Extreme drought impacts have been underestimated in grasslands and shrublands globally.

Climate change is increasing the frequency and severity of short-term (~1 y) drought events-the most common duration of drought-globally. Yet the impact of this intensification of drought on ecosystem functioning remains poorly resolved. This is due in part to the widely disparate approaches ecologists have employed to study drought, variation in the severity and duration of drought studied, and differences among ecosystems in vegetation, edaphic and climatic attributes that can mediate drought impacts. To overcome these problems and better identify the factors that modulate drought responses, we used a coordinated distributed experiment to quantify the impact of short-term drought on grassland and shrubland ecosystems. With a standardized approach, we imposed ~a single year of drought at 100 sites on six continents. Here we show that loss of a foundational ecosystem function-aboveground net primary production (ANPP)-was 60% greater at sites that experienced statistically extreme drought (1-in-100-y event) vs. those sites where drought was nominal (historically more common) in magnitude (35% vs. 21%, respectively). This reduction in a key carbon cycle process with a single year of extreme drought greatly exceeds previously reported losses for grasslands and shrublands. Our global experiment also revealed high variability in drought response but that relative reductions in ANPP were greater in drier ecosystems and those with fewer plant species. Overall, our results demonstrate with unprecedented rigor that the global impacts of projected increases in drought severity have been significantly underestimated and that drier and less diverse sites are likely to be most vulnerable to extreme drought.

Signatures of autumn deluges revealed during spring drought in a semi-arid grassland.

Increases in extremely large precipitation events (deluges) and shifts in seasonal patterns of water availability with climate change will both have important consequences for ecosystem function, particularly in water-limited regions. While previous work in the semi-arid shortgrass steppe of northeastern Colorado has demonstrated this ecosystem's strong sensitivity to growing season deluges, our understanding of ecosystem responses to deluges during the dormant season is limited. Here, we imposed experimental 100 mm deluges (~ 30% of mean annual precipitation) in either September or October in a native C4-dominated shortgrass steppe ecosystem to evaluate the impact of this post-growing season shift in water availability during the autumn and the following growing season. Soil moisture for both deluge treatments remained elevated compared with ambient levels through April as spring precipitation was atypically low. Despite overall low levels of productivity with spring drought, these deluges from the previous autumn increased aboveground net primary production (ANPP), primarily due to increases with C4 grasses. C3 ANPP was also enhanced, largely due to an increase in the annual C3 grass, Vulpia octoflora, in the October deluge treatment. While spring precipitation has historically been the primary determinant of ecosystem function in this ecosystem, this combination of two climate extremes-an extremely wet autumn followed by a naturally-occurring spring drought-revealed the potential for meaningful carryover effects from autumn precipitation. With climate change increasing the likelihood of extremes during all seasons, experiments which create novel climatic conditions can provide new insight into the dynamics of ecosystem functioning in the future.

The detection and attribution of extreme reductions in vegetation growth across the global land surface.

Negative extreme anomalies in vegetation growth (NEGs) usually indicate severely impaired ecosystem services. These NEGs can result from diverse natural and anthropogenic causes, especially climate extremes (CEs). However, the relationship between NEGs and many types of CEs remains largely unknown at regional and global scales. Here, with satellite-derived vegetation index data and supporting tree-ring chronologies, we identify periods of NEGs from 1981 to 2015 across the global land surface. We find 70% of these NEGs are attributable to five types of CEs and their combinations, with compound CEs generally more detrimental than individual ones. More importantly, we find that dominant CEs for NEGs vary by biome and region. Specifically, cold and/or wet extremes dominate NEGs in temperate mountains and high latitudes, whereas soil drought and related compound extremes are primarily responsible for NEGs in wet tropical, arid and semi-arid regions. Key characteristics (e.g., the frequency, intensity and duration of CEs, and the vulnerability of vegetation) that determine the dominance of CEs are also region- and biome-dependent. For example, in the wet tropics, dominant individual CEs have both higher intensity and longer duration than non-dominant ones. However, in the dry tropics and some temperate regions, a longer CE duration is more important than higher intensity. Our work provides the first global accounting of the attribution of NEGs to diverse climatic extremes. Our analysis has important implications for developing climate-specific disaster prevention and mitigation plans among different regions of the globe in a changing climate.

