This RANCHO SECO_readme.txt file was generated on 2020-11-19 by Julia Michaels GENERAL INFORMATION: 2017-2018_Vegetation_Quadrats.csv 1. Title of Dataset #1: 2017-2018_Vegetation_Quadrats.csv 2. Author Information A. Principal Investigator Contact Information Name: Julia Michaels Institution: University of California, Davis Address: One Shields Ave, Davis, CA 95616 Email: jmichaels@ucdavis.edu B. Associate or Co-investigator Contact Information Name: Valerie Eviner Institution:University of California, Davis Address: One Shields Ave, Davis, CA 95616 Email: veviner@ucdavis.edu 3. Date of data collection (single date, range, approximate date) 2017-05-1 to 2018-08-01>: 4. Geographic location of data collection: Herald, California 5. Information about funding sources that supported the collection of the data: This research was made possible by support from the U.C. Davis Graduate Group in Ecology, U.C. Davis Plant Sciences GSR Fellowship, Sacramento Municipal Utilities District, Area West Consulting, the Davis Botanical Society, CNPS Santa Clara Valley Chapter, and the Northern California Botanists. This work is also supported by the USDA National Institute of Food and Agriculture, Hatch project #1013397. SHARING/ACCESS INFORMATION 1. Licenses/restrictions placed on the data: N/A 2. Links to publications that cite or use the data: In progress 3. Links to other publicly accessible locations of the data: N/A 4. Recommended citation for this dataset: Michaels, Julia (2020), Rancho Seco vernal pool community data, Dryad, Dataset, https://doi.org/10.25338/B8DW6F DATA & FILE OVERVIEW 1. File List: Rancho Seco Vernal Pool Community Data https://doi.org/10.25338/B8DW6F 2. Relationship between files, if important: 3. Additional related data collected that was not included in the current data package: 4. Are there multiple versions of the dataset? no METHODOLOGICAL INFORMATION 1. Description of methods used for collection/generation of data: Site selection Rancho Seco (38.34˚ N,-121.11˚ W) is an 458.10-acre site in Northern California that consists of Mediterranean grassland with vernal pools. Rancho Seco hosts Northern Hardpan vernal pools which span a range of characteristics, including soil type (Redding Gravelly Loam and Corning Complex) and pool size (9.16-336.94 m2), which may influence plant composition (Croel & Kneitel, 2011; Kneitel, 2015). Most of the site has been grazed continuously with commercial beef cattle for over 100 years. Over the past 20 years mean annual stocking rate has been 0.31 Animal Unit Year (AUY) per hectare with a nine-month grazing season from October through June, which is considered low to moderate for commercial cattle operations in the region. Livestock were excluded from a portion of this site in the 1970’s. In January 2016, a subset of this excluded area was reintroduced to cattle grazing at the stocking rate as long-term grazing sites, creating three different grazing management areas across the vernal pool landscape. We selected 12 pools in each management area (Long-term grazed, Grazing exclusion, and Short-term grazing reintroduction), making sure that each group of selected pools had a similar balance of soil type and pool size (m2) using ArcGIS and USGS SoilWeb (Appendix A). Pool basin depth and catchment size After the pools were selected, we also characterized each pool by basin depth (cm) and catchment area (m2) in order to account for these additional environmental factors that may affect vegetation characteristics at the pool scale. We calculated basin depth using a Real Time Kinematic (RTK) geographic positioning system (GPS), subtracting the height of the lowest point in the pool from points taken at the highest edge of the pool perimeter. Basin depth remains constant in vernal pools while surface water level fluctuates. Basin depth ranged from 53.95 cm to 516.64 cm. We calculated catchment as the area of upland from which the pool collects surface and groundwater, for each individual vernal pool using RTK GPS points collected at intervals while walking transects across the landscape. Since pools can be connected to one another via subsurface groundwater flow (Bartolome et al. 2007) a single upland point could be designated to multiple pool catchemnts. These points were processed using Surfer software (version 16, Golden Software, Golden CO). Pool catchment area ranged 39-5029 m2. Vegetation Monitoring