Correlation of consumer grade hydroacoustic signature to submersed plant biomass

Abstract

Invasive macrophytes, such as non-native Hydrilla verticillata, negatively affect lentic systems of the Southeastern United States by impeding recreational activities and power generation as well as disrupting intrinsic ecological function. Expenditures associated with aquatic weed management include costs accompanied with monitoring, mapping, and implementing control measures. Traditional biomass sampling techniques have been widely utilized to assess the extent and abundance of submersed aquatic vegetation (SAV) incursions, but often require significant labor inputs which limits repeatability, the scale of sampling, and the rapidness of processing. Advances in consumer available hydroacoustic technology and data post-processing platforms offer the opportunity to estimate SAV biomass at scale with reduced labor and economic requirements. Research was conducted at two North Carolina reservoirs to compare acoustically derived cloud-based biovolume estimations from an over-the-counter echosounder, to in situ hydrilla biomass measurements. Temporal patterns, spatial developments, and hydrilla biomass prediction models are presented. Biomass and biovolume measurements were positively correlated at both the Shearon Harris and Roanoke Rapids study locations. The most robust predictive equation employed generalized additive models (GAMs) from the Shearon Harris dataset which, described environmental parameters with the lowest error and greatest agreement compared to other verified models. Each biovolume to biomass relationship supported the initial hypothesis that as biovolume increases, SAV biomass increases in a positive, non-linear trend. Implications from this study may prove useful for comparing seasonal growth patterns, littoral occupancy, and herbicide treatment effects on a spatiotemporal level.