Accounting for herbaceous communities in process-based models will advance our understanding of "grassy" ecosystems.

Grassland and other herbaceous communities cover significant portions of Earth's terrestrial surface and provide many critical services, such as carbon sequestration, wildlife habitat, and food production. Forecasts of global change impacts on these services will require predictive tools, such as process-based dynamic vegetation models. Yet, model representation of herbaceous communities and ecosystems lags substantially behind that of tree communities and forests. The limited representation of herbaceous communities within models arises from two important knowledge gaps: first, our empirical understanding of the principles governing herbaceous vegetation dynamics is either incomplete or does not provide mechanistic information necessary to drive herbaceous community processes with models; second, current model structure and parameterization of grass and other herbaceous plant functional types limits the ability of models to predict outcomes of competition and growth for herbaceous vegetation. In this review, we provide direction for addressing these gaps by: (1) presenting a brief history of how vegetation dynamics have been developed and incorporated into earth system models, (2) reporting on a model simulation activity to evaluate current model capability to represent herbaceous vegetation dynamics and ecosystem function, and (3) detailing several ecological properties and phenomena that should be a focus for both empiricists and modelers to improve representation of herbaceous vegetation in models. Together, empiricists and modelers can improve representation of herbaceous ecosystem processes within models. In so doing, we will greatly enhance our ability to forecast future states of the earth system, which is of high importance given the rapid rate of environmental change on our planet.

Traits that distinguish dominant species across aridity gradients differ from those that respond to soil moisture.

Many plant traits respond to changes in water availability and might be useful for understanding ecosystem properties such as net primary production (NPP). This is especially evident in grasslands where NPP is water-limited and primarily determined by the traits of dominant species. We measured root and shoot morphology, leaf hydraulic traits, and NPP of four dominant North American prairie grasses in response to four levels of soil moisture in a greenhouse experiment. We expected that traits of species from drier regions would be more responsive to reduced water availability and that this would make these species more resistant to low soil moisture than species from wetter regions. All four species grew taller, produced more biomass, and increased total root length in wetter treatments. Each species reduced its leaf turgor loss point (TLP) in drier conditions, but only two species (one xeric, one mesic) maintained leaf water potential above TLP. We identified a suite of traits that clearly distinguished species from one another, but, surprisingly, these traits were relatively unresponsive to reduced soil moisture. Specifically, more xeric species produced thinner roots with higher specific root length and had a lower root mass fraction. This suggest that root traits are critical for distinguishing species from one another but might not respond strongly to changing water availability, though this warrants further investigation in the field. Overall, we found that NPP of these dominant grass species responded similarly to varying levels of soil moisture despite differences in species morphology, physiology, and habitat of origin.

Multiple global change drivers show independent, not interactive effects: a long-term case study in tallgrass prairie.

Ecosystems are faced with an onslaught of co-occurring global change drivers. While frequently studied independently, the effects of multiple global change drivers have the potential to be additive, antagonistic, or synergistic. Global warming, for example, may intensify the effects of more variable precipitation regimes with warmer temperatures increasing evapotranspiration and thereby amplifying the effect of already dry soils. Here, we present the long-term effects (11 years) of altered precipitation patterns (increased intra-annual variability in the growing season) and warming (1 °C year-round) on plant community composition and aboveground net primary productivity (ANPP), a key measure of ecosystem functioning in mesic tallgrass prairie. Based on past results, we expected that increased precipitation variability and warming would have additive effects on both community composition and ANPP. Increased precipitation variability altered plant community composition and increased richness, with no effect on ANPP. In contrast, warming decreased ANPP via reduction in grass stems and biomass but had no effect on the plant community. Contrary to expectations, across all measured variables, precipitation and warming treatments had no interactive effects. While treatment interactions did not occur, each treatment did individually impact a different component of the ecosystem (i.e., community vs. function). Thus, different aspects of the ecosystem may be sensitive to different global change drivers in mesic grassland ecosystems.

