Legacies of Precipitation Extremes: Divergent Responses Across Steppe Types

Legacies of Precipitation Extremes: Divergent Responses Across Steppe Types | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 23 September 2025 V1 Latest version Share on Legacies of Precipitation Extremes: Divergent Responses Across Steppe Types Authors : Ruojun Sun 0000-0002-9597-438X , Kuo Sun , Leren Liu , Feng Zhang , Yanhong Lou , Yuping Zhuge , and Zhenzhu Xu 0000-0001-5246-8981 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.175865459.93061204/v1 290 views 120 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Steppes are essential ecosystems that provide vital services, such as carbon sequestration, biodiversity support, and grazing resources. However, these ecosystems are highly susceptible to the effects of climate change, particularly extreme precipitation events, including droughts and heavy rainfall. This study integrates data from an 11-year desert steppe experiment with multi-decadal observations from China’s meadow, typical, and desert steppes to investigate the legacy effects of extreme precipitation on steppe productivity. The results demonstrate that aboveground net primary productivity (ANPP) responds differently to precipitation extremes across steppe types. In the meadow steppe, ANPP follows a linear relationship with precipitation, whereas in typical and desert steppes, the relationship becomes saturating. Ecosystem resilience varies substantially across the steppes: desert steppe exhibits rapid recovery after drought, meadow steppe shows persistent suppression of productivity, and typical steppe experiences community restructuring. Both drought and excessive rainfall disrupt plant phenology and competition, resulting in long-term changes to ecosystem productivity. Structural equation modeling reveals that ANPP serves as the primary driver of legacy effects, with plant functional group (PFG) shifts influenced by climatic variables. Specifically, grass ratios exacerbate the legacy effects, while higher plant diversity indices appear to mitigate them. This study emphasizes the importance of understanding vegetation-type-specific responses to extreme precipitation in accurately predicting the ecological consequences of climate change. Our findings suggest that Earth System Models must incorporate vegetation-type-specific parameterizations to improve projections of long-term productivity outcomes under future climate scenarios. Overall, this research provides important insights into the resilience of steppe ecosystems and highlights the need for enhanced management strategies to mitigate the impacts of extreme precipitation events in the context of global climate change. Introduction Occupying almost 40% of the Earth’s land surface, grasslands are a major biome offering crucial ecosystem services, including carbon storage, soil nutrient retention, biodiversity support, and grazing resources (Bai & Cotrufo, 2022). However, grassland ecosystems exhibit high sensitivity to climate change, with rapid shifts in species composition and community structure in response to altered temperature and precipitation regimes (e.g., Tilman, Reich & Knops, 2006; Harpole & Tilman, 2007; Craine et al., 2013; Xu et al., 2019; Yu et al., 2025). These dynamics drive significant variations in aboveground net primary productivity (ANPP) —a critical metric of ecosystem function (Yu et al., 2025; Knapp et al., 2020). As a fundamental climatic driver, precipitation changes critically regulates ANPP, increasing rapidly at first, gradually reaching a productivity ceiling, and potentially even declining (Sala et al., 2012; Huxman et al., 2004; Nippert et al., 2006; Wilcox et al., 2015). Furthermore, precipitation patterns also modulate plant biodiversity, especially in these arid ecosystems, ultimately affecting ecosystem stability and sustainability (Craine et al., 2013; Loreau & De Mazancourt, 2013; Wang et al., 2023). Since the Industrial Revolution, intensified extreme precipitation events (e.g., droughts, heavy rain) have substantially impacted arid/semi-arid plant communities (Ma et al., 2020; Xu et al., 2016). For instance, in a steppe ecosystem, severe droughts compromise ecosystem stability, which in turn leads to declines in vegetation biomass and ecosystem degradation (Liu et al., 2024). In contrast, excessive rainfall can cause waterlogging and soil erosion, negatively affecting plant functional traits, plant growth, vegetation structure, and even community stability (Mesele et al., 2025). Additionally, the increasing frequency of these extreme events can disrupt ecological balance, thereby altering species composition and plant community structure (Furtak & Wolińska, 2023; Mahecha et al., 2024). This raises potential challenges for protecting the health of vulnerable terrestrial ecosystems and for sustainable management, particularly in the face of ongoing climate change. The productivity response of steppe ecosystems to extreme precipitation events is often characterized by a delayed recovery, a phenomenon known as the legacy effect (Canarini et al., 2021; Yu et al., 2021). This delayed response is marked by prolonged reductions in ecosystem productivity following periods of extreme precipitation (Hoover, Knapp & Smith, 2014; Sala et al., 2012). The underlying mechanisms involve adaptive adjustments in plant species composition, vegetation structure, and water-use efficiency within steppe ecosystems (Yu et al., 2025; Hoover, Pfennigwerth & Duniway, 2021; Su et al., 2022; Xu et al., 2015). However, whether drought legacy effects occur remains debated and the underlying mechanisms remain unclear (e.g., Yu et al., 2025; Sun et al., 2022; Hoover, Pfennigwerth & Duniway, 2021; Du et al., 2024; Reichmann, Sala & Peters, 2013). Thus, understanding this delayed productivity response is crucial for assessing the resilience and stability of steppe ecosystems in the context of climate change. This legacy effect can present significant challenges for predicting long-term ecosystem responses to extreme precipitation events in regions undergoing climatic shifts (Yu et al., 2025; Mahecha et al., 2024; Hoover, Pfennigwerth & Duniway, 2021). Numerous studies have explored the legacy effects of extreme precipitation events on steppe ecosystems, revealing that their impacts can persist beyond immediate effects and continue to significantly affect vegetation productivity and plant community dynamics (e.g., Yu et al., 2025; Furtak & Wolińska, 2023; Yu et al., 2021; Sun et al., 2023). Despite this growing body of research, notable gaps remain in our understanding of the relationship between precipitation extremes and vegetation productivity (Yu et al., 2025; Mahecha et al., 2024). A key limitation is the scarcity of comprehensive studies examining the multidimensional responses of steppe ecosystems during both pre-drought and post-drought periods, and conversely, during pre-wetting and post-wetting periods. While some studies have focused on the resistance and resilience to drought, they often overlook the temporal stability and complex structural transformations that can occur within these ecosystems (Mahecha et al., 2024). Additionally, understanding of the legacy effects associated with broad range of precipitation pattern remain incomplete. Although advances have been made in identifying the magnitude, direction, and duration of these effects, significant knowledge gaps remain regarding their broad ecological implications and underlying mechanisms, particularly concerning long-term multi-decadal observational datasets (Yu et al., 2025; Mahecha et al., 2024). Furthermore, the response mechanisms of different steppe types to extreme precipitation events remain underexplored. Comparative studies have revealed contrasting drought sensitivity patterns between Eurasian and North American steppes, highlighting the need for further in-depth investigations into the specific responses of different steppes (Yu et al., 2025). Moreover, the cumulative effects of extreme wetness events on steppe ecosystems remain poorly understood, necessitating further research to clarify their legacy effects on steppe productivity and vegetation dynamics (Yu et al., 2025; Liu et al., 2024). In light of the increasing frequency of extreme precipitation events under global change, this study specifically addresses three critical research questions: (1) Can nonlinear models effectively integrate long-term datasets from different steppe types to depict the relationships between precipitation and aboveground net primary productivity (PPT–ANPP)? (2) How do different steppe types respond to pre- and post-drought, and conversely, pre- and post- wetting? (3) What are the underlying mechanisms driving the legacy responses to both drought and wetting, particularly in relation to plant community composition, plant functional types, and climate variables? Thus, this investigation aims to advance our understanding of steppe ecosystem dynamics and provide novel and fundamental insights into the ecological impacts of extreme precipitation events amid continuing climate change. Study area The Inner Mongolia steppes, situated in northern China (spanning from 97°12′E to 126°04′E and 37°24′N to 53°23′N), comprise three major ecosystems: meadow, typical, and desert types of steppe, as depicted in Figure 1 (Li et al., 2020; Chen et al., 2021). To examine the effects of extreme drought and excessive precipitation on productivity loss, representative sampling was conducted across all three steppe types. The meadow steppe research was conducted at two key sites: the Hulunber Grassland Ecosystem National Observation and Research Station, affiliated with the Institute of Agricultural Resources and Regional Planning, and the Erguna Forest-Steppe Ecotone Research Station, part of the Institute of Applied Ecology, Chinese Academy of Sciences. The typical steppe research was conducted at the Inner Mongolia Grassland Ecosystem Research Station, located in the Xilin Gol steppe and operated by the Chinese Academy of Sciences. The desert steppe research was conducted at the Xilamuren steppe in Darhan Muminggan Joint Banner, Inner Mongolia. Figure 7. Geospatial distribution of steppe types and meteorological monitoring network in Northern China. The base color blocks represent steppe types: meadow steppe (dark green), typical steppe (light green), and desert steppe (light yellow). White circles with black centers indicate research plots, while blue circles containing building symbols denote meteorological stations. Data access The ANPP, plant functional groups (PFGs), and plant species diversity indices were primarily obtained from these experimental stations (Table S1). For the meadow steppe, we integrated long-term in-situ observation data during 2006-2015 (Bai et al., 2010; Bai et al., 2012), and 2015–2020 (Han et al., 2011), and our field observation data were obtained during 2021–2022. For the typical steppe, the plant dataset derived from long-term in-situ observations during 1980–2019, and precipitation manipulation experiment data during 2014–2019 (Sun et al., 2022; Sun et al., 2023). The additional observation data during 2005–2008 (Han et al., 2011) and during 1981–2012 (Yan et al., 2015) were also compiled. The desert steppe data were primarily obtained from our long-term precipitation manipulation experiment conducted from 2011 to 2021, with five distinct precipitation treatments applied annually, resulting in a cumulative total of 55 site-years (Yu et al., 2021). We defined plant community aboveground biomass at end of the peaking growth as the aboveground net productivity (ANPP, Hu et al., 2025; Sabzchi-Dehkharghani et al., 2024). Plants were classified into two groups based on their morphological characteristics: grass and forb. The proportions of each group in the community were determined. Furthermore, plats were grouped into C 3 and C 4 types according to their photosynthetic pathways, and their respective proportions in the community were also obtained. Plant species biodiversity was quantified using the following indices: the Patrick species richness index ( S ), the Shannon-Wiener diversity index ( H ’), and the Pielou evenness index ( J ’) (Whittaker,1972; Wilsey et al., 2005). These indices were calculated using the following formulas: S = N (1) Hʹ = - \(\sum{\text{P}_{\text{i}}\text{ln}\text{P}_{\text{i}}}\) (2) Jʹ = \(\frac{\text{Hʹ}}{\text{ln}\text{S}}\) (3) where N is the total number of plant species in the steppe plot, and P i is the importance value corresponding to plant species i in the plot. The importance value is the average of the relative coverage, height, and frequency of each species in the plot (Whittaker,1972). For climate data access during the experimental period and for long-term meteorological observations, including annual precipitation (AP), growing season precipitation (GSP), mean annual temperature (MAT), and growing season temperature (GST), the following principles were followed: if the literature provided original meteorological data, those values were selected. When unavailable, data from the nearest national meteorological station (https://data.cma.cn/) were used. For stations located too far away, interpolated meteorological data were employed through computational methods. The de Martonne aridity index (I dM ), also referred to as the de Martonne index, was used as a surrogate for a bioclimatic index (de Martonne, 1926). This index serves as an indicator of climate aridity. Elevated IdM values correspond to lower aridity and higher air humidity. The IdM is particularly well-suited for large-scale spatial assessments and effectively captures the interplay between vegetation and climate. Its calculation was as follows: I dM = \(\frac{\text{AP}}{\text{MAT}\text{\ +\ 10}}\) (4) Precipitation use efficiency (PUE) is a critical metric for evaluating vegetation productivity particularly in arid and semi-arid regions (Paruelo et al., 1999). It quantifies the vegetation’s ability to convert precipitation into biomass production. Higher PUE values indicate that vegetation can more effectively transform precipitation into biomass, thereby enhancing vegetation productivity (Paruelo et al., 1999). We calculated PUE for each plot using the formula: PUE = \(\frac{\text{ANPP}}{\text{AP}}\) (5) Finally, we determined the legacy effects of aboveground net primary productivity (Legacy ANPP ) for each year at the experimental plots. It was quantified as the difference between observed aboveground net primary (ANPP expected ) (Sun et al., 2022). ANPPexpected was estimated based on the overall relationship between long-term ANPP and growing season precipitation across steppe types (Table S2). The legacy values and long-term observations for the three steppe types were calculated as follows: Legacy ANPP = ANPP observed - ANPP expected (6) To enhance our understanding of how precipitation events influence legacy effects in steppe ecosystems, we calculated the interannual variation in growing season precipitation associated with legacy effects, specially defined as the current year’s growing season precipitation minus that of the