future ecosystem services. For example, annual cropland and perennial grassland differ markedly in plant community composition, root system phenology, and seasonal water demand (Fan et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e). Grasslands, dominated by perennial plant species, typically maintain dense and continuous vegetation cover throughout the year, while croplands undergo seasonal cycles of sowing and harvest, leading to periods of low cover or even bare ground and increased evaporative water loss from the topsoil (Wang et al., \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Bagley et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Moreover, croplands are characterized by shallow, fast-developing root systems that concentrate water uptake during a short peak growth stage, whereas grasslands support deeper and more persistent root networks that sustain water uptake and continuous transpiration throughout the growing season (Holmes and Rice, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Swindon et al., \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). These contrasting root and vegetation cover dynamics result in distinct seasonal patterns of soil water depletion and influence the ecosystem\u0026rsquo;s capacity to withstand climate extremes (Maestre et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Scherzinger et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Additionally, land-use intensity, characterized by fertilization dose, tillage intensity in croplands, or mowing frequency, the use of a few grass cultivars or a more diverse species composition in grasslands, can further modify root development and vegetation cover, this in turn affects water uptake efficiency and soil water distribution (Rose et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Spera et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The interaction between land-use type and intensity and a projected future climate scenario in shaping soil water dynamics, and the potential implications of this for plant productivity under climate extremes, remains unclear.\u003c/p\u003e\u003cp\u003eIn particular, the relationship between rooting density and water availability remains poorly understood, complicating efforts to interpret plant responses to climate extremes. This complexity arises from several reasons: (1) water uptake is not uniform across the entire root system but can be root type specific, is higher close to root tips and can be altered by the presence of root hairs (Segal et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Ahmed et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2018\u003c/span\u003e); (2) spatial heterogeneity in soil water can decouple the water uptake patterns from root density distribution (Carminati et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2016\u003c/span\u003e); and (3) elevated root densities may reflect nutrient foraging strategies rather than water demand (Lynch, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This mismatch between root distribution and water uptake poses a critical challenge for understanding the plant adjustment to extreme climate conditions. The structure and function of deep rooting systems and their contribution to deep soil water uptake remain poorly understood, particularly under field conditions (Draye et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Pierret et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). These uncertainties hinder a comprehensive evaluation of the extent to which shifts in rooting patterns constitute an effective mechanism for coping with water stress under changing climate.\u003c/p\u003e\u003cp\u003ePlants rely on different strategies that enhance water uptake and improve water use efficiency (WUE) to maintain productivity under climate extremes. In agricultural systems, management strategies include selecting or breeding for cultivars with inherently higher WUE, optimizing cropping system and soil tillage regimes in croplands or adjusting mowing schedules in grasslands (Condon et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Rose et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). While such strategies may delay the onset of drought stress, they often have limited effectiveness in sustaining overall productivity under prolonged water deficits (Leakey et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Drought resistance tends to decline with increasing land-use intensity as higher productivity is mostly associated with higher water demand (Van Sundert et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Bazzichetto et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Korell et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), potentially exacerbating ecosystem vulnerability under intensified management regimes. In grasslands, diverse plant communities contribute to ecosystem resistance under climate extremes by combining shallow- and deep-rooting species, which enhances partitioning of soil water resources and minimizes competition under water stress (Craine et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Isbell et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Weides et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Increasing attention has been paid to the legacy effect of climate extremes: persistent alterations in soil water availability that extend beyond the period of stress itself (Bastos et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Muller and Bahn, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Liu et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This legacy of soil water storage may reshape subsequent rooting patterns, alter plant community composition, and influence WUE, with significant implications for the sustainability of plant production systems under ongoing climate change.\u003c/p\u003e\u003cp\u003eThe extent to which deep soil water storage, driven by the legacy effect of climate extremes, can buffer productivity loss and affect WUE under future climate scenario remains poorly understood. Therefore, we made use of a unique experimental platform, Global Change Experimental Facility \u0026ndash; GCEF (see Sch\u0026auml;dler et al. (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)), in which a realistic future climate scenario is compared against ambient climate across five land-use types (two croplands and three grasslands), each representing a gradient of land-use intensities.\u003c/p\u003e\u003cp\u003eThe objective of this study was to investigate how climate extremes influence spatiotemporal patterns of soil water storage under different land-use types and how these patterns are affected by a future climate scenario involving modified seasonal precipitation and higher temperature. We further aimed to investigate whether and how legacy effects of deep soil water storage influence plant productivity and WUE in response to climate extremes that we observed in a three-year monitoring campaign. We hypothesized that, (1) topsoil water content is primarily driven by atmospheric forcing, but deep soil water content is regulated by root water uptake and legacy effect of past soil moisture conditions. (2) On an annual basis, croplands maintain a higher soil water content than grasslands due to their sparser root systems and shorter growing season, whereas the impact of the future climate scenario on deep soil water storage is unclear and should be clarified by this study. Furthermore, we expected that (3) crops achieve higher yields than grasslands due to breeding for resource efficiency and rapid growth, whereas extensive grasslands, with greater plant diversity, are most resistant to climate extremes. Intensive grasslands, characterized by low diversity but permanent cover, are expected to be less resilient than extensive grasslands but more stable than croplands, due to perennial root systems and continuous cover. Finally, we hypothesized that (4) plant adjustment to drought conditions and limitations in deep soil water supply would mainly occur through higher WUE and less through changing rooting patterns or changes in vegetation composition. These hypotheses were tested through three years of monitoring soil water content profiles, yield, and WUE across all five land-use types and two climate scenarios. Interpretation of the observed growth and water uptake patterns was supported by data on plant community composition and root length density profiles collected during peak growth. Our analysis explicitly accounted for site-specific precipitation history and interannual climatic variability, providing a robust context for evaluating land-use and climate interactions on ecosystem water dynamics and productivity.\u003c/p\u003e"},{"header":"2. MATERIAL AND METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Site description\u003c/h2\u003e\u003cp\u003eOur study was conducted within the Global Change Experimental Facility (GCEF) of the Helmholtz-Centre for Environmental Research (UFZ) in Bad Lauchst\u0026auml;dt, Germany (51\u0026deg; 23\u0026prime; N, 11\u0026deg; 53\u0026prime; E, 118 m a.s.l.). The area is characterized by a sub-continental climate, with a mean annual precipitation of 525 mm (1993\u0026ndash;2013) and a mean annual temperature of 9.7\u0026deg;C (1993\u0026ndash;2013) (Sch\u0026auml;dler et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The soil type is a Haplic Chernozem, with on average 21% clay, 69% silt, and 10% sand (Altermann et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The volumetric water content at field capacity (pF1.8 = -60 hPa) and permanent wilting point (pF4.2 = -15000 hPa) amounted to 40.7% and 14.9% in the topsoil, respectively (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe GCEF platform was established in 2013 and consists of a split-plot design with two climate scenarios as the main plot factor and five land-use types as the subplot factor. A more detailed description of this facility is given in Sch\u0026auml;dler et al. (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In brief, the GCEF includes 50 subplots which are arranged in ten main plots of 80 \u0026times; 24 m size with five main plots subjected to ambient climate and to future climate each. The climate treatment is based on a projection of the climate of Central Germany for the years 2070\u0026ndash;2100 based on different dynamic regional climate models (Jacob and Podzun, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; D\u0026ouml;scher, 2002; Rockel et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). The use of automated roofs and side panels in the future climate to passively increase the night temperatures increased mean air temperature on average by 0.55\u0026deg;C. Precipitation was increased by 10% in spring and autumn and reduced by 20% in summer as compared to the ambient climate (Sch\u0026auml;dler et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Yin et al., \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The land-use types include two croplands: conventional farming (CF) with a typical regional crop rotation consisting of winter barley, triticale, and winter wheat, the application of chemical weed control and mineral fertilizers, and organic farming (OF) characterized by legume (white clover) replacing the triticale in the crop rotation, mechanical weed control, non-stained seeds and a restricted use of pesticides. It also includes three grasslands: intensive meadow (IM) sown with a commercially used mixture of cultivars of forage grasses, and managed by moderate fertilization and frequent mowing (2\u0026ndash;3 times per year), extensive meadow (EM) sown with a species-rich mixture of more than 50 regionally typical grassland species and managed by moderate mowing (1\u0026ndash;2 times per year), and extensive pasture (EP) with the same sown species mixture as EM but managed by 2\u0026ndash;3 times one-day grazing events per year with sheep. For more details we refer to Sch\u0026auml;dler et al. (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe precipitation and irrigation patterns of both climate scenarios and the management of each land-use type from 2022 to 2024 are summarized in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The cumulative growing season rainfall (GSR), measured as total precipitation and irrigation from April to October, was almost twice as high in 2023 and 2024 as in 2022 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 Meteorological water deficit and standardized precipitation-evapotranspiration index (SPEI)\u003c/h2\u003e\u003cp\u003eThe meteorological water deficit and SPEI are widely used to identify and quantify wet and dry climate events in ecological studies. Meteorological water deficit is the difference between precipitation and potential evapotranspiration. SPEI is derived from the non-exceedance probability of the water deficit, fitted to a three-parameter log-logistic distribution to account for common negative values (Beguer\u0026iacute;a et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Values below \u0026minus;\u0026thinsp;1.28 and above 1.28 indicate extreme dry and wet conditions, respectively (Isbell et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Precipitation is calculated as the sum of precipitation and irrigation, whereas potential evapotranspiration is estimated using the Penman-Monteith equation, with solar radiation derived from the diurnal temperature range following the Hargreaves method (Hargreaves and Allen, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). As no crop-specific parameters are included, this estimate reflects atmospheric demand rather than actual plant evapotranspiration. The daily precipitation and temperature data used for water deficit and SPEI calculations were obtained from the German Meteorological Service (DWD) weather station in Bad Lauchst\u0026auml;dt (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://opendata.dwd.de/climate_environment/CDC/observations_germany/climate/daily/kl/\u003c/span\u003e\u003cspan address=\"https://opendata.dwd.de/climate_environment/CDC/observations_germany/climate/daily/kl/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and aggregated to monthly values. The water deficit was also aggregated to monthly values, whereas the SPEI of a given month represented the 3-month average of this month and the two antecedent months to account for seasonal hydrological consistency. Detailed descriptions of the water deficit and SPEI calculation can be found in Vicente-Serrano et al. (\u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) and Schnabel et al. (\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The \u0026ldquo;\u003cem\u003eSPEI\u003c/em\u003e\u0026rdquo; package in R (Beguer\u0026iacute;a et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) was used for analysis.