Resolving the Dust Bowl paradox of grassland responses to extreme drought.

During the 1930s Dust Bowl drought in the central United States, species with the C3 photosynthetic pathway expanded throughout C4-dominated grasslands. This widespread increase in C3 grasses during a decade of low rainfall and high temperatures is inconsistent with well-known traits of C3 vs. C4 pathways. Indeed, water use efficiency is generally lower, and photosynthesis is more sensitive to high temperatures in C3 than C4 species, consistent with the predominant distribution of C3 grasslands in cooler environments and at higher latitudes globally. We experimentally imposed extreme drought for 4 y in mixed C3/C4 grasslands in Kansas and Wyoming and, similar to Dust Bowl observations, also documented three- to fivefold increases in C3/C4 biomass ratios. To explain these paradoxical responses, we first analyzed long-term climate records to show that under nominal conditions in the central United States, C4 grasses dominate where precipitation and air temperature are strongly related (warmest months are wettest months). In contrast, C3 grasses flourish where precipitation inputs are less strongly coupled to warm temperatures. We then show that during extreme drought years, precipitation-temperature relationships weaken, and the proportion of precipitation falling during cooler months increases. This shift in precipitation seasonality provides a mechanism for C3 grasses to respond positively to multiyear drought, resolving the Dust Bowl paradox. Grasslands are globally important biomes and increasingly vulnerable to direct effects of climate extremes. Our findings highlight how extreme drought can indirectly alter precipitation seasonality and shift ecosystem phenology, affecting function in ways not predictable from key traits of C3 and C4 species.

Shifting seasonal patterns of water availability: ecosystem responses to an unappreciated dimension of climate change.

Seasonal patterns of water availability can differ dramatically among ecosystems, with well-known consequences for ecosystem structure and functioning. Less appreciated is that climate change can shift the seasonality of water availability (e.g. to wetter springs, drier summers), resulting in both subtle and profound ecological impacts. Here we (1) review evidence that the seasonal availability of water is being altered in ecosystems worldwide, (2) explore several mechanisms potentially driving these changes, and (3) highlight the breadth of ecological consequences resulting from shifts in the seasonality of water availability. We conclude that seasonal patterns of water availability are changing globally, but in regionally specific ways requiring more rigorous and nuanced assessments of ecosystem vulnerability as well as the ecological consequences.

Compound hydroclimatic extremes in a semi-arid grassland: Drought, deluge, and the carbon cycle.

Climate change is predicted to increase the frequency and intensity of extreme events including droughts and large precipitation events or "deluges." While many studies have focused on the ecological impacts of individual events (e.g., a heat wave), there is growing recognition that when extreme events co-occur as compound extremes, (e.g., a heatwave during a drought), the additive effects on ecosystems are often greater than either extreme alone. In this study, we assessed a unique type of extreme-a contrasting compound extreme-where the extremes may have offsetting, rather than additive ecological effects, by examining how a deluge during a drought impacts productivity and carbon cycling in a semi-arid grassland. The experiment consisted of four treatments: a control (average precipitation), an extreme drought (<5th percentile), an extreme drought interrupted by a single deluge (>95th percentile), or an extreme drought interrupted by the equivalent amount of precipitation added in several smaller events. We highlight three key results. First, extreme drought resulted in early senescence, reduced carbon uptake, and a decline in net primary productivity relative to the control treatment. Second, the deluge imposed during extreme drought stimulated carbon fluxes and plant growth well above the levels of both the control and the drought treatment with several additional smaller rainfall events, emphasizing the importance of precipitation amount, event size, and timing. Third, while the deluge's positive effects on carbon fluxes and plant growth persisted for 1 month, the deluge did not completely offset the negative effects of extreme drought on end-of-season productivity. Thus, in the case of these contrasting hydroclimatic extremes, a deluge during a drought can stimulate temporally dynamic ecosystem processes (e.g., net ecosystem exchange) while only partially compensating for reductions in ecosystem functions over longer time scales (e.g., aboveground net primary productivity).

