Nitrogen rather than water availability limits aboveground primary productivity in an arid ecosystem: substantial differences between grasses and shrubs
Short title: Aboveground primary productivity in response to nitrogen and water addition
Luisina Carbonell-Sillettaab, Fabian Gustavo Scholz ab, Antonella Burek ab, Virginia Diaz Villac, Agustin Cavallaro ad, Javier Oscar Askenazi e, Nadia Soledad Ariasab, Guang-You Hao f, Guillermo Goldstein cg, Sandra Janet Bucci ab
a Grupo de Estudios Biofísicos y Ecofisiológicos (GEBEF), Instituto de Biociencias de la Patagonia (INBIOP), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) and Universidad Nacional de la Patagonia San Juan Bosco (UNPSJB), (9000) Comodoro Rivadavia, Argentina
b Facultad de Ciencias Naturales y Ciencias de la Salud, UNPSJB, (9000) Comodoro Rivadavia, Argentina
c Laboratorio de Ecología Funcional. Instituto de Ecología, Genética y Evolución de Buenos Aires. UBA – CONICET (1425) Buenos Aires, Argentina
d INTA EEAf Esquel, Esquel, Chubut, Argentina
e Centro de Investigación y Transferencia del Golfo San Jorge (CONICET – UNPSJB), (9000), Comodoro Rivadavia, Chubut
f CAS Key Laboratory of Forest Ecology and Management & Key Laboratory of Terrestrial Ecosystem Carbon Neutrality, Liaoning Province, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, China
g Department of Biology, University of Miami, Coral Gables, Florida, USA
Author for correspondence: Sandra Janet Bucci
Email: sj_bucci@yahoo.com
Funding: This study was funded by Fondo para la Promoción Científica y Tecnológica (FONCyT; grant PICT 2010-960, PICT 2013-2426, PICT 2016-3019, PICT 2019-1306) and CONICET (PUE INBIOP 0033).
Acknowledgment: We thank to staff of Instituto Nacional de Tecnología Agropecuaria (INTA) for allowing the access and assistance in the Rio Mayo Experimental Field. This work complies with Argentinian law.
Data and code availability: The datasets generated and analyzed and the R code generated during this study are available from https://osf.io/cmj4b/.
Author Contributions: FGS, SJB and GG conceived and designed the experiments. LCS, FGS, SJB, AB, AC and JOA conducted fieldwork. VDV performed imaging analysis. LCS performed statistical analyses. LCS, SJB and GG wrote the manuscript. GYH and NSA contributed to the discussion of the results.
Abstract
Changes in water and nitrogen availability can affect the structure and function of arid ecosystems. How these resources affect aboveground primary productivity (ANPP) remains far from clear. We examined the N and water limitation of ANPP from the species to the community level and the response of ANPP to annual precipitation in a Patagonian steppe. We conducted a 7-year field experiment with water addition (+W), nitrogen addition (+N), and +NW. Destructive methods for grasses and allometric relationships for shrubs were used to assess ANPP and vegetation indices (NDVI and MSAVI2) to estimate community ANPP. An increase in ANPP of one grass species (Papposstipa humilis ) and a decrease of the grassPoa ligularis under +N were observed. Some shrubs species exhibited mortality under nitrogen addition. Nitrogen exerted a positive effect on grass ANPP and amplified the sensitivity of grass ANPP to annual precipitation. However, +N had not effects on the shrub ANPP and shrub ANPP-precipitation relationship. Water addition by itself had no effect on ANPP for either shrubs or grasses. However, shrubs responded positively to an unusually wet year regardless of treatment and were also more sensitive to changes in annual precipitation than grasses. Total ANPP increased significantly in +N relative to the C and +W, but without changes in the sensitivity to annual precipitation. The results suggest that the responses of grasses and shrubs to water inputs is driven by soil moisture redistribution and rooting depth and that grass and community ANPP is more limited by N than by water.
Keywords: anomalous precipitation, water addition, nitrogen addition, Patagonian steppes, rooting depth, soil water distribution