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Soil water content and storage profiles\u003c/h2\u003e\u003cp\u003eSoil moisture profiles were monitored at each subplot using time domain reflectometry (TDR) probe at approximately weekly to bi-weekly intervals from April 2022 to December 2024. The device was a TRIME-PICO T3/IPH44 probe (IMKO Micromodultechnik GmbH, Ettlingen, Germany), which allows non-destructive determination of soil volumetric water content in a soil profile (Kano-Nakata et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). On each subplot of five land-use types, a soil auger was used to drill down to 110\u0026ndash;120 cm and install the plastic access tubes. The soil volumetric water content of each layer was calculated as the average of two measurements and were made at depths of 10, 20, 30, 50, 70, 90 and 110 cm. The probe was rotated by 90\u0026deg; between two measurements to compensate for the elliptical sensitivity of the measuring field. Missing data on the croplands from July to October/November in three years were due to the removal of access tubes between harvest in summer and tillage in fall. Topsoil and deep soil water storage were calculated as the average soil water content during the growing season (the period from the first day when the mean daily temperature consistently exceeds 5\u0026deg;C (the minimum threshold for the growth of most grassland and crop species in Central Germany) until harvest in early July) in the 0\u0026ndash;30 cm and 30\u0026ndash;110 cm soil layers, respectively. The change in topsoil and deep soil water storage was assessed by comparing at the beginning and end of the growing season.\u003c/p\u003e\u003cp\u003eTo check the accuracy of the TDR, the monitoring data were calibrated by a comparison with four volumetric water content measurements ranging from saturated soil water content to the permanent wilting point using a balance and the TDR probe in an access tube at the center of a homogeneous soil column (34 cm diameter, 35 cm height). The soil column consisted of sieved topsoil material from the GCEF that was compacted to a representative bulk density for the site (1.44 g cm\u003csup\u003e\u0026minus;\u003c/sup\u003e\u0026sup3;). To facilitate a homogeneous moisture profile across the sensor depth, a new soil column was packed for each water content and the water was already added before packing for the lower target water contents and partly added from the top after packing for the higher target water content with sufficient time for equilibration. Volumetric water contents were derived from the known dry weight and bulk density of the soil column determined after oven-drying at 105\u0026deg;C for 48 hours. This calibration resulted in the regression equation Y\u0026thinsp;=\u0026thinsp;1.63X \u0026ndash; 1.63, R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.997, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, n\u0026thinsp;=\u0026thinsp;4, where X is the soil moisture measured by TDR and Y is the calculated soil moisture. The calibration was determined once and used for the entire monitoring period.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Root sampling and analysis\u003c/h2\u003e\u003cp\u003eCylindrical soil cores were taken in May 2022 and 2023 using a 5-cm-diameter soil auger. Two cores were taken per subplot and sectioned into three depth layers: 0\u0026ndash;15 cm, 15\u0026ndash;30 cm and 30\u0026ndash;50 cm, respectively. In total, 200 soil cores were collected (2 climate scenarios \u0026times; 5 land-use types \u0026times; 5 treatment replicates \u0026times; 2 sampling replicates per subplot \u0026times; 2 sampling years), resulting in 600 depth-specific samples. All samples were stored at 4\u0026deg;C immediately to inhibit microbial activity and prevent root decay.\u003c/p\u003e\u003cp\u003eAll soil cores were washed within a week after sampling with tap water over 0.63 mm sieves to extract the roots from the soil, with each sample requiring about 40 minutes to process. Care was taken to minimize damage to the root system during washing. The collected roots were stored in a 50% ethanol solution (Rotisol) prior to analysis. Roots were scanned on a flatbed scanner at 400 dpi (EPSON Perfection V700). Then the obtained images were analyzed using WinRHIZO Pro\u0026trade; (Version 2019a, Regent Instruments, Canada) to measure root characteristics, i.e., root length density and root length distribution in different root diameter range.\u003c/p\u003e\u003cp\u003eIn addition, root length density was adjusted using correction coefficients specific to land use and year to obtain correlated root length density, as fresh roots primarily contribute to root water uptake. Fresh roots cannot be distinguished from root residues, that were dead before sampling, during root washing and sorting and also not with image analysis with WinRhizo software. These correction coefficients were estimated independently as the relative proportion of fresh roots to total roots (Fig. S2), derived from X-ray CT images of intact soil cores (Phalempin et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), which were taken in the same soil depths shortly after root sampling. The proportion of fresh roots was 85.1\u0026ndash;91.2% in croplands and 50.4\u0026ndash;59.4% in grasslands.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Vegetation recording\u003c/h2\u003e\u003cp\u003eShortly before root sampling, an area of 30 cm in diameter, centered on the sampling points for soil coring, was marked with a metal frame in the grassland subplots. All plant species rooting within this area were recorded and species-level cover as well as total vegetation cover were estimated to the nearest percentage.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e2.6 Evapotranspiration and water use efficiency\u003c/h2\u003e\u003cp\u003eEvapotranspiration (ET) was estimated using the soil water balance equation (Zeleke and Wade, \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) as follows:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:ET=P+I+\\varDelta\\:S-R-D+CR\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\left[mm\\right]$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003ewhere \u003cem\u003eET\u003c/em\u003e is the evapotranspiration (mm), \u003cem\u003eP\u003c/em\u003e is precipitation (mm) (referring to the values measured under the automated proofs), \u003cem\u003eI\u003c/em\u003e is irrigation application (mm), \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varDelta\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cem\u003eS\u003c/em\u003e is soil water storage change (mm) (soil water content at the start minus soil water content at the end of a given period for the 0-110 cm depth), \u003cem\u003eR\u003c/em\u003e is surface runoff (mm), \u003cem\u003eD\u003c/em\u003e is drainage from the root zone and \u003cem\u003eCR\u003c/em\u003e is capillary rise to the root zone. Due to the deep soil profile and the large water holding capacity, runoff is zero in the field. Due to the deep groundwater table (about 32 m below surface), capillary rise is negligible. Lysimeters with 1 m deep intact soil monoliths on-site featured no seepage at the lower boundary during three years. Therefore, \u003cem\u003eR\u003c/em\u003e, \u003cem\u003eD\u003c/em\u003e and \u003cem\u003eCR\u003c/em\u003e were set to zero in this study (Sun et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The net change in water content during peak growth used to link with root length density at 0\u0026ndash;50 cm depth was calculated as the cumulative water input (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;+\u0026thinsp;\u003cem\u003eI\u003c/em\u003e) and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varDelta\\:\\)\u003c/span\u003e\u003c/span\u003e\u003cem\u003eS\u003c/em\u003e in the 0\u0026ndash;50 cm depth over eight weeks spanning four weeks before to four weeks after root sampling.\u003c/p\u003e\u003cp\u003eWater use efficiency (WUE) is defined as yield divided by \u003cem\u003eET\u003c/em\u003e. The cumulative \u003cem\u003eET\u003c/em\u003e was calculated from the first day when the mean temperature consistently exceeds 5\u0026deg;C until the day before harvest. Yield data was obtained from machine harvest records on all subplots, where the sum of grain and straw was taken for the croplands. Harvest dates for croplands and grasslands were synchronized through linear interpolation of the biomass production between two cuts in the meadows to ensure comparability between land-use types. Extensive pastures were excluded from WUE estimation because biomass removal by sheep grazing could not be accurately determined.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e2.7 Statistical analysis\u003c/h2\u003e\u003cp\u003eAll statistical analyses were conducted using R 4.1.3. Linear mixed-effects models (LMMs) were fitted using the \u0026ldquo;\u003cem\u003elmer\u0026rdquo;\u003c/em\u003e function in the \u0026ldquo;\u003cem\u003elme4\u0026rdquo;\u003c/em\u003e package (Bates, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), incorporating random intercepts to account for the split-plot design and repeated measures. These models were applied to analyze soil water content and storage at different depths, root length density at different depths, total vegetation cover, and water use efficiency. To account for the compositional nature of functional group data, we calculated log-ratios using grass as the reference group. Specifically, we computed the log-ratio of forb to grass (log(forb/grass)) and legume to grass (log(legume/grass)). These log-ratios were then analyzed jointly in a linear mixed-effects model to capture functional group dependence. Land use, climate, year, and their interactions were fixed factors, while year (sampling time), and subplot nested within main plot served as random effects. For models involving root length density, sampling depth, as well as its interaction with land use and year, were additionally tested. Models were fitted using maximum likelihood (ML) estimation, and likelihood ratio tests (χ\u003csup\u003e2\u003c/sup\u003e ratio) were used to assess the significance of fixed effects through stepwise model comparison. We used the \u0026ldquo;emmeans\u0026rdquo; package (Lenth et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) to further analyze significant interactions, by running post-hoc pairwise comparisons of estimated marginal means between treatment levels if LMMs yielding significant effects of land-use types and interactions between land use and climate. Linear regression analysis was performed to assess relationships between net change in water content and correlated root length density. We used the \u0026ldquo;lme\u0026rdquo; function in the \u0026ldquo;piecewiseSEM\u0026rdquo; package (Lefcheck et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) to develop structure equation modeling (SEM) to further uncover the direct and indirect