previous year. Positive values indicate an increase in precipitation during the current growing season relative to the preceding year, while negative values denote a corresponding decrease in precipitation. Calculations and statistics The data processing and analysis in this study were executed within the R language environment (version 4.5.1, R Core Team, 2014). The core workflow includes data preprocessing and modeling through the construction of structured data frameworks using the tibble package. Data manipulation—including filtering for vegetation types and extreme climate years, grouped aggregation via ”group_by()” and ”summarise()”, column operations, and integration of Tukey test results with ”left_join()”—was performed using dplyr to ensure data uniformity and downstream compatibility. Missing values were addressed using the mice package’s random forest imputation via ”(mice(…, method = ”rf”))”, generating five complete datasets where the fifth imputation result ”(complete(imp_total, 5))” was retained for reproducibility. For meadow steppe, linear models ”(lm())” quantified relationships between ANPP and environmental factors, extracting intercepts, slopes, and statistical significance. For typical and desert steppes, saturation curves were fitted using nonlinear models ”(nls())”, with prediction intervals calculated using the investr package’s ”predFit(…, interval = ”prediction”)” to delineate ecological thresholds. Statistical inference employed one-way ANOVA ”(aov())” on grouped data, with post hoc multiple comparisons conducted using the ”HSD.test()” from package agricolae and ”TukeyHSD()” from package multcomp . The ”multcompLetters4()” function from multcompView generated compact letter displays to signify group-wise differences. Piecewise structural equation modeling utilized piecewiseSEM , integrating mixed-effects models with the nlme package’s ”lme()” to incorporate year as a random effect and vegetation type as a grouping variable. Variables were standardized to eliminate scale discrepancies, enabling quantification of multi-path effects linking environmental factors, PFGs, and ANPP. Global model fit was assessed via ”fisherC()” tests. Visualization leveraged ggplot2 and ggpubr for foundational graphics, with ”geom_ribbon()” rendering prediction bands and ”theme_pubr()” ensuring publication-ready styling. The scales package formatted axes via ”sec_axis()” for dual-scale plots and ”comma()” for numeric labels. Multi-panel figures were assembled using patchwork . Principal Component Analysis via FactoMineR ”PCA()” visualized ecological gradients across steppe types via factoextra ”fviz_pca_biplot()”, while variable loadings ”(fviz_pca_var())” and correlation heatmaps ( ggcorrplot ) diagnosed multicollinearity. Correlation matrices were visualized using the corrplot package to elucidate relationships between vegetation productivity and environmental drivers across steppe types. Reproducibility was ensured by setting ”set.seed(1234)” for stochastic processes and exporting high-resolution figures as PNG files via ”ggsave”. Dynamic annotations embedded model equations and statistical metrics using ”annotate()”. Results ANPP sensitivity to precipitation and temperature First, we analyzed the relationship between Aboveground net primary productivity (ANPP) and key climatic factors (precipitation and temperature) across different steppe types. Across all steppe types, ANPP followed a logarithmic relationship with growing season precipitation (GSP) (Figure 1a, R² = 0.31, P < 0.001), while it showed a significant linear decrease with growing season temperature (GST) (Figure 1c, R² = 0.09, P < 0.001). Figure 1. Differential responses of steppe ANPP to climatic drivers in Northern China. Relationships between ANPP and (a) growing season precipitation (GSP) for total steppes, (b) GSP partitioned by steppe types, (c) growing season temperature (GST) for total steppes, and (d) GST partitioned by types. Shaded areas, 95% CIs. When steppes were classified into meadow steppe, typical steppe, and desert steppe, the relationship between ANPP and GSP showed a significant linear relationship in meadow steppe (Figure 1b, R² = 0.54, P < 0.001). In contrast, the relationships in typical steppe and desert steppe were better described by the Michaelis-Menten model (R² = 0.25 and R² = 0.23 for typical and desert steppe, respectively). When steppe types were classified, no relationship was observed between ANPP and GST in typical steppe (Figure 1d). Nonlinear PUE Dynamics Along Aridity Gradients We compared the de Martonne aridity index (I dM ) and precipitation use efficiency (PUE) across steppe types. Across all steppe types, PUE showed a significant inverse proportional decline with increasing I dM (Figure 2a, R² = 0.51, P < 0.001). When steppe types were analyzed separately, meadow steppe exhibited a slight upward trend in PUE with rising I dM (R² = 0.04, P < 0.01), while in both typical steppe and desert steppe PUE significantly decreased with increasing I dM ( P < 0.001). Notably, desert steppe demonstrated a rapid inverse proportional decline (R² = 0.84), contrasting with the less pronounced but still significant trend in typical steppe (R² = 0.37, Figure 2b). Figure 2. Divergent nonlinear responses of PUE to aridity gradient in meadow, typical, and desert steppes, and across all the three vegetation types. Relationships between PUE and I dM in (a) total steppes and (b) type-specific steppes with distinct fitting models. Shaded areas, 95% CIs; PUE, precipitation-use efficiency; I dM , De Martonne aridity index. Legacy effects of extreme drought and wetting events We first identified the years of extreme drought and excessive precipitation for the three steppe types (i.e., meadow, typical, and desert steppes) based on local climate variations, data completeness, and aridity index (I dM ) (Figure S1 and Table S3). For meadow steppe, extreme drought occurred in 2007 and excessive precipitation in 2013. Typical steppe experienced extreme drought in 2005 and excessive precipitation in 2008, while desert steppe was subjected to extreme drought in 2017 and excessive precipitation in 2020 (Figure 3a-c). Figure 3. ANPP responses to GSP in meadow (a), typical (b), and desert steppes (c) during extreme dry (orange bars)/wet (blue bars) years in Northern China. Different lowercase letters and uppercase letters indicate significant differences in ANPP among years under extreme dry/wet years, respectively (two-sided Tukey’s HSD tests). Error bars, ±1 SE; Blue line, GSP interannual fluctuation (right y-axis). ANPP, above-ground net primary production; GSP, growing season precipitation. By comparing ANPP between these steppe types, we observed distinct patterns: In the meadow steppe (Figure 3a), both extreme drought and excessive precipitation significantly affected ANPP ( P < 0.05). Drought reduced ANPP by 47.15% compared to the previous year, and ANPP did not recover even after precipitation returned to normal levels, indicating a pronounced negative legacy effect. Excessive precipitation increased ANPP by 41.21% in the excessive precipitation year, but subsequent normalization of precipitation led to a decline in ANPP. In the typical steppe (Figure 3b), extreme drought did not significantly reduce ANPP. Post-drought precipitation recovery reduced ANPP by 79.42% relative to pre-drought year. The 2008 excessive precipitation event failed to boost ANPP; and after precipitation recovered normally, ANPP dropped by 34.63% compared to the excessive precipitation year. In the desert steppe (Figure 3c), extreme