I. Monitoring studies. Our monitoring studies assess how key populations and processes respond to long-term climate variability, and determine important feedbacks between consumer and producer populations, all of which are explicit components of the TDND model.
A. Producers. Plant community composition and NPP are governed by seasonal and interannual variation in precipitation. The goals of our producer studies are to quantify above- and belowground NPP, and plant community composition, structure and dynamics in response to climate variability, and to follow grassland recovery following the 2009 wildfire. NPP is a key integrating variable (McNaughton et al. 1989) and understanding pattern and control of NPP is required of all LTER sites. Our TDND model includes drivers that affect NPP (Muldavin et al. 2008). We measure seasonal aboveground NPP at our core blue grama, black grama and creosote sites, NFert, WENNDEx, MRME, in burned and unburned mixed grassland, in burned and unburned grassland-shrubland transition areas, and in understory vegetation in piñon-juniper woodland. We use a non-destructive allometric approach (Huenneke et al. 2001, Muldavin et al. 2008) that it allows us to measure aboveground NPP by species at the same point over time. Results from our long-term NPP studies show that grassland productivity is highly variable over time and poorly correlated with seasonal precipitation or soil moisture. Production in creosote-dominated areas is much less variable, and related to spring soil moisture (Muldavin et al. 2008).
Starting in 2005, belowground net primary production (BNPP) has been measured using root ingrowth donuts (Milchunas et al. 2005, Milchunas 2009) at NFert, in burned and unburned mixed grass sites, at the creosote core site, and starting in 2011 at MRME. Donuts are harvested yearly at two depths, 0-15 cm and 15-30 cm. Newly collected soil from adjacent areas is then sifted and used to reconstruct the root ingrowth donut for the next annual harvest. In addition to BNPP, we have greatly increased studies of other belowground processes (Stursova et al. 2006, Crenshaw et al. 2007, Zeglin et al. 2007, Green et al. 2008, Porras-Alfaro et al. 2008, 2009, 2010, Vargas et al. 2010, in review, Pockman & Small 2010, Ladwig et al. 2011). We will continue seasonal measurements of fungal hyphae and fine root dynamics in a series of minirhizotrons located in 1) mixed black grama-blue grama grassland, 2) creosote shrubland, 3) MRME and 4) NFert.
Differences in winter vs. summer precipitation regimes are reflected in species, functional group, and community responses as mediated by soil water content, both of which affect consumer abundance and dynamics (see below). We measure vegetation composition and structure seasonally each year at 1-cm resolution along two 400-m, permanently located line-intercept transects. One of these transects was burned in 2009. Prior to the fire (1989-2008), these data supported our patch dynamics model (Peters et al. 2006) prediction that patch boundaries are either dynamic, static, or shifting leading to a large-scale patch mosaic that changes (pulses) at different rates over time (Collins & Xia in prep).
B. Consumers. Animal abundances integrate across multiple pulse events (seasonal and interannual), creating lag effects between precipitation and population dynamics (Yates et al. 2002, Chesson et al. 2004, Warne et al. 2010). The Sevilleta ecosystem is strongly bottom-up driven (Baez et al. 2006) and heavily depended on seasonal patterns of NPP. The goals of our consumer studies are to quantify population abundance, interannual variability, and physiological responses of key taxa (small mammals, bees, grasshoppers) in response to seasonal and interannual variation in precipitation and fluctuations in resource availability. Mammal trapping webs were established in black grama grassland, creosote shrubland, and piñon-juniper woodland in 1989, and in blue grama grassland in 2000 (Parmenter et al. 2003). Populations have been monitored twice each year (spring, fall), but in LTERV we will add a third trapping period each year to better quantify population dynamics (Ernest et al. 2000). Grasshoppers, another key consumer in aridland ecosystems (Ritchie 2000), have been shown to vary along spatial and elevational gradients at SEV (Rominger et al. 2009). To determine distribution and abundance of these important consumers, grasshopper abundance and community composition are measured annually along survey transects adjacent to the mammal trapping webs. Numerous populations of native bee species occur at SEV where the rarity of Apis mellifera makes SEV ideal for the study of native pollinators. Blue and yellow funnel traps located near the trapping webs have been used to sample bees every two weeks from March through October since 2002. Traps are located at core blue grama, black grama, and creosote sites. Bee composition and abundance are related to temporal pulses in the abundance of plant populations (Mulhouse et al. in prep) along with interannual variation in floral phenology (Wetherill et al. in review).