Global change effects on plant communities are magnified by time and the number of global change factors imposed.

Global change drivers (GCDs) are expected to alter community structure and consequently, the services that ecosystems provide. Yet, few experimental investigations have examined effects of GCDs on plant community structure across multiple ecosystem types, and those that do exist present conflicting patterns. In an unprecedented global synthesis of over 100 experiments that manipulated factors linked to GCDs, we show that herbaceous plant community responses depend on experimental manipulation length and number of factors manipulated. We found that plant communities are fairly resistant to experimentally manipulated GCDs in the short term (<10 y). In contrast, long-term (≥10 y) experiments show increasing community divergence of treatments from control conditions. Surprisingly, these community responses occurred with similar frequency across the GCD types manipulated in our database. However, community responses were more common when 3 or more GCDs were simultaneously manipulated, suggesting the emergence of additive or synergistic effects of multiple drivers, particularly over long time periods. In half of the cases, GCD manipulations caused a difference in community composition without a corresponding species richness difference, indicating that species reordering or replacement is an important mechanism of community responses to GCDs and should be given greater consideration when examining consequences of GCDs for the biodiversity-ecosystem function relationship. Human activities are currently driving unparalleled global changes worldwide. Our analyses provide the most comprehensive evidence to date that these human activities may have widespread impacts on plant community composition globally, which will increase in frequency over time and be greater in areas where communities face multiple GCDs simultaneously.

Determinants of community compositional change are equally affected by global change.

Global change is impacting plant community composition, but the mechanisms underlying these changes are unclear. Using a dataset of 58 global change experiments, we tested the five fundamental mechanisms of community change: changes in evenness and richness, reordering, species gains and losses. We found 71% of communities were impacted by global change treatments, and 88% of communities that were exposed to two or more global change drivers were impacted. Further, all mechanisms of change were equally likely to be affected by global change treatments-species losses and changes in richness were just as common as species gains and reordering. We also found no evidence of a progression of community changes, for example, reordering and changes in evenness did not precede species gains and losses. We demonstrate that all processes underlying plant community composition changes are equally affected by treatments and often occur simultaneously, necessitating a wholistic approach to quantifying community changes.

How big is big enough? Surprising responses of a semiarid grassland to increasing deluge size.

Climate change has intensified the hydrologic cycle globally, increasing the magnitude and frequency of large precipitation events, or deluges. Dryland ecosystems are expected to be particularly responsive to increases in deluge size, as their ecological processes are largely dependent on distinct soil moisture pulses. To better understand how increasing deluge size will affect ecosystem function, we conducted a field experiment in a native semiarid shortgrass steppe (Colorado, USA). We quantified ecological responses to a range of deluge sizes, from moderate to extreme, with the goal of identifying response patterns and thresholds beyond which ecological processes would not increase further (saturate). Using a replicated regression approach, we imposed single deluges that ranged in size from 20 to 120 mm (82.3rd to >99.9th percentile of historical event size) on undisturbed grassland plots. We quantified pre- and postdeluge responses in soil moisture, soil respiration, and canopy greenness, as well as leaf water potential, growth, and flowering of the dominant grass species (Bouteloua gracilis). We also measured end of season above- and belowground net primary production (ANPP, BNPP). As expected, this water-limited ecosystem responded strongly to the applied deluges, but surprisingly, most variables increased linearly with deluge size. We found little evidence for response thresholds within the range of deluge sizes imposed, at least during this dry year. Instead, response patterns reflected the linear increase in the duration of elevated soil moisture (2-22 days) with increasing event size. Flowering of B. gracilis and soil respiration responded particularly strongly to deluge size (14- and 4-fold increases, respectively), as did ANPP and BNPP (~60% increase for both). Overall, our results suggest that this semiarid grassland will respond positively and linearly to predicted increases in deluge size, and that event sizes may need to exceed historical magnitudes, or occur during wet years, before responses saturate.