drought sharply reduced ANPP by 54.87% relative to the preceding year, meanwhile ANPP completely recovered to baseline levels when precipitation increased largely. In contrast, excessive precipitation in 2020 decreased ANPP by 36.24%, and no recovery occurred even after precipitation returned to normal level. Legacy effect magnitudes scale with precipitation variability We compared the legacy effects of three steppe types and their interannual variations in GSP, revealing differential responses in legacy effects of ANPP across steppe types under varying GSP fluctuations (Figure 4). In meadow steppe, higher GSP variance corresponded to lower legacy effect values, though this trend was not pronounced. Conversely, typical steppe exhibited an opposite pattern: increasing positive GSP variance enhanced legacy effects, suggesting amplified legacies when current-year GSP exceeded those of preceding years. Desert steppe demonstrated distinct bidirectional responses: negative GSP variance values showed intensified legacy effects with decreasing variance, while positive values exhibited enhanced legacies with increasing variance. This implies that regardless of directional change, the legacy effect magnitude in desert steppe escalates proportionally with the absolute value of GSP variance Figure 4. ANPP legacy values in response to interannual GSP variance in meadow, typical, and desert steppes, respectively. Black-bordered points, legacy<0; Error bars, ±1 SE; Shaded areas, 95% CI. ANPP, above-ground net primary production; GSP, growing season precipitation. Legacy effects via climate-mediated plant functional group shifts We integrated multi-year observational data from meadow steppe, typical steppe, and desert steppe to exhibit legacy effects. Through correlation analysis of multiple factors and principal component analysis (PCA) dimensionality reduction (Figure 5, Figures S2, Table S4), we identified core drivers based on climatic variables, PFGs, diversity indices, and ANPP. Distinct patterns were observed in the three vegetation types and across all vegetation types. For meadow steppe, the first two PCs explained 54.4% of total variance (Figure 5a). ANPP was positively correlated with GSP, AP and PUE, while negatively with GST. However, no tight associations of the biodiversity indices with these climatic variables were observed, whereas they negatively closely related with grass ratio indices. In the typical steppe, the first two PCs explained 44.6% of the total variance, three variables (GSP, AP, I dM ) contributed primarily to PC1 while PFGs (e.g., forb, grass, and annual ratios) played a major role in determining the PC2 scores. GST was closely associated with plant species diversity indices, whereas both GST and diversity indices were nearly orthogonal to ANPP in the PCA biplot, indicating a lack of strong relationships with productivity (Figure 5b). In the desert steppe, the first two PCs explained 54.3% of the total variance (PC1: 30.9%, PC2: 23.4%; Figure 5c). ANPP showed a positive association with the forb ratio but a negative association with the grass ratio, indicating opposite contributions of these two contrasting PFGs to vegetation production. Finally, across the three vegetation types, climatic variables contributed mainly to PC1, while PFGs contributed primarily to PC2. The loading points for the desert steppe were predominantly situated in the second quadrant, those for the meadow steppe mainly in the fourth quadrant, and the typical steppe clustered around the origin (Figure 5d). This pattern highlights that the underlying mechanisms driven by climatic variables strongly depend on PFGs and vegetation types. Figure 5. PCA of biotic and abiotic factors in meadow (a), typical (b), desert (c) steppes, and across all steppe types (d). The length of the arrows in each subgraph represents the contribution of the factor to the principal component (cos² value), and the degree of color red and blue reflects the explanatory strength (the higher the R² value, the redder the color). S , Patrick species richness index; H’ , Shannon-Wiener diversity index; and J’ , Pielou’s evenness index; PUE, precipitation use efficiency AP, total annual precipitation; GSP growing season precipitation; GST, growing season temperature; I dM , de Martonne aridity index. Grass, forb, annual, and C 3 represent their respective biomass ratios within the total plant community. A piecewise structural equation modeling framework was developed to decipher the synergistic regulatory mechanisms governing legacy effect values in steppe ecosystems. Five linear mixed-effects models (LMEs) characterized complex variable relationships (Figure 6, Table S6), incorporating vegetation type as random effects (with additional year random effects for legacy effects) to account for hierarchical data structures. The model identified ANPP as the primary driver of ecosystem legacy effects, exhibiting the dominant positive influence (standardized path coefficient = 1.08, P < 0.001). GSP significantly suppressed legacy effects (-0.54, P < 0.001), meanwhile GST had a significant yet weak positive direct effect on legacy effects (0.13, P < 0.001). Additionally, grass ratio showed a weak positive effect on legacy effects (0.07, P < 0.05), while diversity index had a weak negative effect (-0.07, P < 0.05). Figure 6. ANPP legacy effects driven by climate-mediated plant functional groups. *** P <0.001, ** P <0.01, * P <0.05, P<0.1. GSP: growing season precipitation; GST: growing season temperature; Grass ratio: grass biomass proportion; C 3 ratio: C 3 plant biomass proportion; Annual ratio: annual plant biomass proportion; Diversity index: the first principal component of Patrick’s species richness index ( S ), Shannon-Wiener diversity index ( H’ ), and Pielou’s evenness index ( J’ ). Fisher’s C = 8.54, χ² = 5.92, P = 0.314. ANPP was regulated by GSP (0.52, P < 0.001) and GST (-0.10, P < 0.05). PFGs exhibited contrasting effects on ANPP: grass ratio significantly suppressed ANPP (-0.16, P < 0.001), whereas annual ratio enhanced it (0.13, P < 0.001). C₃ plant proportion and diversity index also negatively affected ANPP (-0.08, P < 0.05; -0.13, P < 0.001, respectively). GST indirectly influenced ANPP through multiple pathways: by promoting annual ratio (0.42, P < 0.001) and suppressing grass ratio (-0.27, P < 0.001) and diversity index (-0.11, P < 0.05). C₃ plant ratio showed significant positive correlation with grass ratio (0.27, P < 0.001) but negative correlation with annual ratio (-0.23, P < 0.001) and diversity index (-0.11, P < 0.01). Grass ratio significantly suppressed diversity index (-0.26, P < 0.001), while annual ratio weakly promoted it (0.09, P < 0.05). GSP had a greater indirect effect on legacy via ANPP, whereas GST influenced legacy through multiple indirect pathways—starting from the annual ratio to ANPP, from diversity to ANPP, and from grass ratio to diversity to ANPP—ultimately affecting legacy. The model explained 65% (marginal) and 75% (conditional) of the variation in legacy effects, with marginal R² values ranging from 0.08 to 0.24 for vegetation components. All standardized path coefficients and significance levels are presented in TableS6. Discussion Climatic sensitivity divergence among steppe types Climate factors, including precipitation and temperature, play crucial roles in determining the plant community structure, function, and productivity of grasslands (Tilman, Reich & Knops, 2006; Loreau & De Mazancourt, 2013; Sun et al., 2020; Wang et al., 2024). Theoretically, water availability becomes an increasingly