contributions of land use, climate scenario, and climate extremes (SPEI) to yield and water use efficiency (Fig. S3). Prior to scaling, the categorical levels of land use were ordered as CF, IM, OF, and EM to reflect the gradient of land-use intensity. SPEI, topsoil water storage, deep soil water storage, yield, and water use efficiency were the measured values during the growing seasons until crop harvest. The dataset under extensive pasture was excluded because the yield and water use efficiency were missing. If the initial model was identified as saturated based on Shipley\u0026rsquo;s test of \u003cem\u003ed\u003c/em\u003e-separation, indicating by a non-significant \u003cem\u003eFisher\u0026rsquo;s C\u003c/em\u003e statistic (\u003cem\u003eChi-square\u003c/em\u003e distributed), non-significant paths were subsequently removed to gain a reduced/best-fitted model (Shipley, \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Figures were produced with the ggplot2 package (Wickham et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e"},{"header":"3. RESULTS","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003e3.1 SPEI and soil water content spatiotemporal dynamics for three years\u003c/h2\u003e\u003cp\u003eThe SPEI values indicate mild- to extreme drought conditions (-2.39\u0026thinsp;\u0026lt;\u0026thinsp;SPEI \u0026lt; -0.09) throughout 2022, followed by wetter conditions in mid- and late- 2023 and moderate drought to extreme wet conditions (-1.20\u0026thinsp;\u0026lt;\u0026thinsp;SPEI\u0026thinsp;\u0026lt;\u0026thinsp;2.08) in 2024 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Notably, a heavy rainfall event in August 2023 resulted in a water surplus of 142 mm, causing the SPEI to shift abruptly from \u0026minus;\u0026thinsp;2.17 in July to 0.77 in August. The consequences of interannual changes in atmospheric forcing on soil water profiles are demonstrated exemplarily for extensive meadows under ambient climate scenario (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The soil moisture at 0\u0026ndash;20 cm depth was closely related with the SPEI (looking three months backward in time, R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.57, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and to a lesser degree also with monthly water deficit (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.23, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.005) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, S4). Summer drought conditions persisted in deeper soil layers, with water content approaching the permanent wilting point, which for our site is at 14.9% volumetric water content. However, precipitation alleviated these conditions in topsoil, resulting in drier deep soil than topsoil throughout most of the monitoring period. The legacy effect of summer drought was manifested by the long duration for infiltration fronts to reach the depth of 100 cm (five months in 2022, two months in 2023). In 2024, increased precipitation led to higher soil water content well above the permanent wilting point throughout the profile and the entire time, making it the wettest year in the monitoring period (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The heavy rain event in August 2023 (on average 5.8 mm per day, with a maximum daily rainfall of 59.3 mm and a cumulative total of 180.4 mm) caused short-term water ponding, but eventually the entire volume of precipitation was taken up by the soil profile without raising the water content to field capacity (at 40.7% soil water content) due to the enormous soil water deficit at the time (305 mm available capacity) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea, \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\u003ch2\u003e3.1.1 Comparison of soil water content between cropland and grassland\u003c/h2\u003e\u003cp\u003eCropland exhibited consistently higher soil water content than grassland across the entire soil profile, except for 0\u0026ndash;10 cm depth in which cropland was drier (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). The discrepancy in average soil water content between cropland and grassland increased with depth. Specifically, when averaged across the period for which data is available in both croplands and grasslands, soil water content in cropland was 7.6\u0026ndash;7.8% lower than in grassland at 0\u0026ndash;10 cm depth, but 4.0-15.8% greater at 20\u0026ndash;50 cm depth, and 17.8\u0026ndash;31.9% higher at 50\u0026ndash;110 cm depth (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The future climate scenario did not affect average soil water content compared to ambient climate irrespective of soil depth (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn the topsoil (0\u0026ndash;30 cm), soil moisture exhibited strong seasonal fluctuations, characterized by pronounced increases during winter and declines in summer (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). In contrast, the deep soil layer (30\u0026ndash;110 cm) displayed a more gradual, delayed response, with cropland maintaining higher soil moisture than grassland during winter 2022, most of 2023, and spring 2024. Differences between climate scenarios were evident, but they were episodic and depended on land-use types. From January to June 2023, the water content of deep cropland soil under the future climate scenario was higher than under the ambient climate, reflecting the percolation of added irrigation in autumn of 2022 and spring of 2023. In contrast, the water content of deep grassland soil remained consistently lower after the heavy rainfall event in August 2023, as 20% of this event was shielded in the future climate scenario.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section3\"\u003e\u003ch2\u003e3.1.2 Comparison of soil water content between conventional and organic farming\u003c/h2\u003e\u003cp\u003eNo significant differences were observed in 0\u0026ndash;20 cm depth, but organic farming consistently exhibited higher soil water content than conventional farming in deeper soil layers (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The differences between the two farming systems were more pronounced under the ambient climate scenario, with organic farming maintaining 6.4\u0026ndash;17.9% higher water content levels (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Higher soil moisture in the deeper soil under organic farming was most evident during the dry years of 2022 and 2023, but disappeared during the wetter year of 2024 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Under the future climate scenario, soil water content in conventional farming was significantly higher from January to June 2023 in the deep soil layers as a delayed response to irrigation, but declined below ambient climate levels after August 2023 because of precipitation shielding.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\u003ch2\u003e3.1.3 Comparison of soil water content between intensive and extensive grasslands\u003c/h2\u003e\u003cp\u003eCompared to extensive meadows and pastures, intensive meadows had drier topsoil but moister deeper soil layers (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). On average, soil water content in intensive meadows across the entire profile was 21.9\u0026ndash;41.8% and 13.1\u0026ndash;18.6% higher under ambient and future climate scenarios, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea). The gap between deep soil water contents of intensive and extensive grasslands opened when the storage was declining in late summer and became smaller, when the storage was refilled in spring (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb). Under the future climate scenario, deep soil water content significantly declined after September 2023 compared to ambient climate, when the 20% deficit of the August rain event started to reach the subsoil. This discrepancy was strongest in intensive meadows, as the antecedent soil moisture was the same and persisted until the end of the monitoring period in December 2024 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). However, in extensive pastures, soil water content under the future climate scenario was initially higher, a result of the additional irrigation in autumn and spring, but dropped below ambient climate levels after September 2023 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.2 Root length density (RLD), root length distribution, and vegetation cover\u003c/h2\u003e\u003cp\u003eRLD in grasslands was 3.6\u0026ndash;5.5 times higher than in croplands, with the highest values observed in extensive meadows and lowest values in organic farming (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). From 0\u0026ndash;15 cm to 15-30cm and 30\u0026ndash;50 cm soil depth, RLD in grasslands were 3.0\u0026ndash;12.0 times, 2.7\u0026ndash;7.4 times and 1.6\u0026ndash;3.4 times as high as in croplands (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb, Table S2-3). Across all land-use types, RLD decreased with increasing soil depth at similar proportions.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eA significant interaction of land use and year suggested interannual variability in RLD, particularly in croplands at 0\u0026ndash;30 cm depths (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). Compared to 2022, the RLD in 2023 increased by 69.9% in conventional farming at 0\u0026ndash;15 cm depth, 102.9% and 166.2% in organic farming at 0\u0026ndash;15 cm and 15\u0026ndash;30 cm depth, but decreased by 32.3% in intensive meadow at 0\u0026ndash;15 cm depth (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb, Table S2-3).\u003c/p\u003e\u003cp\u003eRLD was greater in grasslands than in croplands for all root diameter classes, but especially for fine roots (\u0026lt;\u0026thinsp;0.2 mm) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec, S5). Fine roots (\u0026lt;\u0026thinsp;0.1 mm) were the most abundant diameter class in intensive meadows, whereas other land-use types had as many or more roots in the 0.1\u0026ndash;0.2 mm diameter range. Notably, the length density of roots with 0.1\u0026ndash;0.3 mm diameter range were significantly higher in extensive meadows and pastures than in intensive meadows (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec).\u003c/p\u003e\u003cp\u003eAcross all three grassland types, the total vegetation cover of the soil surface was higher in 2023 than in 2022 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed, Table S2-3). The relative abundance of plant functional groups (grass, forb, legume) and functional group dependence varied both across land-use types and between years (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ee, Table S2-3). The higher total vegetation cover for extensive meadows and pastures in 2023 as compared to 2022, was mainly caused by a higher relative abundance (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ee) and summed cover per functional group (Fig. S6) of legumes.\u003c/p\u003e\u003cp\u003eFurthermore, the future climate scenario had no significant effect on overall RLD (0\u0026ndash;50 cm), RLD at individual depth intervals, root length distribution, total vegetation cover, and plant functional group dependence (Table S2).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.3 Soil water depletion and root water uptake patterns\u003c/h2\u003e\u003cp\u003eChanges in both topsoil and deep soil water storage during the growing season significantly varied across land-use types and years (Fig. S7, Table S2, S4). In both the topsoil (0\u0026ndash;30 cm) and deep soil layer (30\u0026ndash;110 cm), croplands generally experienced greater water depletion than grasslands, particularly in the drought years of 2022 and 2023. The highest topsoil water depletion occurred in 2023, while deep soil water storage depletion peaked in 2022, suggesting pronounced deep water uptake during the extreme drought in 2022. In contrast, soil water storage in 2024 showed minimal change or slight increases across all land-use types.