Temporal design for aquatic organismal sampling across the National Ecological Observatory Network

Abstract

Abstract The National Ecological Observatory Network (NEON) is a continental‐scale research platform designed to assess the impacts of climate change, land‐use change and invasive species on ecosystem structure and function at field sites distributed across 20 ecoclimatic domains (or regions) from Alaska to Puerto Rico. Aquatic sites within the NEON network include 24 streams, 7 lakes and 3 rivers among 19 domains. A significant challenge to this effort is defining standardized methodology for sampling at sites with substantially variable geomorphology and hydrology. The aquatic temporal sampling design provides timing windows for seasonal sampling to best assess organismal diversity and abundance. The need to establish a rule set for temporal sampling was addressed via site‐specific seasonal sampling windows, defined using a suite of environmental variables collected from publicly available meteorological data. Variables integrated into the design include stream flow, growing degree days and riparian phenology. Thresholds for these variables were determined using published literature and used to create three sampling windows for each sample site. Sampling windows target biological community diversity and seasonal changes in abundance, and roughly align with spring, mid‐summer and autumn. NEON‐generated organismal data from 2014 to 2021 were analysed for inter‐ and intra‐annual variability to quantify community‐scale variability among seasonal sampling windows and years at each site. Algae, macroinvertebrate and zooplankton community structure was significantly different within sites across sampling windows, supporting the sampling design of three separate windows per year. Moreover, 93% of sampling windows were completed from 2014 to 2021. An analysis of macroinvertebrate data at 14 sites shows that seasonal β‐diversity represents an important attribute of the stream community, even if interannual trends among sites may differ. We conclude that the temporal sampling design for NEON aquatic biological sampling is justified in having three sampling windows per site and is achievable at most NEON aquatic sites. Although sampling windows have been adjusted at a select number of sites, the original sampling windows are used successfully at the majority of NEON aquatic sites. Sampling windows will be updated with NEON continuous sensor data from each site once sufficient (>3 years) data are available.