We focused our sampling efforts in the transition zone ecotones within the pools where the pool basin (longest inundated area of pool) meets the non-inundated uplands surrounding the pool (Marty 2005, 2015, Robbins and Vollmar 2002, Michaels et al. 2020). The transition zones are small in area compared to the area of the uplands and pool basins, but are areas of conservation concern because of their high diversity and sensitivity to grazing (Marty 2005, Michaels 2020). We delineated these transition zones in early spring 2016, after the pools dried down and before plant taxa were identifiable in each pool, by recording water marks across the slope that indicated differences in flooding duration. In May 2018, during peak flowering season, we sampled species using 50x50 cm PVC quadrat frames divided into one hundred 5x5 cm squares. For each species, we recorded the number of cells in which the canopy of the species occurred. This sampling method is beneficial for detecting less common species as well as tiny forbs that are common in vernal pools (Gerhardt and Collinge 2003). We repeated this procedure three times in each pool (3 quadrats/pool X 12 pools per management type X 3 management types=118 quadrats total). Water level: We measured standing water depth in each pool in winter 2017-18 by recording the height of standing water above the lowest point of the pool basin. These depths, checked weekly from November 2017 to June 2018, were used to calculate the total number of days of inundation for each pool per season. We assumed that hydrology remained stable in the time between consecutive measurement dates (i.e., if a pool held water at three points taken 7 days apart from each other, we recorded this as two weeks of inundation) (Shaffer et al. 2000). This measure is the most robust way to assess the hydrology of a vernal pool, since inundation of the transition zone itself can be highly variable, depending on elevational profile and greater short-term temporal variability in inundation in the transition zone. Hoof print cover: We measured hoof prints in the transition zones during the May 2018 vegetation quadrat surveys. In each quadrat, the number of 5x5 cm cells in which a hoof print that depressed and exposed the soil at least 13 mm deep was counted and recorded (Burton, Smith, and Cowley, 2011) Residual dry matter (RDM): In order to measure upland biomass that could potentially reduce standing water in the pools, we collected RDM in the upland zones immediately adjacent to the transition zones, following removal of livestock for the 2017-2018 grazing season (July 2018). Residual dry matter is the mass (kg/ha) of dry, senescent annual herbaceous vegetation remaining following the growing-grazing season (October through June), and provides a relative estimate of vegetation biomass. Quadrats (15x15 cm) were placed in the upland zones within 5 feet of the pool edges and all biomass per quadrat was clipped and air dried at 60oC to a constant weight. 2. Methods for processing the data: Data analyses We tested for differences between livestock management treatments on three types of plant community response variables: species richness, abundance-weighted diversity (Shannon-Weiner Index), and the relative cover of natives. We used linear mixed effects models (LMMs) with pool identity specified as a random intercept to account for correlation among quadrats at the same pool (n = 3 per pool). We specified a normal distribution in the analysis, after confirming residual normality via standard diagnostic plots. Days of Inundation: To test the prediction that (1) continuously grazed pools would have the longest inundation, and (2) that pools recently reintroduced to grazing would exhibit increases in inundation period over time, we used LMM with all pool covariables and Pool ID as a random effect. Hoof prints: To test the prediction that the pools recently reintroduced to grazing would exhibit higher hoof print microtopography than continuously grazed pools, we used an LMM on our quadrats individually with all pool variables and pool ID as a random effect. Relating grazing mechanisms to plant diversity and cover: We tested (1) the relative strength of grazing treatment and grazing-related environmental covariables (days of inundation, hoof print microtopography, and RDM) and (2) the level with which they interact with other environmental