Precipitation-productivity relationships and the duration of precipitation anomalies: An underappreciated dimension of climate change.

In terrestrial ecosystems, climate change forecasts of increased frequencies and magnitudes of wet and dry precipitation anomalies are expected to shift precipitation-net primary productivity (PPT-NPP) relationships from linear to nonlinear. Less understood, however, is how future changes in the duration of PPT anomalies will alter PPT-NPP relationships. A review of the literature shows strong potential for the duration of wet and dry PPT anomalies to impact NPP and to interact with the magnitude of anomalies. Within semi-arid and mesic grassland ecosystems, PPT gradient experiments indicate that short-duration (1 year) PPT anomalies are often insufficient to drive nonlinear aboveground NPP responses. But long-term studies, within desert to forest ecosystems, demonstrate how multi-year PPT anomalies may result in increasing impacts on NPP through time, and thus alter PPT-NPP relationships. We present a conceptual model detailing how NPP responses to PPT anomalies may amplify with the duration of an event, how responses may vary in xeric vs. mesic ecosystems, and how these differences are most likely due to demographic mechanisms. Experiments that can unravel the independent and interactive impacts of the magnitude and duration of wet and dry PPT anomalies are needed, with multi-site long-term PPT gradient experiments particularly well-suited for this task.

Seasonal changes in GPP/SIF ratios and their climatic determinants across the Northern Hemisphere.

Satellite-derived sun-induced chlorophyll fluorescence (SIF) has been increasingly used for estimating gross primary production (GPP). However, the relationship between SIF and GPP has not been well defined, impeding the translation of satellite observed SIF to GPP. Previous studies have generally assumed a linear relationship between SIF and GPP at daily and longer time scales, but support for this assumption is lacking. Here, we used the GPP/SIF ratio to investigate seasonal variations in the relationship between SIF and GPP over the Northern Hemisphere (NH). Based on multiple SIF products and MODIS and FLUXCOM GPP data, we found strong seasonal hump-shaped patterns for the GPP/SIF ratio over northern latitudes, with higher values in the summer than in the spring or autumn. This hump-shaped GPP/SIF seasonal variation was confirmed by examining different SIF products and was evident for most vegetation types except evergreen broadleaf forests. The seasonal amplitude of the GPP/SIF ratio decreased from the boreal/arctic region to drylands and the tropics. For most of the NH, the lowest GPP/SIF values occurred in October or September, while the maximum GPP/SIF values were evident in June and July. The most pronounced seasonal amplitude of GPP/SIF occurred in intermediate temperature and precipitation ranges. GPP/SIF was positively related to temperature in the early and late parts of the growing season, but not during the peak growing months. These shifting relationships between temperature and GPP/SIF across different months appeared to play a key role in the seasonal dynamics of GPP/SIF. Several mechanisms may explain the patterns we observed, and future research encompassing a broad range of climate and vegetation settings is needed to improve our understanding of the spatial and temporal relationships between SIF and GPP. Nonetheless, the strong seasonal variation in GPP/SIF we identified highlights the importance of incorporating this behavior into SIF-based GPP estimations.

Moisture availability mediates the relationship between terrestrial gross primary production and solar-induced chlorophyll fluorescence: Insights from global-scale variations.

