A feasible method to assess inaccuracy caused by patchiness in water quality monitoring

Abstract

Patchiness is a typical property of water quality in lakes. However, in conventional water quality monitoring, patchiness is usually too expensive to take into account, due to the high number of required samples. This study examines a feasible methodology developed for estimating the representativeness of discrete chlorophyll a measurements. Four spatially extensive data sets were collected from the Enonselkä basin of Lake Vesijärvi in Southern Finland, using a flow trough system with a fluorometer in a moving boat. Data sets were used to estimate (1) the spatial representativeness of discrete sampling; (2) the effect of varying sample size on the detected mean chlorophyll a concentration and on the observed proportion of variance. Spatial representativeness was assessed using semivariogram analysis. Results indicate that the spatial representativeness of discrete sampling can remain undesirably low. Furthermore, in monitoring programs involving just one or only a few samples, there is a significant risk of obtaining a false estimate for the mean value and variance of chlorophyll a concentration over the whole monitoring area.

Spatial Heterogeneity of Cyanobacteria and Diatoms in a Thermally Stratified Canyon‐Shaped Reservoir

“…In agreement, FEE (1976) alerted on the necessity of taking into consideration spatial heterogeneity in phytoplankton dynamic studies and pointed out that estimations www.revhydro.com of total Chl-a derived from a single station survey in the deepest part of the lake often presents errors of 25-45%. Nowadays, most experts believe that discrete sampling station monitoring programmes can only very poorly describe the whole lake water quality (ANT-TILA et al, 2005) as they do not consider spatial heterogeneity (VOS et al, 2003) and many studies have concluded that patchiness represents a major challenge for monitoring program design (HORPPILA et al, 1998;KALLIO et al, 2001;ANTTILA et al, 2008). In this sense, it has been stressed that monitoring regimes must be calibrated to the specific spatial dynamics of the lake under investigation (DUTILLEUL, 1993;HEDGER et al, 2001).…”

“…Accordingly, MORENO-OSTOS et al (2008a) suggested that a complete set of spatially-distributed sampling stations should be implemented in El Gergal monitoring routines if a reliable estimation of phytoplankton biomass and community composition were needed. Nevertheless, aquatic ecosystem managers often claim the adoption of such an extensive survey procedure into their regular water quality monitoring routines could be expensive in terms of time, staff and economic resources (ANTTILA et al, 2008).…”

“…Spectrofluorometer with a flow-through system in a moving boat was chosen as a data source due to its ability to cover large areas with extensive number of records (GEORGE, 1976;ANTTILA et al, 2008). LINDFORDS et al (2005) stated that this methodology prevents the misinterpretation due to the patchiness in ground measurements and makes measurements more representative.…”

“…Even though the rather constant wind regime in the reservoir makes this prediction reliable, our results also suggest that the occurrence of exceptional and infrequent wind events should be considered as an error source, as the mean cyanobacteria Chl-a concentration would occur on an area much closer to the downwind shore (around 0.1 L). Although more complex methods based in spatial statistics have been proposed in the literature (i.e., ANTTILA et al, 2008) the simplicity of this feasible low-cost empirical method encourages its use for dynamic cyanobacteria survey design for canyon-shaped reservoirs.…”

Ultraviolet Fluorescence LiDAR (UFL) as a Measurement Tool for Water Quality Parameters in Turbid Lake Conditions

“…As in marine settings, fluorescence spectroscopy has provided invaluable insight into a range of processes and phenomena in lakes, including the characterization of phytoplankton, pollutants and dissolved organic matter [18][19][20][21][22][23]. Combining the technologies of LiDAR and fluorescence spectroscopy allows the potential to detect fluorescence features of lake water in situ and non-intrusively, while adding a cohesive spatial component.…”