\u003c/p\u003e\u003cp\u003eThe relationship between corrected RLD and net change in water content (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;+\u0026thinsp;ΔS) in the same depth profiles (0\u0026ndash;50 cm) during the period of peak growth (eight weeks centered around root sampling in May) differed between croplands and grasslands in both years (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). In croplands, a strong negative correlation (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.53, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) emerged in 2022 (dry period), but disappeared in 2023 (wet period), indicating that the annual crops relied on the limited shallow soil water storage in 2022, but decoupled the water uptake from the actual RLD under sufficient water supply in 2023. In contrast, perennial grasslands with an extensive root system showed no correlation between shallow RLD and net change in water content at 0\u0026ndash;50 cm depth in the dry period of 2022, perhaps because of a shift toward deep root water uptake. A negative correlation (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.33, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) for grasslands emerged in 2023, when the initial soil water contents in the topsoil were higher (close to field capacity, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb) and shallow soil water storage change was greater (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003e3.4 Yield, evapotranspiration and water use efficiency\u003c/h2\u003e\u003cp\u003eAnnual yield across all land-use types increased consistently with water supply from 2022 to 2024 (Fig. S8). Using the normal precipitation regime in 2024 as a reference point, the yield drop in the severe drought year of 2022 was found to be 27.8%, 30.0%, 32.6% and 34.5% for conventional farming, organic farming, intensive meadow and extensive meadow, respectively. Despite interannual variability, the yield differences among land-use types remained stable: conventional farming\u0026thinsp;\u0026gt;\u0026thinsp;intensive meadow\u0026thinsp;\u0026gt;\u0026thinsp;organic farming\u0026thinsp;\u0026gt;\u0026thinsp;extensive meadow. Conventional farming produced 55.40-136.50% higher yields than organic farming, while intensive meadow yielded 61.69-100.61% more plant biomass than extensive meadow.\u003c/p\u003e\u003cp\u003eYield increased with increasing evapotranspiration (ET) during the growing season across all land-use types and both climate scenarios (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003ea), reflecting interannual changes in drought intensity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Interannual variability in ET was evident. In the absence of drought in 2024, ET was highest on average, regardless of land use. It was 9.5% lower under the future climate scenario than under the ambient climate and this gap was consistent across land-use types, indicating that it is mainly caused by the precipitation shielding in summer, when ET demand is highest. During the drought years of 2022 and 2023 (yield in dry periods of 2023 before heavy rainfall in August), ET was lower on average. It was higher under conventional (and organic) farming compared to grasslands. Consistent trends imposed by the future climate scenario vanished.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWater use efficiency (WUE) varied significantly across land-use types and years, with 17.6\u0026ndash;54.7% higher WUEs in croplands than in grasslands (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eb, Table S2, S4). Conventional farming consistently exhibited the highest WUE (4.48\u0026ndash;6.60 kg m\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e), whereas extensive meadows had the lowest (1.89\u0026ndash;2.78 kg m\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e), suggesting that agricultural intensification increases WUE. WUE in intensive meadows was 65.3\u0026ndash;87.4% higher than in extensive meadows (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Table S2, S4). The drought years of 2022 and 2023 caused an increased water use efficiency compared to the normal year of 2024 in all land uses except for the white clover in the organic farming crop rotation of 2023. The future climate scenario only affected WUE under conventional farming, increasing it by 23.6% in 2023 and decreasing it by 19.6% in 2024 compared to ambient climate (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Table S2, S4).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003e3.5 Relationship between soil water storage and plant productivity\u003c/h2\u003e\u003cp\u003eThe SEM revealed that SPEI during the growing season significantly influenced both topsoil and deep soil water storage, although its effect was comparatively weaker on deep soil water storage because of legacy effects (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e). In contrast, land use and the future climate scenario exhibited smaller our even negligible effects on water storage terms, respectively (paths removed in the final model). WUE was directly and strongly affected by land use and SPEI, with small indirect effects through water storage terms. Yield was closely associated with WUE, directly and strongly affected by land use, but the strong SPEI effect on yield was clearly mediated by both water storage terms. These combined observations indicated that yield was mainly driven by water availability and the trade-off between growth and maintenance. The model explained a substantial proportion of the variance in WUE (R\u0026sup2; = 0.76), yield (R\u0026sup2; = 0.81) and topsoil water storage during the growing season (R\u0026sup2; = 0.81), all of which exceeded the explained variance in deep soil water storage (R\u0026sup2; = 0.45) suggesting again the impact of legacy effects on deep water cycling.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"4. DISCUSSION","content":"\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003e4.1 Effect of climate extremes on topsoil and deep soil water storage\u003c/h2\u003e\u003cp\u003eOur results supported the first hypothesis that topsoil water dynamics were primarily controlled by atmospheric forcing in terms of radiation, temperature, and precipitation, and were closely coupled to short-term climatic variability. Shallow soil water content at 0\u0026ndash;20 cm depth was tightly linked to the SPEI throughout a three-year monitoring period with two extended droughts (2022, 2023), a heavy summer rain event (2023) and a normal year (2024), indicating that the shallow soil water storage closely tracked precipitation inputs and evapotranspiration (ET) losses, with a rapid turnover and limited buffering capacity under changing climatic conditions (Good et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Warter et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn contrast, deep soil water exhibited a stronger \u0026ldquo;memory\u0026rdquo; effect on climate variation, particularly during extreme drought events. This finding further supported our hypothesis that deep soil water storage is governed by the legacy effect of past soil moisture conditions. Prolonged periods of negative SPEI led to persistent depletion of deep soil water storage, resulting in an inversion of the soil moisture profiles with depth (soil water content under equilibrium would normally increase with depth due to gravity) that reflected the legacy effects of droughts, while the shallow soil water fluctuated around higher levels following moderate rain events that failed to induce deep infiltration. Notably, infiltration fronts required approximately five months in 2022 and two months in 2023 to reach 100 cm depth after drought, illustrating the delayed replenishment of deep soil moisture. This delay is not only governed by the low unsaturated hydraulic conductivity of dry soils but is further exacerbated by concurrent plant water uptake during infiltration, which reduces the amount of water available for percolation (Bens et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Rickard et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The persistence of drought in deep soil layers restricts plant water access, delaying phenological development, and influencing ecosystem productivity (Wu et al., \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Bastos et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Furthermore, extreme droughts pushed deep soil water content toward the plant wilting point for extended periods, while topsoil water storage recovered more quickly following precipitation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). These patterns aligned with a global modeling study demonstrating increased drying of deep soil layers (\u0026gt;\u0026thinsp;20 cm) during the growing season under a future climate scenario, although not specific to our study site (Schlaepfer et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAs drought extremes intensify, understanding how rooting strategies influence soil water use becomes increasingly critical. Deep rooting can confer drought resistance by accessing subsoil water (Lynch, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Fan et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), but its effectiveness remains debated due to high metabolic costs and limited deep soil water storage (Palta and Turner, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rasmussen et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Figueroa-Bustos et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The effectiveness of the rooting strategy is highly context-dependent, influenced by field conditions and land-use systems (Lynch, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; van der Bom et al., \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). If the unsaturated hydraulic conductivity of the bulk soil is high, like for the silt loam at this site, then water flow towards the roots sustains efficient root water uptake, reducing the need for extensive root systems (Schl\u0026uuml;ter et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Jorda et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Additionally, a heavy rain event in August 2023 substantially replenished deep soil water (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb), offering a long-lasting storage for subsequent dry spells (Feldman et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). However, this recharge occurred post-harvest for the croplands (outside the main growing season), limiting its immediate benefit for crop productivity. In grasslands, although harvest timing is more flexible and depends on the biomass accumulation (typically one or two cuts in extensive meadow, and up to three in intensive meadow), late-summer rainfall was also unlikely to substantially improve productivity. This is probably because the peak growth of grasses generally occurs in late spring during flowering, and later in the season, declining photoperiod and photosynthesis limit further growth (Brookshire and Weaver, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Thus, deep water storage remains a vital buffer in upland soils without groundwater access, mitigating mismatches between water supply and plant demand.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003e4.2 Effect of land use and future climate scenario on deep water storage\u003c/h2\u003e\u003cp\u003eCroplands significantly reduced both the extent and duration of water shortage in deep soil compared to grasslands, primarily due to the 3\u0026ndash;4 months bare soil phase after harvest in croplands, reducing water loss by transpiration, as compared to a longer period of ET losses in grasslands through the continuous transpiration of perennial plant species (Spera et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Wu et al., \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Our results showed that croplands had 4.0-31.9% higher average soil water content than grasslands across a 10\u0026ndash;110 cm depth profile. This finding primarily supported the second hypothesis that croplands maintain a higher soil water content than grasslands on an annual basis. These results are in accordance with previous studies showing that the conversion of cropland to grassland primarily lead to reduced soil moisture (Gao et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Wang et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2024b\u003c/span\u003e). Another reason for higher soil water content in croplands may be reduced rainfall interception due to sparser and seasonal variable canopies, which allow for more efficient precipitation infiltration into the topsoil, especially during early growth stages (Lin et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Lian et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The third reason may be the lower root water uptake on the annual basis in croplands than grasslands due to the lower root length density (Bayala and Prieto, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), which might be more critical in coarse textured soil with less efficient water flow through soil due to low unsaturated hydraulic conductivity. Notably, the differences in soil water content among different land-use types became increasingly pronounced with depth and were especially evident during dry years, further supporting the legacy effects of deep soil water storage discussed earlier and suggesting that soil water \u0026ldquo;legacy effects\u0026rdquo; following droughts are influenced by both vegetation phenology and water use patterns (Zeng et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFurthermore, land-use intensity significantly influences the deep soil water storage, particularly