Introduction

Aboveground net primary productivity (ANPP) is an attribute that integrates key aspects of ecosystem functioning. The ANPP is related to carbon and water cycling and energy fluxes. In arid and semiarid ecosystems, it has been generally observed that ANPP is strongly limited by water and nutrient availability (Guo et al., 2022; Lü et al., 2014; Ma et al., 2020; X. Zhang et al., 2021). Several studies have indicated a positive linear relationship between ANPP and annual precipitation across multiples sites (Bai et al., 2008; A. K. Knapp & Smith, 2001; Sala et al., 2012). At a specific site, the response of ANPP to precipitation can be linear (Sala et al., 2012; Wang et al., 2014; Wilcox et al., 2017; Yahdjian & Sala, 2006) or non-linear, with positive (Ahlström et al., 2015; Felton et al., 2019; L. Zhang et al., 2022; T. Zhang et al., 2020) or negative effects (Hou et al., 2021; Wu et al., 2018). These inconsistent relationships between ANPP and precipitation may be due to differences in species composition, the range of precipitation considered, or biogeochemical factors affecting the sensitivity of species to changes in precipitation (Deng et al., 2021; Felton et al., 2019; Knapp et al., 2017). For example, in some cases nutrient limitations restrict the effect of water on ANPP (Austin & Sala, 2002; Harpole et al., 2007; X. Zhang et al., 2021). On the other hand, increased soil N content alone usually stimulates primary productivity (Austin & Sala, 2002; Harpole et al., 2007; LeBauer & Treseder, 2008; Tang et al., 2017; Xia & Wan, 2008; Yue et al., 2020). Furthermore, many studies indicate that the ANPP of arid ecosystems is co-limited by water and nutrients (Guo et al., 2022; Harpole et al., 2007; Lü et al., 2014),
In addition to the typical water and nutrient restrictions to which arid ecosystems are exposed, it should be considered that these ecosystems are sensitive to climate change and other global changes (Li et al., 2021; Niu & Wan, 2008; Rudgers et al., 2018). Changes in the amount and temporal pattern of precipitation are predicted for arid regions, with increases or decreases depending on the region, but with a significant increase in extreme events (Drumond et al., 2019; Giorgi et al., 2019; Saurral et al., 2017). Moreover, global N deposition is increasing as a consequence of intensive anthropic activities (agriculture, oil, gas production, among others) even in arid and semiarid regions (Bobbink et al., 2010; Dentener et al., 2006; Gruber & Galloway, 2008). Increased N deposition and altered precipitation regimes can have significant impacts on carbon, water and nutrient cycling (Gruber & Galloway, 2008; Harpole et al., 2007; Nielsen & Ball, 2015), through changes in vegetation structure and ANPP (Song et al., 2019; Stevens et al., 2015). Some studies (Ahlström et al., 2015; c et al., 2020; Poulter et al., 2014) have shown that semiarid regions play a key role, more than other ecosystems, in regulating intra- and interannual variability of the global carbon cycle. These changes in N and water inputs to which ecosystems are exposed may occur at the same time, and their interaction may not be directly predicted by the additive effects of individual resources. Some studies have shown that the effects of water and N amendment on primary productivity are additive (Gong et al., 2011; X. Zhang et al., 2021), while others show non-additive effects (Gao et al., 2011; Niu et al., 2009).
In addition to changes in ANPP due to changes in water and nutrient availability, species composition, species interaction, and stability of dominant species may be modified with the addition of limiting resources (Guo et al., 2022; Lannes et al., 2016; Yue et al., 2020; X. Zhang et al., 2021). The addition of water usually increases species composition and ecosystem functioning (Hu et al., 2022; Weltzin et al., 2003; Yue et al., 2020). Plant diversity can be affected by N addition through competition exclusion, species invasion, or soil nutrient imbalances (Liu et al., 2019; Tian et al., 2016). However, plant community responses to N addition vary between ecosystems and, although the most common response is the reduction of species diversity by competitive exclusion, the opposite effect has been found in resource-poor ecosystems (Bai et al., 2010; Chalcraft et al., 2008; Ladwig et al., 2012). Long-term N addition can decrease soil pH and affect the growth and survival of species sensitive to soil acidification caused by excessive N input, thus reducing plant diversity (Fang et al., 2012; Ke et al., 2023; Wallace et al., 2007).
The Patagonia region of southern Argentina is mainly represented by arid/semiarid ecosystems with pronounced interannual precipitation variability (Jobbágy & Sala, 2000; Paruelo et al., 1998) and a low pool of available soil N (Carbonell Silletta et al., 2019; Yahdjian et al., 2014). Climate change predictions indicate a reduction in precipitation in Patagonia, except in some areas such as the northwest of Chubut, where there is a tendency to increase rainfall (Barros et al., 2015; Saurral et al., 2017). Also, although N deposition in arid areas is lower than in other ecosystems, it is expected to increase, depositing mostly on natural vegetation (Dentener et al. 2006; Gruber & Galloway 2008). Several studies have evaluated the patterns and controls of aboveground production in relation to climatic variables and vegetation structure in the Patagonian steppe at both spatial and temporal scales (Austin & Sala, 2002; Fernández et al., 1991; Gaitán et al., 2014; Jobbágy & Sala, 2000; Sala et al., 2012;Yahdjian & Sala, 2010). Other studies have evaluated the ecosystem response to reduced water inputs (Yahdjian & Sala, 2006), increased water inputs following previous years of water restriction (Yahdjian & Sala, 2006), or a single relatively large experimental rainfall event applied during the summer (Golluscio et al., 1998). These studies have been short-term (1 to 2 years) and have evaluated the immediate plant response to changes in water inputs. However, there are no studies that have evaluated the longer-term effects of both water and nitrogen additions, as well as the interaction of the addition of both resources.
In this research, we explore how soil N additions and water inputs and their interaction affect the structure of vegetation and aboveground primary productivity in a Patagonian steppe. We hypothesized that N and water addition increases ANPP and that this effect is higher with the co-addition of both resources, and higher in species with shallow root systems (grasses) than in higher in species with deeper root systems (shrubs). Furthermore, we hypothesized that at the temporal scale ANPP is positively correlated with precipitation, a relationship that is strengthened with increased nutrient availability. The last hypothesis proposes that nitrogen addition, alone or together with water addition, induces an increase in the abundance of nitrophilous grasses. We carried out a field experiment with water and nitrogen addition during 2013-2019. We also took advantage of the large interannual variability in precipitation observed between 2015 and 2019 to test the second hypothesis. In March 2017, a large part of the Patagonian steppe experienced an extreme rain event that added about 80% more water than the mean annual precipitation of the last four decades in the study area. This was a unique opportunity to assess the ecosystem response to an extreme precipitation event.

Materials and Methods