Effective use of solar-induced chlorophyll fluorescence (SIF) to estimate and monitor gross primary production (GPP) in terrestrial ecosystems requires a comprehensive understanding and quantification of the relationship between SIF and GPP. To date, this understanding is incomplete and somewhat controversial in the literature. Here we derived the GPP/SIF ratio from multiple data sources as a diagnostic metric to explore its global-scale patterns of spatial variation and potential climatic dependence. We found that the growing season GPP/SIF ratio varied substantially across global land surfaces, with the highest ratios consistently found in boreal regions. Spatial variation in GPP/SIF was strongly modulated by climate variables. The most striking pattern was a consistent decrease in GPP/SIF from cold-and-wet climates to hot-and-dry climates. We propose that the reduction in GPP/SIF with decreasing moisture availability may be related to stomatal responses to aridity. Furthermore, we show that GPP/SIF can be empirically modeled from climate variables using a machine learning (random forest) framework, which can improve the modeling of ecosystem production and quantify its uncertainty in global terrestrial biosphere models. Our results point to the need for targeted field and experimental studies to better understand the patterns observed and to improve the modeling of the relationship between SIF and GPP over broad scales.

Is a drought a drought in grasslands? Productivity responses to different types of drought.

Drought, defined as a marked deficiency of precipitation relative to normal, occurs as periods of below-average precipitation or complete failure of precipitation inputs, and can be limited to a single season or prolonged over multiple years. Grasslands are typically quite sensitive to drought, but there can be substantial variability in the magnitude of loss of ecosystem function. We hypothesized that differences in how drought occurs may contribute to this variability. In four native Great Plains grasslands (three C4- and one C3-dominated) spanning a ~ 500-mm precipitation gradient, we imposed drought for four consecutive years by (1) reducing each rainfall event by 66% during the growing season (chronic drought) or (2) completely excluding rainfall during a shorter portion of the growing season (intense drought). The drought treatments were similar in magnitude but differed in the following characteristics: event number, event size and length of dry periods. We observed consistent drought-induced reductions (28-37%) in aboveground net primary production (ANPP) only in the C4-dominated grasslands. In general, intense drought reduced ANPP more than chronic drought, with little evidence that drought duration altered this pattern. Conversely, belowground net primary production (BNPP) was reduced by drought in all grasslands (32-64%), with BNPP reductions greater in intense vs. chronic drought treatments in the most mesic grassland. We conclude that grassland productivity responses to drought did not strongly differ between these two types of drought, but when differences existed, intense drought consistently reduced function more than chronic drought.

Precipitation amount and event size interact to reduce ecosystem functioning during dry years in a mesic grassland.

Ongoing intensification of the hydrological cycle is altering rainfall regimes by increasing the frequency of extreme wet and dry years and the size of individual rainfall events. Despite long-standing recognition of the importance of precipitation amount and variability for most terrestrial ecosystem processes, we lack understanding of their interactive effects on ecosystem functioning. We quantified this interaction in native grassland by experimentally eliminating temporal variability in growing season rainfall over a wide range of precipitation amounts, from extreme wet to dry conditions. We contrasted the rain use efficiency (RUE) of above-ground net primary productivity (ANPP) under conditions of experimentally reduced versus naturally high rainfall variability using a 32-year precipitation-ANPP dataset from the same site as our experiment. We found that increased growing season rainfall variability can reduce RUE and thus ecosystem functioning by as much as 42% during dry years, but that such impacts weaken as years become wetter. During low precipitation years, RUE is lowest when rainfall event sizes are relatively large, and when a larger proportion of total rainfall is derived from large events. Thus, a shift towards precipitation regimes dominated by fewer but larger rainfall events, already documented over much of the globe, can be expected to reduce the functioning of mesic ecosystems primarily during drought, when ecosystem processes are already compromised by low water availability.

Asymmetric responses of ecosystem productivity to rainfall anomalies vary inversely with mean annual rainfall over the conterminous United States.