during the climate extremes. The organic farming without mineral fertilization consistently exhibited higher soil water content in deep soil layer compared to conventional farming (highest land-use intensity), particularly during the dry years of 2022 and 2023, likely because nutrient limitations constrained plant growth and thus reduced transpiration and overall ET demand (Barton et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). In grassland systems, intensive meadow with fertilization, only four forage grass species, and frequent mowing showed lower soil water levels in the topsoil but significantly and consistently higher soil water content in the deeper layers than unfertilized extensive meadows and pastures (lowest land-use intensity). This contrast may be explained by a combination of factors: reduced rooting depth due to early stage of root system development, rapid shoot re-growth after frequent mowing events, and regular fertilization that promote aboveground biomass production and increase water uptake from shallow layers at initially low transpiration rates (Rose et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Prechsl et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Fan et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Additionally, limited canopy cover after mowing reduces interception surface shading, while low plant diversity constrains vertical ET buffering (He and Richards, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This pattern likely also reflects the relatively recent establishment of the intensive meadow, which was ploughed and re-sown in 2020, and may not yet have developed fully functional and deeper rooting networks characteristic of older, undisturbed grassland (Korell et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Notably, the differences in the deep soil water storage between intensive meadow and extensive grasslands were amplified following the heavy rainfall event in August 2023, with the discrepancy becoming even stronger under the wettest conditions in 2024. The capacity of land-use systems to retain or deplete deep soil water following recharge events depend on vegetation community composition and root distribution, which in turn affect infiltration pathways and soil hydraulic properties (Fischer et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Fischer et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2014b\u003c/span\u003e; Cui et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Shi et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Thus, our findings suggest that both land-use types and intensities modulated the capacity to access and retain deep water, with croplands and intensive grasslands retaining more water than extensive grasslands due to lower transpiration and shallower rooting.\u003c/p\u003e\u003cp\u003eOur results indicate that the future climate scenario further modulated deep soil water storage through interactions with land-use types, especially during the extreme events, by the altered precipitation regimes and increased temperature. Under conventional farming, the future climate scenario exhibited higher deep soil water content than ambient climate when the legacy effect of drought was most dominant (January to June 2023), likely due to the added irrigation during fall and spring, which allowed deeper infiltration and water retention when atmospheric demand was low (Peterson and Westfall, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). In contrast, intensive meadow under the future scenario exhibited lower deep soil water content following heavy rain events (August 2023 and June 2024), as 20% of summer rain events were shielded and large events cause large absolute differences. The future climate scenario effect might be greater in intensive meadow than in extensive meadow, probably due to the dominance of shallow-rooted grass species in these systems, which more rapidly depleted surface moisture but were less effective at utilizing or retaining water at depth. Additionally, lower plant diversity and frequent mowing could have influenced water use patterns, leading to less effective recharge in deeper layer under high-intensity rainfall (Rodriguez-Iturbe et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). These patterns highlight the importance of both seasonal deep soil water dynamics and system-specific water use strategies in mediating land-use responses to altered precipitation regimes. Deep soil water storage acts as a critical buffer during climatic extremes, but its effectiveness depends heavily on land-use practices, rooting depth distributions and vegetation composition.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003e4.3 Responses of plant productivity on climate extreme and land-use-based adjustment strategies\u003c/h2\u003e\u003cp\u003eOur results supported the third hypothesis: croplands generally produced 47.0-63.5% higher yields than grasslands, reflecting crop breeding for high resource use efficiency and rapid aboveground biomass production within a short growth period (Leakey et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In contrast, grasslands comprising perennial and diverse species, exhibited extended growth periods and lower yield optimization. Productivity was further influenced by land-use intensity, with the lowest yields observed in extensive meadows across years, likely due to lower initial water content and the absence of fertilization. Furthermore, yield was modulated by interannual climate variability and its interaction with land-use systems. Highest yields occurred in the normal year 2024, while significant declines were observed in the drought year 2022 and to a lesser extent in 2023. Notably, the decrease of yields in 2022 relative to 2024 was similar (~\u0026thinsp;30%) across croplands and grasslands, indicating that the extensive grasslands did not exhibit higher resistance to climate extremes. This contradicts our hypothesis and previous findings from the same field site (Isbell et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Korell et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). That is probably due to the differences in the study period (2015\u0026ndash;2022), the intensive meadow system being established in 2020, and the fact that the yield analyzed in the current study was based on machine harvesting, as is usual in agriculture. In contrast, Korell et al. (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) analyzed manually harvested biomass cut close to the ground level, which also included the pastures.\u003c/p\u003e\u003cp\u003eDeep soil water storage partially buffered yield losses against drought extremes. In 2022, the pronounced decline in deep soil water storage indicated its use to maintain productivity during extreme drought. In 2023, despite a summer drought, yields were higher, likely due to uptake of topsoil water and sustained by the infiltration of winter and spring precipitation, with the wetting front reaching the deep soil by March, and supplemented by late-season rainfall recharge. In contrast, the normal conditions in 2024 eliminated the need to access deeper water reservoirs. Our SEM revealed that extreme climates exerted indirect effect on yield through changes in both topsoil and deep soil water storage during the growing season. These findings suggested that deep soil water provides critical buffering, but is insufficient to fully compensate for severe topsoil moisture deficits, particularly under prolonged drought.\u003c/p\u003e\u003cp\u003eAdditional coping mechanisms, such as shifts in root patterns, plant community composition, and improved WUE are essential under extreme drought (Zwicke et al., \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Qi et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). We observed positive correlation between RLD and net change in water content during peak growth, emerging under grasslands in 2023 and croplands in 2022, indicating that root water uptake shifted between topsoil and deep layer depending on drought severity (Zwicke et al., \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Bristiel et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In grasslands, deep-rooted perennial species, especially the forbs, likely facilitate deep soil water uptake during extreme drought through their dense and persistent root system networks (Barkaoui et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Hanslin et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; K\u0026uuml;nzi et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). However, deeper rooting may not be the dominant adjustment mechanism, as efficient water uptake can occur with sparse roots, and high RLD is often more critical for nutrient acquisition (Jorda et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In croplands, RLD varied more across years, with higher values in the topsoil in 2023 linked to better moisture availability. Yet, yield variation likely reflects both climate-driven root responses and crop-specific traits, such as finer roots under organic management (Wong et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). We observed that the total vegetation cover and relative abundance of legumes slightly increased in 2023, suggesting the ability of grasslands with high species richness to undergo year-to-year changes in species composition depending on environmental conditions and species interactions. However, this interpretation is limited by several factors. RLD data only extend to a depth of 50 cm and may be influenced by the effects of crop rotation. Furthermore, vegetation community data are only available for grasslands and do not include observations from 2024. This restricts our ability to fully assess root and community-level responses.\u003c/p\u003e\u003cp\u003eCroplands exhibited significantly higher ET than grasslands in 2022 and 2023, likely due to higher transpiration from fast-growing annual crops with greater leaf area (Rajan et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). In 2024, ET rates became comparable across all land-use types, suggesting that increased transpiration potential in grasslands (Spera et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Croplands consistently had a higher WUE than grasslands, and the WUE showed higher correlation coefficient with yield than RLD, supporting our fourth hypothesis that adaptations to drought and deep soil water limitations would rely more on WUE than on shifts in rooting patterns or plant community. This likely reflects breeding strategies that enhance biomass production, drought tolerance, and stomatal regulation (Turner, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Haworth et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In contrast, deep-rooted and drought-resilient grassland perennials often show higher ET losses and lower yield responsiveness, as well as the water loss during maintenance without growth, may contribute to lower WUE (Ponton et al., \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Poppe Ter\u0026aacute;n et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Furthermore, extensive grasslands include a high proportion of perennial species capable of clonal regeneration through rhizomes or other belowground tissues, which can sustain transpiration without proportional aboveground biomass gain (Volaire, \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Notably, WUE was 65.3\u0026ndash;87.4% higher in intensive meadow than extensive meadow, probably due to the high frequency of mowing, and the application of fertilizer promoting aboveground growth during the peak productivity (Rose et al., \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe future climate scenario showed limited effect on yield but altered water cycling. In 2024, yields were unaffected by the reduced in precipitation, likely due to other limiting factors such as nutrients regulating plant growth during this normal year (Basso and Ritchie, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Fu et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). During the drought years, legacy effects of additional irrigation in autumn and spring became more relevant, causing greater deep soil water storage, which elevated the ET in the growing season to levels comparable to ambient climate, rendering its effect on yield again negligible. However, root biomass and plant community composition remained unchanged, possibly due to shallow sampling, moderate climatic changes, and ecosystem buffering capacity (K\u0026uuml;hn et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eOverall, while deep soil water storage provided partial drought buffering, WUE optimization emerges as the dominant adjustment strategy under climate extremes. High-intensity systems, such as conventional farming, maintain productivity through efficient water use and flexible responses to future scenarios. These findings highlight the need for adaptive management to sustain productivity under intensifying climate variability.