limiting factor for ANPP along the gradient from humid to arid conditions (Huxman et al., 2004; Sala et al., 2012). Our results demonstrate the distinct ANPP responses to precipitation and temperature across the three steppe types (Figure 2). The precipitation-ANPP relationship in meadow steppe aligns with the pattern reported across various steppes by Guo et al. (2012), showing the steepest linear slope that decreases with increasing aridity (Ruppert et al., 2012). However, Wu et al. (2014) observed an exponential increase in ANPP with GSP in alpine meadow steppe, potentially due to the positive linear correlation between precipitation and temperature on the plateau. Higher precipitation alleviates low-temperature stress, allowing combined thermal and hydrological effects to drive asymmetric positive growth. Our findings in typical and desert steppes match those of Wu et al. (2018), where productivity shifted from precipitation limitation to soil nutrient constraints (e.g., nitrogen) under experimental irrigation (Ma et al., 2020; Sala et al., 2012). Our study indicates that under arid conditions, water is the primary limiting factor for ANPP. Plants adapt to low moisture through strategies like optimized root distribution and enhanced water-use efficiency (Lin et al., 2023). Minor precipitation increases can significantly boost productivity in such environments. However, when precipitation exceeds a threshold, other factors like soil nutrients, light, or temperature become limiting factors (Liu et al., 2018; Wang et al., 2025), leading to an asymmetric negative growth pattern (Knapp, Ciais & Smith, 2017). Temperature is another key climatic driver of ANPP. In our study, ANPP showed negative correlations with temperature in all steppe types except typical steppe. This may result from temperature-induced reductions in soil nutrient availability (Wu et al., 2023) or high temperature-induced drought (Li et al., 2020). Distinct geographical and climatic conditions could shift PFG compositions in different steppe types (Yang et al., 2011; Zhang et al., 2022). Climate-mediated PFGs exhibit varying precipitation sensitivities, leading to divergent ANPP responses (Luo et al., 2023). During drought in meadow steppe, grass biomass proportion increased significantly (Figure S3), indicating superior drought tolerance compared to forbs. This aligns with findings from Liu et al. (2018) on the Tibetan Plateau, where soil moisture declines elevated grass dominance. Thus, PFG shifts inevitably alter ANPP responses to climatic change (Sun et al., 2022). In meadow steppe, the previously dominant Carex pediformis declined from 23.12% to 10.13% during extreme drought and became undetectable by the third year. Conversely, drought-tolerant Leymus chinensis reached peak dominance (32.65%). The forb, Artemisia tanacetifolia , increased by 23.38% just one year post-drought. These compositional shifts drive ANPP fluctuations and low drought resistance (Resistance = 38.53%; Figure S4, Table S5). This persistent shift in species composition is the primary reason preventing meadow steppe ecosystems from fully recovering functionality even after precipitation normalization (Hoover et al., 2014; Yu et al., 2022). For typical steppes, species composition data are relatively limited. However, they exhibit stronger resistance to extreme drought in terms of total ANPP (Resistance = 79.55%, Table S5). Although extreme drought did not significantly suppress ANPP in the same year, it likely induced a systemic state shift. This shift resulted in reduced productivity during subsequent years when precipitation returned to normal levels, indicating a potential delayed deleterious effect of drought or its complex impact on ecosystem resilience in typical steppes (Yu et al., 2022). In desert steppe, annual Salsola collina previously absent emerged during (Figure S4, increasing by 34.30% ANPP relative to current year) and after extreme drought (increasing 20.98% ANPP). Soil desiccation and resource redistribution may reshape competitive relationships, enabling this C 4 species to exploit high photosynthetic efficiency under arid conditions (Schwinning & Sala, 2004). Desert steppe exhibited the highest recovery capacity (Recovery = 194.18%) and resilience (Resilience = 87.64%, Figure 4c; Table S5) following an extreme drought, reflecting long-term adaptation to aridity. Specialized PFGs may result in a high water-use efficiency (WUE) and thus high drought-resistance (Schwinning & Sala, 2004). Annuals like Salsola collina utilizing C 4 photosynthesis can maintain basic physiological function during drought (Taylor et al., 2014), while their short life cycles may enable rapid community reassembly post-disturbance. This agrees with the report by Wang et al. (2024), who indicated high desert steppe resilience to rapid vegetation turnover. Notably, among the three steppe types, only desert steppe exhibited sustained ANPP suppression both during and after extreme wet events. For example, the 2020 wet extreme induced a 36.24% ANPP decline with no subsequent recovery (Figure 3c). This contrasts sharply with the transient ANPP increase observed in meadow steppes and the lack of response in typical steppes. Such sustained suppression may stem from altered plant phenology disrupting the balance between vegetative and reproductive growth. Long-term irrigation experiments (Wang et al., 2023b) support this, showing that sustained wetting shortens growth periods and reduces biomass accumulation. Furthermore, soils characteristic of desert steppes has limited nutrient retention capacity. Excessive precipitation can leach bioavailable nitrogen beyond levels plants can absorb, creating a negative feedback loop where water surplus indirectly constrains productivity. Consequently, wet extremes can lead to a legacy effect of delayed recovery. PFG-mediated PUE dynamics across aridity gradients Precipitation use efficiency (PUE) exhibited distinct steppe-type-dependent responses to the aridity index (I dM ), reflecting divergent water-use strategies driven by plant functional group (PFG) composition and soil nutrient limitations (Ruppert et al., 2012; Bai et al., 2008). Meadow steppe displayed a positive logarithmic relationship between PUE and I dM (y = 0.09ln(x) + 0.23; R² = 0.41, P < 0.05), indicating enhanced water-use efficiency under moderate aridity (Ruppert et al., 2012). In contrast, typical steppe shows a significant negative logarithmic response (y = -0.30ln(x) + 1.53; R² = 0.73, P < 0.001), indicating decline in water use efficiency with increasing moisture. Most notably, desert steppe adheres to a hyperbolic decay model (y = 3.64/x; R² = 0.89, P < 0.001), with PUE peaking under the lowest I dM conditions. This inverse PUE-aridity relationship aligns with findings on alpine steppes in Northern Tibet (Zhang, Du & Zhu, 2020), where an increase in annual precipitation reduces both net primary productivity (NPP) and PUE. Our study demonstrates steppe-type-specific PUE responses to local I dM , with generally higher PUE under lower I dM conditions due to PFG differentiation among steppe types. A rainfall manipulation experiment initiated in 2017 similarly documents reduced PUE under increased rainfall regime (Hai et al., 2022). This phenomenon may arise from drought-induced physiological adaptations: suppressed stomatal conductance and transpiration reduce water loss while photosynthetic rates remain relatively stable, enhancing carbon accumulation efficiency per unit water. Desert steppe consistently achieves the highest PUE in arid environments, corroborating