PS 38-51 - Optimizing sampling across methods improves the power of ecological monitoring data

Abstract: Measuring ecosystem processes and functions is critical to providing data to managers and researchers seeking to understand ecological patterns and change. Transect-based monitoring methods have long been a valuable tool in ecosystem monitoring and often the same transects are used to measure multiple ecosystem attributes. The line-point intercept, vegetation height, and canopy gap intercept monitoring methods have together been identified as a set of “core methods” which can provide indicators of dryland condition. There is a lack of understanding of the impact of sampling decisions with respect to the number, length, and sampling intensity for each of these monitoring methods. No resource is available to help users design a sampling strategy that optimizes the ability to detect ecological change using transect-based methods. We present an assessment of the sampling sensitivity of the core methods under different transect lengths, numbers, and sampling intervals to determine: 1) minimum sampling required to describe ecosystem characteristics and detect ecosystem change for each method, 2) the optimal transect length and transect number for measuring all three methods, and 3) efficacy sampling approaches of widely established monitoring and research programs to make recommendations for future monitoring efforts. We use data from the National Wind Erosion Research Network to explore these questions as the network spans the western US, includes 118 measurements over time across 5 dryland biomes which is unique among method research studies. We found that cumulative transect length sampled ( >100 m) is likely the most important feature to maximize, particularly for measuring the spatial distribution of plant canopies. Reducing sampling effort was more likely to indicate no ecosystem change when one occurred. Ecosystem monitoring should be designed with research questions and monitoring objectives in mind, as well as consider inference at site and landscape scale. This study demonstrates site-scale inference is strongly affected by sample design and consequently our understanding of ecological dynamics may be influenced by sampling decisions.