\u003c/p\u003e\u003c/div\u003e"},{"header":"5. CONCLUSION","content":"\u003cp\u003eOur study highlights the role of deep soil water storage in buffering plant productivity against climate extremes. Unlike shallow soil moisture, the storage of water in deeper soil responds more slowly to climatic fluctuations. This can help to sustain the supply of water to plants and their productivity during prolonged droughts. However, this buffering capacity is not unlimited. Extreme droughts can thoroughly exploit deep storage, particularly when topsoil moisture is severely depleted. This drives deep soil water content toward the plant wilting point, resulting in a situation that persists until the subsequent year. Conventional farming can enhance deep soil water uptake during drought extremes by proper root systems and increased fertilization, whereas intensive meadow management was beneficial to deep soil water retention through higher mowing frequency. These practices enhance water uptake efficiency and help maintain productivity under extreme conditions. The future climate scenarios had weaker impact on the water cycle than climate extremes. Management strategies such as optimizing water use efficiency and promoting species or cultivars with appropriate growth spans and rooting systems that align with seasonal water availability can help stabilize productivity. These findings underscore the importance of deep soil water storage and context-specific management practices for maintaining ecosystem resistance and sustainability in an increasingly variable climate.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eDECLARATION OF COMPETING INTEREST\u003c/h2\u003e\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eACKNOWLEDGEMENTS\u003c/h2\u003e\u003cp\u003eWe appreciate the Helmholtz Association, the Federal Ministry of Education and Research, the State Ministry of Science and Economy of Saxony-Anhalt and the State Ministry for Higher Education, Research and the Arts Saxony to fund the Global Change Experimental Facility (GCEF) project. We thank the staff of the Bad Lauchst\u0026auml;dt Experimental Research Station (especially Ines Merbach and Konrad Kirsch) for their work in in maintaining the plots and infrastructures of the GCEF, and Harald Auge, Fran\u0026ccedil;ois Buscot, and Stefan Klotz for their role in setting up the GCEF. We thank Ralf Gruendling, Max Koehne, and Eric Braatz from the UFZ, Tobias Klauder at the Technical University of Dresden, Vanessa Schwanengel, Jakob Apelt and Elin Briese at the Martin Luther University Halle-Wittenberg for assistance during field work and laboratory analyses.\u003c/p\u003e\n\u003ch3\u003eAUTHORSHUIP CONTRIBUTION STATAMENT\u003c/h3\u003e\n\u003cp\u003e\u003cb\u003eMengqi Wu\u003c/b\u003e: Writing \u0026ndash; original draft, Visualization, Validation, Investigation, Formal analysis, Data curation, Methodology. \u003cb\u003eChristiane Roscher\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Investigation, Data curation, Validation, Methodology. \u003cb\u003eMartin Sch\u0026auml;dler\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Investigation, Data curation, Validation. \u003cb\u003eMika Tarkka\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Methodology, Conceptualization. \u003cb\u003eDoris Vetterlein\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Methodology, Conceptualization. \u003cb\u003eSteffen Schl\u0026uuml;ter\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Writing \u0026ndash; original draft, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhmed, M.A., Zarebanadkouki, M., Meunier, F., Javaux, M., Kaestner, A., Carminati, A., 2018. Root type matters: measurement of water uptake by seminal, crown, and lateral roots in maize. Journal of Experimental Botany 69, 1199\u0026ndash;1206.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAli, G., Wang, Z., Li, X., Jin, N., Chu, H., Jing, L., 2021. Deep soil water deficit and recovery in alfalfa fields of the Loess Plateau of China. Field Crops Research 260, 107990.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAltermann, M., Rinklebe, J., Merbach, I., K\u0026ouml;rschens, M., Langer, U., Hofmann, B., 2005. Chernozem\u0026mdash;Soil of the Year 2005. Journal of Plant Nutrition and Soil Science 168, 725\u0026ndash;740.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBagley, J.E., Kueppers, L.M., Billesbach, D.P., Williams, I.N., Biraud, S.C., Torn, M.S., 2017. The influence of land cover on surface energy partitioning and evaporative fraction regimes in the U.S. Southern Great Plains. Journal of Geophysical Research: Atmospheres 122, 5793\u0026ndash;5807.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarkaoui, K., Roumet, C., Volaire, F., 2016. Mean root trait more than root trait diversity determines drought resilience in native and cultivated Mediterranean grass mixtures. Agriculture, Ecosystems \u0026amp; Environment 231, 122\u0026ndash;132.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarton, L., Wan, G.G.Y., Buck, R.P., Colmer, T.D., 2009. Nitrogen Increases Evapotranspiration and Growth of a Warm-Season Turfgrass. Agronomy Journal 101, 17\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBasso, B., Ritchie, J.T., 2018. Evapotranspiration in High-Yielding Maize and under Increased Vapor Pressure Deficit in the US Midwest. Agricultural \u0026amp; Environmental Letters 3, 170039.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBastos, A., Ciais, P., Friedlingstein, P., Sitch, S., Pongratz, J., Fan, L., Wigneron, J.P., Weber, U., Reichstein, M., Fu, Z., Anthoni, P., Arneth, A., Haverd, V., Jain, A.K., Joetzjer, E., Knauer, J., Lienert, S., Loughran, T., McGuire, P.C., Tian, H., Viovy, N., Zaehle, S., 2020. Direct and seasonal legacy effects of the 2018 heat wave and drought on European ecosystem productivity. Science Advances 6, eaba2724.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBates, D., 2005. Fitting linear mixed models in R. R News 5, 27\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBayala, J., Prieto, I., 2019. Water acquisition, sharing and redistribution by roots: applications to agroforestry systems. Plant and Soil 453, 17\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBazzichetto, M., Sperandii, M.G., Penone, C., Keil, P., Allan, E., Lepš, J., Prati, D., Fischer, M., Bolliger, R., Gossner, M.M., de Bello, F., 2024. Biodiversity promotes resistance but dominant species shape recovery of grasslands under extreme drought. Journal of Ecology 112, 1087\u0026ndash;1100.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeguer\u0026iacute;a, S., Vicente-Serrano, S.M., Maintainer Santiago, B., 2017. Package \u0026lsquo;spei\u0026rsquo;. Calculation of the Standardised Precipitation-Evapotranspiration Index.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBeguer\u0026iacute;a, S., Vicente-Serrano, S.M., Reig, F., Latorre, B., 2014. Standardized precipitation evapotranspiration index (SPEI) revisited: parameter fitting, evapotranspiration models, tools, datasets and drought monitoring. International Journal of Climatology 34, 3001\u0026ndash;3023.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBens, O., Wahl, N.A., Fischer, H., H\u0026uuml;ttl, R.F., 2006. Water infiltration and hydraulic conductivity in sandy cambisols: impacts of forest transformation on soil hydrological properties. European Journal of Forest Research 126, 101\u0026ndash;109.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBristiel, P., Roumet, C., Violle, C., Volaire, F., 2018. Coping with drought: root trait variability within the perennial grass Dactylis glomerata captures a trade-off between dehydration avoidance and dehydration tolerance. Plant and Soil 434, 327\u0026ndash;342.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrookshire, E.N., Weaver, T., 2015. Long-term decline in grassland productivity driven by increasing dryness. Nature Communications 6, 7148.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCarminati, A., Zarebanadkouki, M., Kroener, E., Ahmed, M.A., Holz, M., 2016. Biophysical rhizosphere processes affecting root water uptake. Annals of Botany 118, 561\u0026ndash;571.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, L., Brun, P., Buri, P., Fatichi, S., Gessler, A., McCarthy, M.J., Pellicciotti, F., Stocker, B., Karger, D.N., 2025. Global increase in the occurrence and impact of multiyear droughts. Science 387, 278\u0026ndash;284.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCondon, A.G., Richards, R.A., Rebetzke, G.J., Farquhar, G.D., 2004. Breeding for high water-use efficiency. Journal of Experimental Botany 55, 2447\u0026ndash;2460.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCraine, J.M., Ocheltree, T.W., Nippert, J.B., Towne, E.G., Skibbe, A.M., Kembel, S.W., Fargione, J.E., 2012. Global diversity of drought tolerance and grassland climate-change resilience. Nature Climate Change 3, 63\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCui, Z., Wu, G.-L., Huang, Z., Liu, Y., 2019. Fine roots determine soil infiltration potential than soil water content in semi-arid grassland soils. Journal of Hydrology 578, 124023.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eD\u0026ouml;scher, R.W.N., U.; Jones, C.; Rutgersson, A.; Meier, H. E. M.; Hansson, U., 2002. The Development of The Coupled Ocean-atmosphere Model Rcao For The Northern European Area. EGS XXVII General Assembly, Nice.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDraye, X., Kim, Y., Lobet, G., Javaux, M., 2010. Model-assisted integration of physiological and environmental constraints affecting the dynamic and spatial patterns of root water uptake from soils. Journal of Experimental Botany 61, 2145\u0026ndash;2155.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan, J., Wang, Q., Jones, S.B., Shao, M., 2015. Soil water depletion and recharge under different land cover in China's Loess Plateau. Ecohydrology 9, 396\u0026ndash;406.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFan, Y., Miguez-Macho, G., Jobbagy, E.G., Jackson, R.B., Otero-Casal, C., 2017. Hydrologic regulation of plant rooting depth. Proceedings of the National Academy of Sciences 114, 10572\u0026ndash;10577.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFeldman, A.F., Feng, X., Felton, A.J., Konings, A.G., Knapp, A.K., Biederman, J.A., Poulter, B., 2024. Plant responses to changing rainfall frequency and intensity. Nature Reviews Earth \u0026amp; Environment 5, 276\u0026ndash;294.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFigueroa-Bustos, V., Palta, J.A., Chen, Y., Stefanova, K., Siddique, K.H.M., 2020. Wheat Cultivars With Contrasting Root System Size Responded Differently to Terminal Drought. Frontiers in Plant Science 11, 1285.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFischer, C., Roscher, C., Jensen, B., Eisenhauer, N., Baade, J., Attinger, S., Scheu, S., Weisser, W.W., Schumacher, J., Hildebrandt, A., 2014. How do earthworms, soil texture and plant composition affect infiltration along an experimental plant diversity gradient in grassland? PLoS One 9, e98987.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFischer, C., Tischer, J., Roscher, C., Eisenhauer, N., Ravenek, J., Gleixner, G., Attinger, S., Jensen, B., de Kroon, H., Mommer, L., Scheu, S., Hildebrandt, A., 2014b. Plant species diversity affects infiltration capacity in an experimental grassland through changes in soil properties. Plant and Soil 397, 1\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFu, Z., Ciais, P., Feldman, A.F., Gentine, P., Makowski, D., Prentice, I.C., Stoy, P.C., Bastos, A., Wigneron, J.P., 2022. Critical soil moisture thresholds of plant water stress in terrestrial ecosystems. Science Advances 8, eabq7827.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGao, X., Wu, P., Zhao, X., Wang, J., Shi, Y., 2011. Effects of Land Use on Soil Moisture Variations in a Semi-Arid Catchment: Implications for Land and Agricultural Water Management. Land Degradation \u0026amp; Development 25, 163\u0026ndash;172.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGood, S.P., Noone, D., Bowen, G., 2015. WATER RESOURCES. Hydrologic connectivity constrains partitioning of global terrestrial water fluxes. Science 349, 175\u0026ndash;177.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHanslin, H.M., Bischoff, A., Hovstad, K.A., 2019. Root growth plasticity to drought in seedlings of perennial grasses. Plant and Soil 440, 551\u0026ndash;568.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHargreaves, G.H., Allen, R.G., 2003. History and evaluation of Hargreaves evapotranspiration equation. Journal of Irrigation and Drainage Engineering 129, 53\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaworth, M., Marino, G., Loreto, F., Centritto, M., 2021. Integrating stomatal physiology and morphology: evolution of stomatal control and development of future crops. Oecologia 197, 867\u0026ndash;883.