meteorological inversion studies in Inner Mongolia (Hua, Ma & Siqin, 2021) that identified strong negative PUE-precipitation correlations. This likely reflects water-nutrient decoupling: low-organic-matter soils in desert steppe have limited nutrient retention capacity. When precipitation exceeds plant-absorption thresholds, water leaching reduces available nitrogen, establishing a negative feedback loop where moisture surpluses constrain productivity. Thus, PUE exhibited significant negative correlations with annual precipitation. Legacy effects from interannual precipitation variability Our study reveals significant differences in ANPP legacy effects (Legacy ANPP ) in response to interannual GSP variance across steppe types (Figure 5), reflecting ecosystem-specific modulation of vegetation adaptation strategies. In meadow steppe, Legacy ANPP decreased with increasing GSP variance (albeit weakly), likely due to resource-conservative strategies. This ecosystem exhibited the lowest resilience (Dry resilience = 38.53%), limiting positive legacy generation during wetting periods. Typical steppe showed the opposite pattern, with Legacy ANPP increasing with GSP variance. This suggests the enhancement of the sensitivity to wetting events: improved soil moisture stimulates growth and biomass investment, leading to positive effects into subsequent years (Sun et al., 2022). Community structure dependence on current-year precipitation further enables carryover effects (Renne et al., 2019). Desert steppe displayed a unique nonlinear response, with Legacy ANPP correlating positively with the absolute value of GSP variance. This reflects adaptive strategies to extreme precipitation variability: plants could conserve resources through dormancy during drying, while rapidly exploiting ephemeral moisture for growth and reproduction during wetting (Schwinning & Sala, 2004). This plasticity underlies strong legacies regardless of the precipitation change direction. Integrated mechanisms of legacy effects across steppe types Structural equation modelling (SEM) revealed that ANPP exerted the strongest positive effect on legacy effect values. This finding might support the positive feedback theory between productivity and stability (Craven et al., 2018). Although GSP directly promoted ANPP, it significantly constrained legacy effects. This apparent paradox may arise from the dual role of drought events: a reduced GSP enhances positive legacy effects by triggering compensatory mechanisms in plant community (Sun et al., 2022), while extreme precipitation increases may disrupt community stability (Knapp et al., 2008). Growing-season temperature (GST) indirectly enhanced legacy effects by directly increasing annual plant proportion. This aligns with the strategy of short-lived plants accelerating resource utilization under warming conditions: annuals can exploit resource vacancies through rapid life cycles, accelerating community reassembly and enhancing system resilience to generate positive legacy effects (Gherardi & Sala, 2015). Notably, ANPP is antagonistically regulated by GSP positively and GST negatively, respectively, with a suppression mediated by plant diversity (Figure 7; Wu et al., 2023). In our study, an increase in graminoid proportion hindered ANPP, while higher annual plant proportion contributed positively to ANPP. This opposition reflects divergence in resource-use strategies: the slow resource turnover of graminoid-dominated communities may constrain productivity elasticity, whereas only fast-growing annuals can easily fill the resource gaps (Gherardi & Sala, 2015; Karitter et al., 2025). Moreover, according to plant economics spectrum, annuals exhibit overyielding due to higher photosynthetic rates and nitrogen-use efficiency under resource abundance (Sun et al., 2023; Garnier et al., 2025; Reich, 2014). GST significantly reduced graminoid proportion (Figure 7), representing a climate-mediated filtering process on plant functional trait: it weakens niches of stable species while strengthening resource capture by opportunistic taxa (Richardson et al., 2025). Although this reorganization may reduce structural complexity, it maintains productivity resilience in arid climates through accelerated resource cycling efficiency. This explains why Craven et al. (2018) emphasized the prominence of productivity–stability relationships in drylands—the synergy between rapid resource turnover strategies and climate variability makes ANPP a core buffer against environmental disturbances (Wang et al., 2024; Lisner et al., 2024; Mahaut et al., 2023). References Bai, Y., & Cotrufo, M. F. (2022). Grassland soil carbon sequestration: Current understanding, challenges, and solutions. Science , 377 (6606), 603-608. Craine, J. M., Ocheltree, T. W., Nippert, J. B., Towne, E. G., Skibbe, A. M., Kembel, S. W., & Fargione, J. E. (2013). Global diversity of drought tolerance and grassland climate-change resilience. Nature Climate Change , 3 (1), 63-67. Harpole, W. S., & Tilman, D. (2007). Grassland species loss resulting from reduced niche dimension. Nature , 446 (7137), 791-793. Tilman, D., Reich, P. B., & Knops, J. M. (2006). Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature , 441 (7093), 629-632. Xu, C., McDowell, N. G., Fisher, R. A., Wei, L., Sevanto, S., Christoffersen, B. O., … & Middleton, R. S. (2019). Increasing impacts of extreme droughts on vegetation productivity under climate change. Nature Climate Change , 9 (12), 948-953. Yu, Q., Xu, C., Wu, H., Ke, Y., Zuo, X., Luo, W., … & Han, X. (2025). Contrasting drought sensitivity of Eurasian and North American grasslands. Nature , 639 (8053), 114-118. Knapp, A. K., Ciais, P., & Smith, M. D. (2017). Reconciling inconsistencies in precipitation–productivity relationships: implications for climate change. New Phytologist , 214 (1), 41-47. Huxman, T. E., Smith, M. D., Fay, P. A., Knapp, A. K., Shaw, M. R., Loik, M. E., … & Williams, D. G. (2004). Convergence across biomes to a common rain-use efficiency. Nature , 429 (6992), 651-654. Knapp, A. K., & Smith, M. D. (2001). Variation among biomes in temporal dynamics of aboveground primary production. Science , 291 (5503), 481-484. Sala, O. E., Parton, W. J., Joyce, L. A., & Lauenroth, W. K. (1988). Primary production of the central grassland region of the United States. Ecology , 69 (1), 40-45. 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 (47), eadd6249. Loreau, M., & De Mazancourt, C. (2013). Biodiversity and ecosystem stability: a synthesis of underlying mechanisms. Ecology Letters , 16 , 106-115. Wang, X. Y., Xu, Y. X., Li, C. H., Yu, H. L., & Huang, J. Y. (2023a). Changes of plant biomass, species diversity, and their influencing factors in a desert steppe of northwestern China under long-term changing precipitation. Chinese Journal of Plant Ecology , 47 (4), 479. Ma, Q., Liu, X., Li, Y., Li, L., Yu, H., Qi, M., … & Xu, Z. (2020). Nitrogen deposition magnifies the sensitivity of desert steppe plant communities to large changes in precipitation. Journal of Ecology , 108 (2), 598-610. Xu, Z., Hou, Y., Zhang, L., Liu, T., & Zhou, G. (2016). Ecosystem responses to warming and watering in typical and desert steppes. Scientific Reports , 6 (1), 34801. Liu, P., Chi, Y., Huang, Z., Zhong, D., & Zhou, L. (2024). Multidimensional response of China’s grassland stability to drought. Global Ecology and Conservation , 52 , e02961. Mesele, S. A., Mechri, M., Okon, M. A., Isimikalu, T. O., Wassif, O. M., Asamoah, E., … & Khurshid, C. (2025). Current problems leading to soil degradation in africa: Raising