annuals_composition: general protocol

Overview: In spring 2008 within each site, the CAP LTER randomly chose and permanently marked four 1 m x 1 m sub-plots within each 20 m x 20 m main plot, split between two patch types, under L. tridentata shrubs and inter-plant spaces (hereafter, under L. tridentata, P and Inter-plant space; IP). At peak herbaceous biomass (in approximately March of each year), percent cover of all annual herbaceous species rooted within these four plots (two each under L. tridentata and in interplant spaces i.e. between shrubs) will be estimated. Note that there are also four 1 m x 1 m Biomass sub-plots in each 20 m x 20 m plot – these are different than the community composition plots described here because they are clipped each year while the community composition plots are not. Care should be taken in the field to identify the correct four plots for community composition and avoid the four plots to be harvested for biomass. Steps: 1. On the top of the data sheet (See Table 3 for sample), record the date; names of the sampling crew (i.e. plant identifier, recorder); site (three letter code); plot number (1 to 75); treatment (C1 – no sampling from C2 plots, Nitrogen [N], Phosphorus [P], and Nitrogen+Phosphorus [NP]). Double check the plot numbers and sub-plot numbers! 2. All observers should first calibrate their percent cover estimates on several test plots in the field prior to percent cover measurements. You want to make sure all botanists are seeing the plots similarly. 3. Place the 1 m x 1 m quadrat over the corner rebar at each 1-m2 plot. Working in teams of two (one recorder, one plant identifier), proceed through the steps below. 4. Estimate and record (at the bottom of the data sheet) the relative areal cover of plant matter (e.g. green material) versus bare ground (e.g. brown material) at the soil surface in the plot (e.g. 70% green, 30% brown). If there are soil crusts, measure those as well in a separate category. 5. Identify which annual species are present in the plot and mark a small dot in the corner of the datasheet to easily find later. Use roughly the same amount of effort in each plot to find plant species. As you get better, it will take less time because you are getting more skilled; as you get tired, it will take more time; in simpler systems, it will take less time. There is no a priori way to know when you are using the same amount of effort, but this is usually approximated by time (in minutes: typically 1-3 minutes per species). Each botanist should do some test plots to see how long it takes him/her to find all the species in the plot. Different botanists will take different amounts of time to apply the same effort– this is ok – but the same botanist should use approximately the same amount of time per plot. 6. Next, estimate and record the cover of each species individually to the nearest whole number (e.g., 1%, 2%, 3%...97%, 98%, 99%, 100%). If a species is present in a plot, but does not cover a full 1%, record it as <1%. The sum of all individual species cover can be greater than 100% since some species overlap. 7. Finally, estimate and record the cover of the overstory shrubs within the plot or overhanging the plot. We estimate cover for all annuals rooted in the quadrat, but areal cover of any shrubs rooted in or hanging over the plot. For IP plots, this will be zero most of the time (i.e. unless there is a large shrub nearby that has grown over the plot). For P plots, this will be have two components: (1) the areal projection of the shrub onto the 2-D grid, and (2) the area of the stem base or any other shrub stems. It is expected that the cover of the shrubs overlap the herbaceous species. However if two shrub species overlap in their cover aggregate these into a single larger individual (rare case). 8. Unknown species are recorded with unique identifiers (e.g., Unknown 1, Unknown 2, etc) along with a visual description of the plant and a picture to be used to help identify the unknown in the lab. When possible, a sample of the unknown species should be collected from outside of the plot in a labeled coin envelope and brought to the herbarium to be identified. Unknown species are usually very small and_or undeveloped, and may not be able to be identified to species – this is ok. Err on the side of lumping unknown species into other known species in the same plot. 9. Use caution not to include the percent cover of desiccated/dead plants within the plots from the previous season (these can usually be identified easily as white, old plant material). However, depending on the timing of data collection, some herbaceous plants from the current growing season may have passed peak biomass and already begun to desiccate (dry out or senesce). In this case, estimate the percent cover of these plants as if they were at peak biomass. 10. Before you leave a site, confirm that you have datasheets for all plots and sub-plots, and that there are no duplicates or ambiguous plot labels on your sheet.

annuals_composition: 2008 methods

In the spring of 2008, the estimation of the cover of annuals in the study plots was conducted uniquely relative to subsequent years: collector SLC allowed for layering within the herbaceous layer such that total cover when you summed across species could be > 100% (potentially limitless), whereas collector CMC did not allow for layering within the herbaceous layer such that the cumulative sum of cover of annuals cannot be more than 100%. To facilitate comparison, data collected by SLC were relatavized by the total annual cover in the study plot (i.e., the percent of the plot covered by annuals) and the cumulative sum of the cover of annual species in the study plot (i.e., % cover of plant * (Total Plant Cover in Plot / sum of all species (sumSpp))) such that the estimates by both collectors reflect the relative cover of each plant taxon. In subsequent years, values reflect a collection method that allows for layering. Owing to the requisite relativization, data collected by SLC in 2008 were reconstructed from original field data sheets. As such, information gained from post-processing that resolved the identity of species that were not identified in the field was not available resulting in a greater number of unidentified species (unknowns and variants) for data collected by SLC relative to CMC.