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHe, S., Richards, K., 2015. Impact of Meadow Degradation on Soil Water Status and Pasture Management\u0026mdash;A Case Study in Tibet. Land Degradation \u0026amp; Development 26, 468\u0026ndash;479.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHolmes, T.H., Rice, K.J., 1996. Patterns of growth and soil-water utilization in some exotic annuals and native perennial bunchgrasses of California. Annals of Botany 78, 233\u0026ndash;243.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIsbell, F., Craven, D., Connolly, J., Loreau, M., Schmid, B., Beierkuhnlein, C., Bezemer, T.M., Bonin, C., Bruelheide, H., de Luca, E., Ebeling, A., Griffin, J.N., Guo, Q., Hautier, Y., Hector, A., Jentsch, A., Kreyling, J., Lanta, V., Manning, P., Meyer, S.T., Mori, A.S., Naeem, S., Niklaus, P.A., Polley, H.W., Reich, P.B., Roscher, C., Seabloom, E.W., Smith, M.D., Thakur, M.P., Tilman, D., Tracy, B.F., van der Putten, W.H., van Ruijven, J., Weigelt, A., Weisser, W.W., Wilsey, B., Eisenhauer, N., 2015. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574\u0026ndash;577.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJacob, D., Podzun, R., 1997. Sensitivity studies with the regional climate model REMO. Meteorology and Atmospheric Physics 63, 119\u0026ndash;129.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJorda, H., Ahmed, M.A., Javaux, M., Carminati, A., Duddek, P., Vetterlein, D., Vanderborght, J., 2022. Field scale plant water relation of maize (Zea mays) under drought \u0026ndash; impact of root hairs and soil texture. Plant and Soil 478, 59\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKano-Nakata, M., Nakamura, T., Mitsuya, S., Yamauchi, A., 2019. Plasticity in root system architecture of rice genotypes exhibited under different soil water distributions in soil profile. Plant Production Science 22, 501\u0026ndash;509.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKorell, L., Andrzejak, M., Berger, S., Durka, W., Haider, S., Hensen, I., Herion, Y., Hofner, J., Kindermann, L., Klotz, S., Knight, T.M., Linstadter, A., Madaj, A.M., Merbach, I., Michalski, S., Plos, C., Roscher, C., Schadler, M., Welk, E., Auge, H., 2024. Land use modulates resistance of grasslands against future climate and inter-annual climate variability in a large field experiment. Global Change Biology 30, e17418.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eK\u0026uuml;hn, N., Spiegel, M.P., Tovar, C., Willis, K.J., Macias-Fauria, M., 2022. Seeing roots from space: aboveground fingerprints of root depth in vegetation sensitivity to climate in dry biomes. Environmental Research Letters 17, 114062.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eK\u0026uuml;nzi, Y., Zeiter, M., Fischer, M., Stampfli, A., 2025. Rooting depth and specific leaf area modify the impact of experimental drought duration on temperate grassland species. Journal of Ecology 113, 445\u0026ndash;458.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeakey, A.D.B., Ferguson, J.N., Pignon, C.P., Wu, A., Jin, Z., Hammer, G.L., Lobell, D.B., 2019. Water Use Efficiency as a Constraint and Target for Improving the Resilience and Productivity of C(3) and C(4) Crops. Annual Review of Plant Biology 70, 781\u0026ndash;808.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLefcheck, J., Byrnes, J., Grace, J., 2015. piecewiseSEM: Piecewise Structural Equation Modeling. R package version 1.1.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLenth, R., Henrik, S., Jonathon, L., Paul, B., Maxime, H., 2018. Emmeans: Estimated marginal means. AKA least-squares means 1.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, B., Zhang, X., Morita, S., Sekiya, N., Araki, H., Gu, H., Han, J., Lu, Y., Liu, X., 2022. Are crop deep roots always beneficial for combating drought: A review of root structure and function, regulation and phenotyping. Agricultural Water Management 271, 107781.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Y., Fan, J., Hu, Z., Shao, Q., Zhang, L., Yu, H., 2015. Influence of Land Use Patterns on Evapotranspiration and Its Components in a Temperate Grassland Ecosystem. Advances in Meteorology 2015, 1\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, Y., Li, K., Zhou, Q., Zhao, Y., Cai, L., Yang, Z., 2024. Spatiotemporal dynamics and similarity in soil moisture in shallow soils on karst slopes. Journal of Hydrology 639, 131655.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLian, X., Zhao, W., Gentine, P., 2022. Recent global decline in rainfall interception loss due to altered rainfall regimes. Nature Communications 13, 7642.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLin, M., Sadeghi, S.M.M., Van Stan, J.T., 2020. Partitioning of Rainfall and Sprinkler-Irrigation by Crop Canopies: A Global Review and Evaluation of Available Research. Hydrology 7, 76\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu, Y., Zhang, Y., Pe\u0026ntilde;uelas, J., Kannenberg, S.A., Gong, H., Yuan, W., Wu, C., Zhou, S., Piao, S., 2025. Drought legacies delay spring green-up in northern ecosystems. Nature Climate Change 15, 444\u0026ndash;451.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLynch, J.P., 2013. Steep, cheap and deep: an ideotype to optimize water and N acquisition by maize root systems. Annals of Botany 112, 347\u0026ndash;357.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLynch, J.P., 2019. Root phenotypes for improved nutrient capture: an underexploited opportunity for global agriculture. New Phytologist 223, 548\u0026ndash;564.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaestre, F.T., Eldridge, D.J., Soliveres, S., Kefi, S., Delgado-Baquerizo, M., Bowker, M.A., Garcia-Palacios, P., Gaitan, J., Gallardo, A., Lazaro, R., Berdugo, M., 2016. Structure and functioning of dryland ecosystems in a changing world. Annual Review of Ecology, Evolution, and Systematics 47, 215\u0026ndash;237.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMaurel, C., Nacry, P., 2020. Root architecture and hydraulics converge for acclimation to changing water availability. Nature Plants 6, 744\u0026ndash;749.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMuller, L.M., Bahn, M., 2022. Drought legacies and ecosystem responses to subsequent drought. Global Change Biology 28, 5086\u0026ndash;5103.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePalta, J.A., Turner, N.C., 2018. Crop root system traits cannot be seen as a silver bullet delivering drought resistance. Plant and Soil 439, 31\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeterson, G.A., Westfall, D.G., 2005. Managing precipitation use in sustainable dryland agroecosystems. Annals of Applied Biology 144, 127\u0026ndash;138.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePhalempin, M., Jentzsch, N., K\u0026ouml;hne, J.M., Schreiter, S., Gr\u0026uuml;ndling, R., Vetterlein, D., Schl\u0026uuml;ter, S., 2025. Soil structure development in a five-year chronosequence of maize cropping on two contrasting soil textures. Soil and Tillage Research 251, 106561.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePierret, A., Maeght, J.L., Clement, C., Montoroi, J.P., Hartmann, C., Gonkhamdee, S., 2016. Understanding deep roots and their functions in ecosystems: an advocacy for more unconventional research. Annals of Botany 118, 621\u0026ndash;635.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePonton, S., Flanagan, L.B., Alstad, K.P., Johnson, B.G., Morgenstern, K.A.I., Kljun, N., Black, T.A., Barr, A.G., 2006. Comparison of ecosystem water-use efficiency among Douglas‐fir forest, aspen forest and grassland using eddy covariance and carbon isotope techniques. Global Change Biology 12, 294\u0026ndash;310.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePoppe Ter\u0026aacute;n, C., Naz, B.S., Graf, A., Qu, Y., Hendricks Franssen, H.-J., Baatz, R., Ciais, P., Vereecken, H., 2023. Rising water-use efficiency in European grasslands is driven by increased primary production. Communications Earth \u0026amp; Environment 4, 95.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePrechsl, U.E., Burri, S., Gilgen, A.K., Kahmen, A., Buchmann, N., 2015. No shift to a deeper water uptake depth in response to summer drought of two lowland and sub-alpine C(3)-grasslands in Switzerland. Oecologia 177, 97\u0026ndash;111.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQi, J., Markewitz, D., Radcliffe, D., 2018. Modelling the effect of changing precipitation inputs on deep soil water utilization. Hydrological Processes 32, 672\u0026ndash;686.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajan, N., Maas, S.J., Cui, S., 2014. Extreme drought effects on summer evapotranspiration and energy balance of a grassland in the Southern Great Plains. Ecohydrology 8, 1194\u0026ndash;1204.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRasmussen, C.R., Thorup-Kristensen, K., Dresb\u0026oslash;ll, D.B., 2019. Uptake of subsoil water below 2 m fails to alleviate drought response in deep-rooted Chicory (Cichorium intybus L.). Plant and Soil 446, 275\u0026ndash;290.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRickard, W., Hossain, I., Zhang, X., Cooper, H.V., Mooney, S.J., Hawkesford, M.J., Whalley, W.R., 2025. Field plants strategically regulate water uptake from different soil depths by spatiotemporally adjusting their radial root hydraulic conductivity. New Phytologist 247, 546\u0026ndash;561.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRockel, B., Will, A., Hense, A., 2008. The Regional Climate Model COSMO-CLM (CCLM). Meteorologische Zeitschrift 17, 347\u0026ndash;348.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRodriguez-Iturbe, I., Porporato, A., Laio, F., Ridolfi, L., 2001. Intensive or extensive use of soil moisture: Plant strategies to cope with stochastic water availability. Geophysical Research Letters 28, 4495\u0026ndash;4497.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRose, L., Coners, H., Leuschner, C., 2011. Effects of fertilization and cutting frequency on the water balance of a temperate grassland. Ecohydrology 5, 64\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSch\u0026auml;dler, M., Buscot, F., Klotz, S., Reitz, T., Durka, W., Bumberger, J., Merbach, I., Michalski, S.G., Kirsch, K., Remmler, P., Schulz, E., Auge, H., 2019. Investigating the consequences of climate change under different land-use regimes: a novel experimental infrastructure. Ecosphere 10, e02635.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScherzinger, F., Schadler, M., Reitz, T., Yin, R., Auge, H., Merbach, I., Roscher, C., Harpole, W.S., Blagodatskaya, E., Siebert, J., Ciobanu, M., Marder, F., Eisenhauer, N., Quaas, M., 2024. Sustainable land management enhances ecological and economic multifunctionality under ambient and future climate. Nature Communications 15, 4930.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchlaepfer, D.R., Bradford, J.B., Lauenroth, W.K., Munson, S.M., Tietjen, B., Hall, S.A., Wilson, S.D., Duniway, M.C., Jia, G., Pyke, D.A., Lkhagva, A., Jamiyansharav, K., 2017. Climate change reduces extent of temperate drylands and intensifies drought in deep soils. Nature Communications 8, 14196.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchl\u0026uuml;ter, S., Vogel, H.-J., Ippisch, O., Vanderborght, J., 2013. Combined Impact of Soil Heterogeneity and Vegetation Type on the Annual Water Balance at the Field Scale. Vadose Zone Journal 12, 1\u0026ndash;17.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchnabel, F., Purrucker, S., Schmitt, L., Engelmann, R.A., Kahl, A., Richter, R., Seele-Dilbat, C., Skiadaresis, G., Wirth, C., 2022. Cumulative growth and stress responses to the 2018\u0026ndash;2019 drought in a European floodplain forest. Global Change Biology 28, 1870\u0026ndash;1883.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSegal, E., Kushnir, T., Mualem, Y., Shani, U., 2008. Water Uptake and Hydraulics of the Root Hair Rhizosphere. Vadose Zone Journal 7, 1027\u0026ndash;1034.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShi, X., Qin, T., Yan, D., Tian, F., Wang, H., 2021. A meta-analysis on effects of root development on soil hydraulic properties. Geoderma 403.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShipley, B., 2009. Confirmatory path analysis in a generalized multilevel context. Ecology 90, 363\u0026ndash;368.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSpera, S.A., Galford, G.L., Coe, M.T., Macedo, M.N., Mustard, J.F., 2016. Land-use change affects water recycling in Brazil's last agricultural frontier. Global Change Biology 22, 3405\u0026ndash;3413.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun, H., Shen, Y., Yu, Q., Flerchinger, G.N., Zhang, Y., Liu, C., Zhang, X., 2010. Effect of precipitation change on water balance and WUE of the winter wheat\u0026ndash;summer maize rotation in the North China Plain. Agricultural Water Management 97, 1139\u0026ndash;1145.