awareness and finding potential solutions. European Journal of Soil Science , 76 (1), e70069. Furtak, K., & Wolińska, A. (2023). The impact of extreme weather events as a consequence of climate change on the soil moisture and on the quality of the soil environment and agriculture–A review. Catena , 231 , 107378. Mahecha, M. D., Bastos, A., Bohn, F. J., Eisenhauer, N., Feilhauer, H., Hickler, T., … & Quaas, J. (2024). Biodiversity and climate extremes: Known interactions and research gaps. Earth’s Future , 12 (6), e2023EF003963. Canarini, A., Schmidt, H., Fuchslueger, L., Martin, V., Herbold, C. W., Zezula, D., … & Richter, A. (2021). Ecological memory of recurrent drought modifies soil processes via changes in soil microbial community. Nature Communications , 12 (1), 5308. Yu, H., Ma, Q., Liu, X., Li, Y., Li, L., Qi, M., … & Zhang, F. (2021). Resistance, recovery, and resilience of desert steppe to precipitation alterations with nitrogen deposition. Journal of Cleaner Production , 317 , 128434. Hoover, D. L., Knapp, A. K., & Smith, M. D. (2014). Resistance and resilience of a grassland ecosystem to climate extremes. Ecology , 95 (9), 2646-2656. Sala, O. E., Gherardi, L. A., Reichmann, L., Jobbágy, E., & Peters, D. (2012). Legacies of precipitation fluctuations on primary production: theory and data synthesis. Philosophical Transactions of the Royal Society B: Biological Sciences , 367 (1606), 3135-3144. Hoover, D. L., Pfennigwerth, A. A., & Duniway, M. C. (2021). Drought resistance and resilience: The role of soil moisture–plant interactions and legacies in a dryland ecosystem. Journal of Ecology , 109 (9), 3280-3294. Su, J., Zhao, Y., Xu, F., & Bai, Y. (2022). Multiple global changes drive grassland productivity and stability: A meta‐analysis. Journal of Ecology , 110 (12), 2850-2869. Xu, Z., Ren, H., Li, M. H., van Ruijven, J., Han, X., Wan, S., … & Jiang, L. (2015). Environmental changes drive the temporal stability of semi‐arid natural grasslands through altering species asynchrony. Journal of Ecology , 103 (5), 1308-1316. Du, Q., Guan, Q., Sun, Y., Wang, Q., Zhang, J., Xiao, X., … & Zhang, E. (2024). Legacy effects of extreme drought and wetness events on mountain grassland ecosystems and their elevation dependence. Journal of Hydrology , 630 , 130757. Reichmann, L. G., Sala, O. E., & Peters, D. P. (2013). Precipitation legacies in desert grassland primary production occur through previous‐year tiller density. Ecology , 94 (2), 435-443. Sun, J., Zhang, B., Pan, Q., Liu, W., Wang, X., Huang, J., … & Han, X. (2023). Non‐linear response of productivity to precipitation extremes in the Inner Mongolia grassland. Functional Ecology , 37 (6), 1663-1673. Sun, J., Zhou, T. C., Liu, M., Chen, Y. C., Liu, G. H., Xu, M., … & Li, Y. N. (2020). Water and heat availability are drivers of the aboveground plant carbon accumulation rate in alpine grasslands on the Tibetan Plateau. Global Ecology and Biogeography , 29 (1), 50-64. Wang, S., Hong, P., Adler, P. B., Allan, E., Hautier, Y., Schmid, B., … & Feng, Y. (2024). Towards mechanistic integration of the causes and consequences of biodiversity. Trends in Ecology & Evolution , 39 (7), 689-700. Guo, Q., Hu, Z., Li, S., Li, X., Sun, X., & Yu, G. (2012). Spatial variations in aboveground net primary productivity along a climate gradient in Eurasian temperate grassland: effects of mean annual precipitation and its seasonal distribution. Global Change Biology , 18 (12), 3624-3631. Ruppert, J. C., Holm, A., Miehe, S., Muldavin, E., Snyman, H. A., Wesche, K., & Linstädter, A. (2012). Meta‐analysis of ANPP and rain‐use efficiency confirms indicative value for degradation and supports non‐linear response along precipitation gradients in drylands. Journal of Vegetation Science , 23 (6), 1035-1050. Wu, J., Shen, Z., & Zhang, X. (2014). Precipitation and species composition primarily determine the diversity–productivity relationship of alpine grasslands on the Northern Tibetan Plateau. Alpine Botany , 124 (1), 13-25. Wu, D., Ciais, P., Viovy, N., Knapp, A. K., Wilcox, K., Bahn, M., … & Piao, S. (2018). Asymmetric responses of primary productivity to altered precipitation simulated by ecosystem models across three long-term grassland sites. Biogeosciences , 15 (11), 3421-3437. Lin, S., Wang, G., Hu, Z., Sun, X., Song, C., Huang, K., … & Yang, Y. (2023). Contrasting response of growing season water use efficiency to precipitation changes between alpine meadows and alpine steppes over the Tibetan Plateau. Agricultural Water Management , 289 , 108571. Liu, H., Mi, Z., Lin, L. I., Wang, Y., Zhang, Z., Zhang, F., … & He, J. S. (2018). Shifting plant species composition in response to climate change stabilizes grassland primary production. Proceedings of the National Academy of Sciences , 115 (16), 4051-4056. Wang, Y., Du, Y., Zhao, W., Liu, H., Jiang, J., & He, Z. (2025). Soil drought thresholds of alpine grassland deceased rapidly under the influence of extreme low temperature in northeastern Qinghai-Tibet Plateau. Ecological Processes , 14 (1), 21. Wu, W., Sun, R., Liu, L., Liu, X., Yu, H., Ma, Q., … & Xu, Z. (2023). Precipitation consistently promotes, but temperature inversely drives, biomass production in temperate vs. alpine grasslands. Agricultural and Forest Meteorology , 329 , 109277. Li, Z., Li, Z., Tong, X., Zhang, J., Dong, L., Zheng, Y., … & Li, F. Y. (2020). Climatic humidity mediates the strength of the species richness–biomass relationship on the Mongolian Plateau steppe. Science of the Total Environment , 718 , 137252. Yang, H., Wu, M., Liu, W., Zhang, Z. H. E., Zhang, N., & Wan, S. (2011). Community structure and composition in response to climate change in a temperate steppe. Global Change Biology , 17 (1), 452-465. Zhang, A., Li, X., Zeng, F., Jiang, Y., & Wang, R. (2022). Variation characteristics of different plant functional groups in alpine desert steppe of the Altun Mountains, northern Qinghai-Tibet Plateau. Frontiers in Plant Science , 13 , 961692. Luo, W., Ma, W., Song, L., Te, N., Chen, J., Muraina, T. O., … & Collins, S. L. (2023). Compensatory dynamics drive grassland recovery from drought. Journal of Ecology , 111 (6), 1281-1291. Yu, H., Liu, X., Ma, Q., Li, L., Wu, W., Qi, M., … & Xu, Z. (2022). Nitrogen deposition drives response and recovery in the context of precipitation change and its reversal in an arid ecosystem. Journal of Geophysical Research: Biogeosciences , 127 (9), e2022JG006828. Schwinning, S., & Sala, O. E. (2004). Hierarchy of responses to resource pulses in arid and semi-arid ecosystems. Oecologia , 141 (2), 211-220. Taylor, S. H., Ripley, B. S., Martin, T., De‐Wet, L. A., Woodward, F. I., & Osborne, C. P. (2014). Physiological advantages of C 4 grasses in the field: a comparative experiment demonstrating the importance of drought. Global Change Biology , 20 (6), 1992-2003. Wang, Y., Shen, Y., Xie, Y., Ma, H., Li, W., Luo, X., … & Li, J. (2023b). Changes in precipitation have both direct and indirect effects on typical steppe aboveground net primary productivity in Loess Plateau, China. Plant and Soil , 484 (1), 503-515. Bai, Y., Wu, J., Xing, Q., Pan, Q., Huang, J., Yang, D., & Han, X. (2008). Primary production and rain use efficiency across a precipitation gradient on the Mongolia plateau. Ecology , 89 (8), 2140-2153. Zhang, X., Du, X., & Zhu, Z. (2020). Effects of precipitation and temperature on precipitation use efficiency of alpine grassland in Northern Tibet, China. Scientific Reports , 10 (1), 20309. Hai, X., Li, J., Li, J., Liu, Y., Dong, L., Wang, X., … & Deng, L. (2022). Variations in plant water use