covariables (pool area, soil type, pool basin depth, and catchment area) to affect plant diversity and native cover. We used a conditional inference tree (CIT) in order to select the key subset of these covariates for each diversity metric (Hothorn et al. 2006). We included grazing treatment as a covariable in our CIT in order to represent grazing-related mechanisms that we were unable to account for in our study (i.e., direct defoliation, soil compaction, nutrients, etc). This CIT recursive partitioning algorithm looks for a predictor and a threshold at each stage that divides between low and high responses for the dependent variable. With this subset of covariables, we determined the best LMM model using AIC model selection, testing each predictor separately and combined and in interaction, if an interaction was indicated by the CIT analysis. Pairwise comparisons were made using Tukey’s Honestly Significant Difference (HSD) test (Rohlf et al., 1995). Statistics were conducted using the vegan and lme4 packages in R v. 3.5.2 (R Core Team 2018; Oksanen et al. 2019; Bates et al. 2015). To further identify the specific species that were driving grazing effects on the plant community, we also compared the mean relative abundance of each species for each grazing treatment using two-sample t-tests and the Benjamni-Hochberg correction for multiple comparisons (Benjamini and Hochberg 1995, Waite & Campbell, 2006). Data available from the Dryad Digital Repository: https://doi.org/10.25338/B8DW6F Michaels et al. 2020 3. Instrument- or software-specific information needed to interpret the data: All analyses were completed in R statistical software v3.6. DATA-SPECIFIC INFORMATION FOR: [FILENAME] 1. Number of variables: 82 2. Number of cases/rows:112 3. Variable List: Pool ID: ID code for each vernal pool, corresponding to a labeling system already in use at Rancho Seco Mitigation bank. Quadrat: Quadrat ID #. 3 quadrats/pool Grazing: Grazing management of the pasture in which the vernal pool is located. Pair: ID used to balance each treatment group by pool characteristics. Hoofprint: % cover of hoof prints within each Quadra. SoilType: Soil type. RDM: Residual Dry Matter measured in each pool. Catchment:Size (m) of the local watershed that drains into each pool. Size: Area (m2) of pool Depth: Depth (cm) of each pool Wetup Timing: Time of initial inundation: Early (Nov), Mid (Dec-Jan) Mid-season dry period: Three categories referring to the length of the mid-season dry period for each pool: Long dry (>7 days), Mid dry (3-7 days), Short dry (0-3 days) Inundation timing: Stable (<3 dry down periods), Variable (3 or more dry down periods) BarGr through TriVa: Vernal pool species codes. See abbreviations below. 4. Missing data codes: N/a 5. Specialized formats or other abbreviations used: Species Abbreviation AcmAm Acmispon americanus AirCa Aira caryophella AloSa Alopecurus saccatus AveFa Avena fatua BarGr Bare ground BleNa Blennosperma nanum BriMi Briza minor BroAp Brodiaea appendiculata BroDi Bromus diandrus BroHo Bromus hordeaceous BroMi Brodiaea minor CasAt Castilleja attenuata CasCa Castilleja campestris CenSp Centromadia spp. ChlPo Chlorogalum pomeridianum CinQu Cicendia quadrangularis CotCo Cotula coronopifolia CroSe Croton setigerus DesDa Deschampsia danthanoides DowBi Downingia bicornuta DowCu Downingia cuspidata DowOr Downingia ornatissima EleAr Eleocharis acicularis EleMa Eleocharis macrostachya ElyCa Elymus caput-medusae EroBo Erodium botrys EryCa Eryngium castrense EscLo Eschscholzia lobii FesBr Festuca bromoides FesPe Festuca perennis GerDi Geranium dissectum GlyDe Glyceria declinata GraEb Gratiola ebracteata HolVi Holocarpha virgata HorMa Hordeum marinum HypRa Hypochaeris radicata JunBa Juncus balticus JunBu Juncus bufonius JunCa Juncus capitatus JunPa Juncus patens LasFr Lasthenia fremontii LasGl Lasthenia glaberrima LayFr Layia fremontii LeoSa Leontodon saxatilis LupBi Lupinus bicolor LytHy Lythrum hyssopifolia NavLe Navarretia leucocephala NavMe Navarretia Mersii PilAm Pilularia americana PlaLe Plagiobothrys leptocladus PlaSt Plagiobothrys stipitatus PogZi Pogogyne zizyphoroides PsiBre Psilocarphus brevissimus RanBo Ranunculus bonariensis TriDe Trifolium depauperatum TriDu Trifolium dubium TriEr Trifolium hirtum TriHi Trifolium hirtum TriHy Triphysaria eriantha TriLa Triteleia hyacinthina TriLa Triteleia laxa TriVa Trifolium variegatum