SEV provides an excellent venue for understanding how changing climates may affect producer–consumer interactions and population dynamics. Here, these studies explicitly couple rainfall pulses and NPP with consumer trophic structure in our TDND model. Variation in winter rainfall patterns and inputs directly influences the biomass of C3 species (Muldavin et al. 2008, Xia et al. 2010) whereas C4 productivity is largely driven by the summer monsoon. These differences in photosynthetic pathways among plant functional groups, and hence C isotope ratios, allow us to partition, quantify, and track C (and by proxy N and energy) movements from producers into consumers at a variety of trophic levels. Climate change is likely to differentially influence seasonal patterns of precipitation, and these changes will undoubtedly alter NPP among plant functional types. As a consequence, insight into how consumers interact with specific resource systems (C3 vs. C4 functional types) is critical to understanding how climate change might influence the structure and abundance of consumer populations. Grasshoppers and ants are among the most important primary consumer groups at SEV and provide significant biomass to higher tropic levels. We make monthly collections of grasshoppers and ants during the growing season. Using TDL, we measure the C, N and O isotope ratios of consumer “breath” (Martinez del Rio & Wolf 2005, Engel et al. 2009, Warne et al. 2010, 2011) to assess water stress and determine the relative importance of C and N derived from C3 vs C4 plant production associated with winter or summer rainfall. Also, we periodically measure isotope ratios in prairie dogs to assess their health and status during the colony restoration process.
II. Experimental manipulations. On-going experiments are designed to understand interactions among dominant plant functional types in our model, as well as how external forces (e.g., fire) affect plant community composition and structure.
Removal experiment. At the patch scale, we measure cover and composition of vegetation on 3x4 m plots established in 1995, where we annually remove aboveground biomass of dominant species. Replicate and control plots are located in areas dominated by blue grama, black grama or creosote or in communities co-dominated by blue and black grama, and black grama and creosote. This long-term experiment provides information on the dynamics of dominant species at SEV (Peters & Yao In review). Results thus far show that dominant C4 grasses exert strong controls on forb abundance and diversity.
Seasonal fire experiment. In 2007, we established a replicated experiment to determine the effects of fire seasonality on grassland community composition. Plots are 20mx30m (n=4) and replicates were burned in November 2007, March 2008 and June 2008. Species composition is measured periodically in permanent subplots nested in each replicate. Burning will be repeated once every 6-8 years. Results thus far show that summer fires result in highest short-term plant species diversity (Ladwig et al. in prep).
Gunnison’s prairie dog restoration experiment (GPDREx). Population dynamics of native consumers have been dramatically altered during the past century, including local extirpation of a keystone species, Gunnison’s prairie dog (Cynomys gunnisonii). Prairie dogs are keystone species that physically alter soil structure and nutrient content and create habitat patches high in biodiversity (Fahnestock & Detling 2002, Davidson & Lightfoot 2008, 2009, 2010). In 2005, we initiated an experimental restoration of Gunnison’s prairie dogs at SEV. The primary goals of GPDREx are to determine how prairie dog colonies influence plant community structure and ecosystem processes, and how changes in community structure affect consumer resources within and among seasons. Since 2005 prairie dogs periodically have been added to replicate colonies. Prairie dog population size is estimated each year through trapping and observation towers. Plant species composition and abundance are collected twice each year at 36 permanent points within a 50x50 m grid centered on each experimental colony or paired non-colony area. Vegetation standing crop is estimated each year by clipping all vegetation rooted in a 1-m2 plot near each permanent sampling point. Active mounds are mapped every fall. This design allows us to track spatial and temporal changes in community composition, resource availability and ecosystem dynamics as burrows are abandoned and new burrows are colonized.