Methods

Study System
We conducted fieldwork at McLaughlin Natural Reserve in Northern California (38.860643 N, -122.364156 W; Fig. S1). The Reserve lies within a Mediterranean climate region, characterized by cool wet winters and hot dry summers. During the 2019/2020 growing season, the average temperature was 20.1℃, in line with a five-year average of 19.3℃ (i.e., 2014-2019; range = 17.8℃-20.6℃; SD = 0.63; Western Regional Climate Center 2022). Total annual rainfall during 2019/2020 was 35.89cm, lower than the average of 82.11cm observed during the reference period (range = 45.14cm-129.79cm; SD = 33.92; Western Regional Climate Center 2022). McLaughlin Reserve is dominated by serpentine soils, a soil type that is common in subduction zones. Serpentine soils are characterized by a high heavy metal content, low soil moisture retention, nutrient limitation, and high Mg:Ca ratios. Despite these poor conditions, many plants from serpentine regions have adapted to tolerate serpentine soils (Anacker 2014). However, serpentine soils can vary dramatically in fertility, chemical composition, and physical properties, resulting in stark differences in plant productivity even among sites only 100m apart.
Species and Site Selection
We chose four focal annual plant species to use as our transplants:Bromus hordeaceus (non-native grass), Festuca microstachys(native grass), Micropus californicus (native herb), andPlantago erecta (native herb). We chose these species because they differ in life history (two grasses, two herbs), provenance (three native, one non-native), and are common across the field site. As annual plants, individuals can be transplanted to new locations as seeds, and we can measure germination, survival, and reproduction in a growing season.
We chose six sites for our transplant experiments and to source seeds from. These sites were chosen because they were in close proximity to sites used in previous work (Germain et al.2017), they varied dramatically in total plant production (from mostly bare ground (5% cover) to dense cover (100% cover )), and they were relatively evenly spaced to cover a total spatial extent of 45 km2. In May of 2019, we sourced seeds from hundreds of individuals in approximately equal numbers from a total of 18 populations; three located within each of the six sites. Critically, sourced populations did not overlap with exact transplant locations to minimize any home site advantages of seeds collected from and transplanted back into the same locations. We pooled all collected seeds per species to create a “regional” pool of seeds, so that any differences in success among seeds due to site of origin would be averaged across all of our transplanted individuals.