The CONterminous United States (CONUS) presents a large range of climate conditions and biomes where terrestrial primary productivity and its inter-annual variability are controlled regionally by rainfall and/or temperature. Here, the response of ecosystem productivity to those climate variables was investigated across different biomes from 2010 to 2018 using three climate datasets of precipitation, air temperature or drought severity, combined with several proxies of ecosystem productivity: a remote sensing product of aboveground biomass, an net primary productivity (NPP) remote sensing product, an NPP model-based product and four gross primary productivity products. We used an asymmetry index (AI) where positive AI indicates a greater increase of ecosystem productivity in wet years compared to the decline in dry years, and negative AI indicates a greater decline of ecosystem productivity in dry years compared to the increase in wet years. We found consistent spatial patterns of AI across the CONUS for the different products, with negative asymmetries over the Great Plains and positive asymmetries over the southwestern CONUS. Shrubs and, to a lesser extent, evergreen forests show a persistent positive asymmetry, whilst (natural) grasslands appear to have transitioned from positive to negative anomalies during the last decade. The general tendency of dominant negative asymmetry response for ecosystem productivity across the CONUS appears to be influenced by the negative asymmetry of precipitation anomalies. AI was found to be a function of mean rainfall: more positive AIs were found in dry areas where plants are adapted to drought and take advantage of rainfall pulses, and more negative AIs were found in wet areas, with a threshold delineating the two regimes corresponding to a mean annual rainfall of 200-400 mm/year.

Rainfall manipulation experiments as simulated by terrestrial biosphere models: Where do we stand?

Changes in rainfall amounts and patterns have been observed and are expected to continue in the near future with potentially significant ecological and societal consequences. Modelling vegetation responses to changes in rainfall is thus crucial to project water and carbon cycles in the future. In this study, we present the results of a new model-data intercomparison project, where we tested the ability of 10 terrestrial biosphere models to reproduce the observed sensitivity of ecosystem productivity to rainfall changes at 10 sites across the globe, in nine of which, rainfall exclusion and/or irrigation experiments had been performed. The key results are as follows: (a) Inter-model variation is generally large and model agreement varies with timescales. In severely water-limited sites, models only agree on the interannual variability of evapotranspiration and to a smaller extent on gross primary productivity. In more mesic sites, model agreement for both water and carbon fluxes is typically higher on fine (daily-monthly) timescales and reduces on longer (seasonal-annual) scales. (b) Models on average overestimate the relationship between ecosystem productivity and mean rainfall amounts across sites (in space) and have a low capacity in reproducing the temporal (interannual) sensitivity of vegetation productivity to annual rainfall at a given site, even though observation uncertainty is comparable to inter-model variability. (c) Most models reproduced the sign of the observed patterns in productivity changes in rainfall manipulation experiments but had a low capacity in reproducing the observed magnitude of productivity changes. Models better reproduced the observed productivity responses due to rainfall exclusion than addition. (d) All models attribute ecosystem productivity changes to the intensity of vegetation stress and peak leaf area, whereas the impact of the change in growing season length is negligible. The relative contribution of the peak leaf area and vegetation stress intensity was highly variable among models.

Temporal variability in production is not consistently affected by global change drivers across herbaceous-dominated ecosystems.

Understanding how global change drivers (GCDs) affect aboveground net primary production (ANPP) through time is essential to predicting the reliability and maintenance of ecosystem function and services in the future. While GCDs, such as drought, warming and elevated nutrients, are known to affect mean ANPP, less is known about how they affect inter-annual variability in ANPP. We examined 27 global change experiments located in 11 different herbaceous ecosystems that varied in both abiotic and biotic conditions, to investigate changes in the mean and temporal variability of ANPP (measured as the coefficient of variation) in response to different GCD manipulations, including resource additions, warming, and irrigation. From this comprehensive data synthesis, we found that GCD treatments increased mean ANPP. However, GCD manipulations both increased and decreased temporal variability of ANPP (24% of comparisons), with no net effect overall. These inconsistent effects on temporal variation in ANPP can, in part, be attributed to site characteristics, such as mean annual precipitation and temperature as well as plant community evenness. For example, decreases in temporal variability in ANPP with the GCD treatments occurred in wetter and warmer sites with lower plant community evenness. Further, the addition of several nutrients simultaneously increased the sensitivity of ANPP to interannual variation in precipitation. Based on this analysis, we expect that GCDs will likely affect the magnitude more than the reliability over time of ecosystem production in the future.