efficiency response to manipulated precipitation in a temperate grassland. Frontiers in Plant Science , 13 , 881282. Hua, Y., Ma, X., & Siqin, B. (2021). Spatial and temporal characteristics of precipitation utilization efficiency of desert steppe vegetation in Inner Mongolia. Journal of Desert Research , 41 (4), 51. Renne, R. R., Bradford, J. B., Burke, I. C., & Lauenroth, W. K. (2019). Soil texture and precipitation seasonality influence plant community structure in North American temperate shrub steppe. Ecology , 100 (11), e02824. Craven, D., Eisenhauer, N., Pearse, W. D., Hautier, Y., Isbell, F., Roscher, C., … & Manning, P. (2018). Multiple facets of biodiversity drive the diversity–stability relationship. Nature Ecology & Evolution , 2 (10), 1579-1587. Knapp, A. K., Beier, C., Briske, D. D., Classen, A. T., Luo, Y., Reichstein, M., … & Weng, E. (2008). Consequences of more extreme precipitation regimes for terrestrial ecosystems. Bioscience , 58 (9), 811-821. Gherardi, L. A., & Sala, O. E. (2015). Enhanced interannual precipitation variability increases plant functional diversity that in turn ameliorates negative impact on productivity. Ecology Letters , 18 (12), 1293-1300. Karitter, P., Corvers, E., Karrenbauer, M., March-Salas, M., Stojanova, B., Ensslin, A., … & Scheepens, J. F. (2025). Evolution of competitive ability and the response to nutrient availability: a resurrection study with the calcareous grassland herb, Leontodon hispidus. Oecologia , 207 (1), 17. Garnier, E., Vile, D., Debain, S., Bottin, L., Laurent, G., & Roumet, C. (2025). Photosynthesis, water‐use and nitrogen relate to both plant height and leaf structure in 60 species from the Mediterranean. Functional Ecology , 39 (2), 567-582. Reich, P. B. (2014). The world‐wide ‘fast–slow’plant economics spectrum: a traits manifesto. Journal of Ecology , 102 (2), 275-301. Richardson, J. A., Engelstädter, J., & Letten, A. D. (2025). A unifying principle for multispecies coexistence under resource fluctuations. Proceedings of the National Academy of Sciences , 122 (24), e2424996122. Lisner, A., Segrestin, J., Konečná, M., Blažek, P., Janíková, E., Applová, M., … & Lepš, J. (2024). Why are plant communities stable? Disentangling the role of dominance, asynchrony and averaging effect following realistic species loss scenario. Journal of Ecology , 112 (8), 1832-1841. Mahaut, L., Choler, P., Denelle, P., Garnier, E., Thuiller, W., Kattge, J., … & Violle, C. (2023). Trade‐offs and synergies between ecosystem productivity and stability in temperate grasslands. Global Ecology and Biogeography , 32 (4), 561-572. Zhao, Y., Lu, X., Wang, Y., & Bai, Y. (2022). How precipitation legacies affect broad-scale patterns of primary productivity: Evidence from the Inner Mongolia grassland. Agricultural and Forest Meteorology , 320 , 108954. Yan, R., Chen, B., Zhang, B., Yang, G., & Xin, X. (2021). Dynamic dataset of plant community composition in Stipa baicalensis meadow steppe in HulunBuir of China (2009–2015). DOI: 10.11922/sciencedb.j00001.00243. Yang, G., Tang, H., & Xin, X. (2011). China’s Ecosystem Positioning Observation and Research Dataset: Grassland and Grassland Ecosystem Volume (2006-2008). China Agriculture Press. DOI: 10.12199/nesdc.ecodb.mon.2020.dp2011.hlg.003. Han, X., Bai, Y., Pan, Q., & He, N. (2011). Ecosystem positioning observation and research dataset in China: Grassland & Desert Ecosystems Volume: Inner Mongolia Xilin Gol Station: 2005~2008. China Agriculture Press. DOI: 10.12199/nesdc.ecodb.mon.2020.dp2011.nmg.004. Yan, H., Liang, C., Li, Z., Liu, Z., Miao, B., He, C., & Sheng, L. (2015). Impact of precipitation patterns on biomass and species richness of annuals in a dry steppe. PLoS One , 10 (4), e0125300. Sabzchi-Dehkharghani, H., Biswas, A., Meshram, S. G., & Majnooni-Heris, A. (2024). Estimating gross and net primary productivities using earth observation products: A review. Environmental Modeling & Assessment , 29 (1), 179-200. Whittaker, R. H. (1972). Evolution and measurement of species diversity. Taxon , 21 (2-3), 213-251. Hu, H., Liu, X., He, Y., Feng, J., Xu, Y., & Jing, J. (2025). Legacy effects of precipitation change: Theories, dynamics, and applications. Journal of Environmental Management , 373 , 123729. Wilsey, B. J., Chalcraft, D. R., Bowles, C. M., & Willig, M. R. (2005). Relationships among indices suggest that richness is an incomplete surrogate for grassland biodiversity. Ecology , 86 (5), 1178-1184. Tuhkanen, S. (1980). Climatic parameters and indices in plant geography. Sv. växtgeografiska sällsk. Paruelo, J. M., Lauenroth, W. K., Burke, I. C., & Sala, O. E. (1999). Grassland precipitation-use efficiency varies across a resource gradient. Ecosystems , 2 (1), 64-68. Liu, X., Ma, Q., Yu, H., Li, Y., Li, L., Qi, M., … & Xu, Z. (2021). Climate warming-induced drought constrains vegetation productivity by weakening the temporal stability of the plant community in an arid grassland ecosystem. Agricultural and Forest Meteorology , 307 , 108526. R Core Team, R. (2020). R: A language and environment for statistical computing. Acknowledgments The authors are grateful to Quanhui Ma, Hongying Yu, Wenjuan Wu, Xiaodi Liu, Yibo Li, and Lang Li for their loyal help during the study. Funding: The Strategic Priority Research Program of Chinese Academy of Sciences (XDA26010103) Competing interests: Authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Data and R codes are available from https://doi.org/10.6084/m9.figshare.29597078. Other data from Han et al. (2011), Sun et al. (2022), Sun et al. (2023), Yan et al. (2015), Yan et al. (2021), Yang et al. (2011) and Yu et al. (2021) are openly available at https://doi.org/10.12199/nesdc.ecodb.mon.2020.dp2011.nmg.004, https://doi.org/10.1126/sciadv.add6249, https://doi.org/10.1111/1365-2435.14328, https://doi.org/10.1371/journal.pone.0125300, https://doi.org/10.11922/sciencedb.j00001.00243, https://doi.org/10.12199/nesdc.ecodb.mon.2020.dp2011.hlg.003, and https://doi.org/10.1016/j.jclepro.2021.128434, respectively. Supplementary Materials Figs. S1 to S4 Tables S1 to S6 References (1 to 7) Information & Authors Information Version history V1 Version 1 23 September 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords aboveground net primary productivity climatic drivers divergent responses across steppe types extreme precipitation events legacy effects plant functional groups Authors Affiliations Ruojun Sun 0000-0002-9597-438X Chinese Academy of Sciences Institute of Botany View all articles by this author Kuo Sun Chinese Academy of Sciences Institute of Botany View all articles by this author Leren Liu Chinese Academy of Sciences Institute of Botany View all articles by this author Feng Zhang Institute of Botany Chinese Academy of Sciences View all articles by this author Yanhong Lou Shandong Agricultural University View all articles by this author Yuping Zhuge Shandong Agricultural University View all articles by this author Zhenzhu Xu 0000-0001-5246-8981 [email protected] Chinese Academy of Sciences Institute of Botany View all articles by this author Metrics & Citations Metrics Article Usage 290 views 120 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Ruojun Sun, Kuo Sun, Leren Liu, et al. Legacies of Precipitation Extremes: Divergent Responses Across Steppe Types. Authorea . 23 September 2025. DOI: https://doi.org/10.22541/au.175865459.93061204/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175865459.93061204/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fef45051ce0df88',t:'MTc3OTMyMTI5OA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();