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun, J., Liu, W., Pan, Q., Zhang, B., Lv, Y., Huang, J., Han, X., 2022. Positive legacies of severe droughts in the Inner Mongolia grassland. Science Advances 8, eadd6249.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun, L., Wang, R., Li, J., Wang, Q., Lyu, W., Wang, X., Cheng, K., Mao, H., Zhang, X., 2019. Reasonable fertilization improves the conservation tillage benefit for soil water use and yield of rain-fed winter wheat: A case study from the Loess Plateau, China. Field Crops Research 242, 107589.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun, L., Wang, S., Zhang, Y., Li, J., Wang, X., Wang, R., Lyu, W., Chen, N., Wang, Q., 2018. Conservation agriculture based on crop rotation and tillage in the semi-arid Loess Plateau, China: Effects on crop yield and soil water use. Agriculture, Ecosystems \u0026amp; Environment 251, 67\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSun, W., Zhou, S., Yu, B., Zhang, Y., Keenan, T., Fu, B., 2025. Soil moisture-atmosphere interactions drive terrestrial carbon-water trade-offs. Communications Earth \u0026amp; Environment 6, 169.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSwindon, J.G., Burke, I.C., Lauenroth, W.K., 2019. Seasonal Patterns of Root Production with Water and Nitrogen Additions Across Three Dryland Ecosystems. Ecosystems 22, 1664\u0026ndash;1675.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTissink, M., Radolinski, J., Reinthaler, D., Venier, S., Potsch, E.M., Schaumberger, A., Bahn, M., 2025. Individual Versus Combined Effects of Warming, Elevated CO(2) and Drought on Grassland Water Uptake and Fine Root Traits. Plant Cell \u0026amp; Environment 48, 2083\u0026ndash;2098.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTurner, N.C., 2004. Agronomic options for improving rainfall-use efficiency of crops in dryland farming systems. Journal of Experimental Botany 55, 2413\u0026ndash;2425.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan der Bom, F.J.T., Williams, A., Bell, M.J., 2020. Root architecture for improved resource capture: trade-offs in complex environments. Journal of Experimental Botany 71, 5752\u0026ndash;5763.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan Sundert, K., Arfin Khan, M.A.S., Bharath, S., Buckley, Y.M., Caldeira, M.C., Donohue, I., Dubbert, M., Ebeling, A., Eisenhauer, N., Eskelinen, A., Finn, A., Gebauer, T., Haider, S., Hansart, A., Jentsch, A., Kubert, A., Nijs, I., Nock, C.A., Nogueira, C., Porath-Krause, A.J., Radujkovic, D., Raynaud, X., Risch, A.C., Roscher, C., Scherer-Lorenzen, M., Schuchardt, M.A., Schutz, M., Siebert, J., Sitters, J., Spohn, M., Virtanen, R., Werner, C., Wilfahrt, P., Vicca, S., 2021. Fertilized graminoids intensify negative drought effects on grassland productivity. Global Change Biology 27, 2441\u0026ndash;2457.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVereecken, H., Amelung, W., Bauke, S.L., Bogena, H., Br\u0026uuml;ggemann, N., Montzka, C., Vanderborght, J., Bechtold, M., Bl\u0026ouml;schl, G., Carminati, A., Javaux, M., Konings, A.G., Kusche, J., Neuweiler, I., Or, D., Steele-Dunne, S., Verhoef, A., Young, M., Zhang, Y., 2022. Soil hydrology in the Earth system. Nature Reviews Earth \u0026amp; Environment 3, 573\u0026ndash;587.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVicente-Serrano, S.M., Beguer\u0026iacute;a, S., L\u0026oacute;pez-Moreno, J.I., 2010. A Multiscalar Drought Index Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index. Journal of Climate 23, 1696\u0026ndash;1718.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVicente-Serrano, S.M., Dom\u0026iacute;nguez-Castro, F., Beguer\u0026iacute;a, S., El Kenawy, A., Gimeno-Sotelo, L., Franquesa, M., Azorin-Molina, C., Andres-Martin, M., Halifa-Mar\u0026iacute;n, A., 2025. Atmospheric drought indices in future projections. Nature Water 3, 374\u0026ndash;387.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVolaire, F., 2018. A unified framework of plant adaptive strategies to drought: Crossing scales and disciplines. Global Change Biology 24, 2929\u0026ndash;2938.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang-Erlandsson, L., Tobian, A., van der Ent, R.J., Fetzer, I., te Wierik, S., Porkka, M., Staal, A., Jaramillo, F., Dahlmann, H., Singh, C., Greve, P., Gerten, D., Keys, P.W., Gleeson, T., Cornell, S.E., Steffen, W., Bai, X., Rockstr\u0026ouml;m, J., 2022. A planetary boundary for green water. Nature Reviews Earth \u0026amp; Environment 3, 380\u0026ndash;392.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, S., Fu, B.J., Gao, G.Y., Yao, X.L., Zhou, J., 2012. Soil moisture and evapotranspiration of different land cover types in the Loess Plateau, China. Hydrology and Earth System Sciences 16, 2883\u0026ndash;2892.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Y., Hu, W., Sun, H., Zhao, Y., Zhang, P., Li, Z., Zhou, Z., Tong, Y., Liu, S., Zhou, J., Huang, M., Jia, X., Clothier, B., Shao, M., Zhou, W., An, Z., 2024b. Soil moisture decline in China's monsoon loess critical zone: More a result of land-use conversion than climate change. Proceedings of the National Academy of Sciences 121, e2322127121.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Z., Ma, L., Liu, X., Xue, Z., Wu, J., Wen, X., 2024a. Grass (\u003cem\u003ePoa annua L.\u003c/em\u003e) cover for eight years as an effective strategy for recovering soil moisture. Geoderma 449, 117010.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWarter, M.M., Singer, M.B., Cuthbert, M.O., Roberts, D., Caylor, K.K., Sabathier, R., Stella, J., 2021. Drought onset and propagation into soil moisture and grassland vegetation responses during the 2012\u0026ndash;2019 major drought in Southern California. Hydrology and Earth System Sciences 25, 3713\u0026ndash;3729.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWeides, S.E., Hajek, T., Liancourt, P., Herberich, M.M., Kramp, R.E., Tomiolo, S., Pacheco-Riano, L.C., Tielborger, K., Majekova, M., 2024. Belowground niche partitioning is maintained under extreme drought. Ecology 105, e4198.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWickham, H., Chang, W., Wickham, M.H., 2016. Package \u0026lsquo;ggplot2\u0026rsquo;: An implementation of the Grammar of Graphics. Journal of Statistical Software 1, 189.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWong, K.M., Griffiths, M., Moran, A., Johnston, A., Liu, A.E., Sellers, M.A., Topp, C.N., 2023. Cover crop cultivars and species differ in root traits potentially impacting their selection for ecosystem services. Plant and Soil 500, 279\u0026ndash;296.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu, X., Liu, H., Li, X., Ciais, P., Babst, F., Guo, W., Zhang, C., Magliulo, V., Pavelka, M., Liu, S., Huang, Y., Wang, P., Shi, C., Ma, Y., 2018. Differentiating drought legacy effects on vegetation growth over the temperate Northern Hemisphere. Global Change Biology 24, 504\u0026ndash;516.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu, Z., Behzad, H.M., He, Q., Wu, C., Bai, Y., Jiang, Y., 2021. Seasonal transpiration dynamics of evergreen Ligustrum lucidum linked with water source and water-use strategy in a limestone karst area, southwest China. Journal of Hydrology 597, 126199.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWunsch, A., Liesch, T., Broda, S., 2022. Deep learning shows declining groundwater levels in Germany until 2100 due to climate change. Nature Communications 13, 1221.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang, Q., Fan, J., Luo, Z., 2024. Response of soil moisture and vegetation growth to precipitation under different land uses in the Northern Loess Plateau, China. Catena 236, 107728.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYin, R., Kardol, P., Thakur, M.P., Gruss, I., Wu, G.L., Eisenhauer, N., Schadler, M., 2020. Soil functional biodiversity and biological quality under threat: intensive land use outweighs climate change. Soil Biology \u0026amp; Biochemistry 147, 107847.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu, L., Leng, G., Tang, Q., Lu, C., Gao, Y., Yao, L., Fan, S., Dong, J., Tu, H., Ai, Z., Liao, X., 2025. Land-atmosphere interactions mitigate the direct surface-drying effects of land use and cover change. One Earth 8, 101210.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZeleke, K.T., Wade, L.J., 2012. Evapotranspiration Estimation Using Soil Water Balance, Weather and Crop Data, In: Irmak, A. (Ed.), Evapotranspiration \u0026ndash; Remote Sensing and Modeling. InTech, Rijeka, Croatia.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZeng, Z., Wu, W., Ge, Q., Li, Z., Wang, X., Zhou, Y., Zhang, Z., Li, Y., Huang, H., Liu, G., Pe\u0026ntilde;uelas, J., 2021. Legacy effects of spring phenology on vegetation growth under preseason meteorological drought in the Northern Hemisphere. Agricultural and Forest Meteorology 310, 108630.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang, G., Zhang, Y., Zhao, D., Liu, S., Wen, X., Han, J., Liao, Y., 2023. Quantifying the impacts of agricultural management practices on the water use efficiency for sustainable production in the Loess Plateau region: A meta-analysis. Field Crops Research 291, 108787.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao, C., Jia, X., Zhu, Y., Shao, M.a., 2017. Long-term temporal variations of soil water content under different vegetation types in the Loess Plateau, China. Catena 158, 55\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZuo, Q., Shi, J., Li, Y., Zhang, R., 2006. Root length density and water uptake distributions of winter wheat under sub-irrigation. Plant and Soil 285, 45\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZwicke, M., Picon-Cochard, C., Morvan-Bertrand, A., Prud'homme, M.P., Volaire, F., 2015. What functional strategies drive drought survival and recovery of perennial species from upland grassland? Annals of Botany 116, 1001\u0026ndash;1015.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"climate change, cropland, grassland, water cycling, root length density, plant community composition, yield","lastPublishedDoi":"10.21203/rs.3.rs-7062058/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7062058/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eClimate extremes, including multiyear droughts and extreme rainfall events, are projected to intensify, threatening the global water cycle and reducing agricultural productivity. Deep soil water storage plays a key role in buffering extremes, yet its influence on plant productivity and water use across land-use systems remains insufficiently understood. Here, we monitored soil moisture dynamics over three years and vegetation responses in a long-term field trial comprising five land-use types (two croplands: conventional \u0026amp; organic farming; three grasslands: intensive meadow, extensive meadow \u0026amp; pasture). The monitoring period captured both prolonged droughts and an extreme rainfall. We found strong legacy effects of past droughts on deep soil water storage (30\u0026ndash;110 cm), which decoupled plant productivity from short-term climate fluctuations. Extensive grasslands exploited the deep soil water storage more efficiently than intensive grasslands and croplands, because of longer transpiration demand and higher interception caused by the perennial vegetation cover. In turn, water use efficiency increased with land-use intensity, driven by shorter growing periods in croplands and higher mowing frequency in intensive grasslands. Our findings highlight how land-use practices shape ecosystem responses to climate extremes and underscore the need to incorporate deep soil water dynamics into sustainable land-management strategies under future climate conditions.\u003c/p\u003e","manuscriptTitle":"Legacy effects of climate extremes on deep soil water storage and water use efficiency across different land-use systems","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-21 17:33:29","doi":"10.21203/rs.3.rs-7062058/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7816182d-1a68-4f2a-bfd0-ecd236f822c5","owner":[],"postedDate":"July 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":51668419,"name":"Earth and environmental sciences/Climate sciences/Hydrology"},{"id":51668420,"name":"Earth and environmental sciences/Ecology/Ecosystem services"},{"id":51668421,"name":"Scientific community and society/Agriculture"},{"id":51668422,"name":"Earth and environmental sciences/Environmental social sciences/Climate-change adaptation"}],"tags":[],"updatedAt":"2025-10-23T15:37:08+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-21 17:33:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7062058","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7062058","identity":"rs-7062058","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}