Climate legacies determine grassland responses to future rainfall regimes

Abstract

Climate variability and periodic droughts have complex effects on carbon (C) fluxes, with uncertain implications for ecosystem C balance under a changing climate. Responses to climate change can be modulated by persistent effects of climate history on plant communities, soil microbial activity, and nutrient cycling (i.e., legacies). To assess how legacies of past precipitation regimes influence tallgrass prairie C cycling under new precipitation regimes, we modified a long‐term irrigation experiment that simulated a wetter climate for >25 years. We reversed irrigated and control (ambient precipitation) treatments in some plots and imposed an experimental drought in plots with a history of irrigation or ambient precipitation to assess how climate legacies affect aboveground net primary productivity (ANPP), soil respiration, and selected soil C pools. Legacy effects of elevated precipitation (irrigation) included higher C fluxes and altered labile soil C pools, and in some cases altered sensitivity to new climate treatments. Indeed, decades of irrigation reduced the sensitivity of both ANPP and soil respiration to drought compared with controls. Positive legacy effects of irrigation on ANPP persisted for at least 3 years following treatment reversal, were apparent in both wet and dry years, and were associated with altered plant functional composition. In contrast, legacy effects on soil respiration were comparatively short‐lived and did not manifest under natural or experimentally‐imposed “wet years,” suggesting that legacy effects on CO2 efflux are contingent on current conditions. Although total soil C remained similar across treatments, long‐term irrigation increased labile soil C and the sensitivity of microbial biomass C to drought. Importantly, the magnitude of legacy effects for all response variables varied with topography, suggesting that landscape can modulate the strength and direction of climate legacies. Our results demonstrate the role of climate history as an important determinant of terrestrial C cycling responses to future climate changes.

Climate Legacy Effects Shape Tallgrass Prairie Nitrogen Cycling

“…e ., centuries‐millennia of differences in water availability) do not find changes in N mineralization with mean AP, which has been attributed to concurrent increases in soil organic C that increase potential N immobilization (Barrett et al., 2002; Feyissa et al., 2021; McCulley et al., 2009). We did not find changes in soil C during the duration of our study (Broderick et al., 2022; Wilcox et al., 2016) or legacies of previous irrigation on soil moisture. Moreover, while historic rainfall patterns may have altered N mineralization via effects on litter chemistry (i.e., belowground plant inputs in this annually burned system), the lack of response of root biomass and chemistry to rainfall treatments (Broderick et al., 2022) suggests that changes in plant C inputs were not the driver of N cycling legacies.…”

“…Conversely, in previously irrigated prairie now subject to natural water deficits, drought-induced mortality in the microbial community when soils experienced extended dry periods may result in a flush of labile organic N resources (Xiang et al, 2008). Previous results from this experiment revealed that long-term release from water stress resulted in greater MBC, but upon reversal to ambient conditions there was a sharp drop to biomass levels comparable to non-irrigated controls (Broderick et al, 2022). Therefore, in the I→A reversal treatment, release of nutrients from the death of drought sensitive microbes after repeated dry/wet cycles may have contributed to the increased rates of N mineralization in this treatment (Franzluebbers, 1999;Xiang et al, 2008).…”

“…In order to predict and potentially mitigate the effects of climate change on ecosystem functioning, it will be important to understand the ecosystem properties and processes that underlie climate sensitivity. Long‐term climate patterns can produce legacies that shape C fluxes in the face of novel climate scenarios, and these climate legacies are likely to vary in their magnitude, direction, and causes across and within ecosystems (Broderick et al., 2022; Kannenberg et al., 2020; Sala et al., 2012). In the Central Plains, precipitation is expected to become more variable within and between years, raising questions about how sensitive these ecosystems will be to altered precipitation patterns as well as how previous rainfall history across the region will moderate this sensitivity.…”

“…We also found differences between responses of pools and fluxes; for example, sharp increases in nitrification with previous irrigation in the lowlands did not accompany any response in nitrate pools. This may be due to increased plant uptake in the lowlands with previous irrigation (associated with higher ANPP without changes in plant N content; Broderick et al., 2022), especially in forbs which tend to have a higher N demand (Tjoelker et al., 2005). Importantly, short‐ and long‐term precipitation manipulations had distinct effects on net nitrification rates.…”

“…For example, in many ecosystems the relationship between annual precipitation (AP) and ANPP can be partially explained by precipitation amounts in the year preceding measurement (Griffin‐Nolan et al., 2018; Sala et al., 2012). Accordingly, there is growing evidence that the responses of ecosystems to climate change may depend on historical climate conditions, often referred to as legacy effects (Broderick et al., 2022; Hawkes et al., 2017; Sala et al., 2012). Current climate models do not account for legacy effects that alter the interactions between C and N cycling (Averill et al., 2016), which may be an important driver of sensitivity to future climates.…”