A comparison of two methods for sampling biomass of aquatic plants

We evaluated the performance of a boat-based vertical rake method for sampling aquatic plant biomass to determine whether it was a suitable alternative to diver quadrat sampling. Boat-based methods are generally considered to be safer and require less collection time than diver quadrat sampling, but they may be less accurate and precise. We compared three aspects of the two sampling methods: detection rates for individual taxa, comparability of biomass estimates, and precision of biomass estimates. Detection rates for the two sampling methods were comparable for all evaluated plant taxa. Similarly, rake and quadrat biomass estimates were comparable in magnitude and precision for most individual plant taxa. However, rake estimates were significantly higher than quadrat estimates for coontail (Ceratophyllum demersum L.) and flatstemmed pondweed (Potamogeton zosteriformis Fern.). Linear regression of rake yields against quadrat yields (log-transformed) showed significant positive relationships for all evaluated taxa, and slopes that did not differ significantly from 1 for most taxa. However, rake estimates of total biomass (all taxa combined) were significanuy higher and less precise than quadrat estimates, particularly for samples collected from areas with dense plant growth (>200 dry g m^-2). Although our results suggested that the rake method was not a perfect surrogate for the quadrat method, the increased sample collection afforded by the vertical rake mediod may offset the method's inherentiy lower precision. Consequently, we concluded that the vertical-rake method was a suitable alternative to the diver quadrat method for collection of aquatic plant biomass, particularly in large-scale studies requiring high sampling intensity over expansive areas.