A data-driven modeling approach for simulating algal blooms in the tidal freshwater of James River in response to riverine nutrient loading

Abstract

Algal blooms often occur in the tidal freshwater (TF) of the James River estuary, a tributary of the Chesapeake Bay. The timing of algal blooms correlates highly to a summer low-flow period when residence time is long and nutrients are available. Because of complex interactions 5 between physical transport and algal dynamics, it is challenging to predict interannual variations 6 of bloom correctly using a complex eutrophication model without having a high-resolution model grid to resolve complex geometry and an accurate estimate of nutrient loading to drive the model. In this study, an approach using long-term observational data (from 1990-2013) and the Support vector machine (LS-SVM) for simulating algal blooms was applied. The Empirical Orthogonal Function was used to reduce the data dimension that enables the algal bloom dynamics for the entire TF to be modeled by one model. The model results indicate that the datadriven model is capable of simulating interannual algal blooms with good predictive skills and is capable of forecasting algal blooms responding to the change of nutrient loadings and environmental conditions. This study provides a link between a conceptual model and a dynamic model, and demonstrates that the data-driven model is a good approach for simulating algal blooms in this complex environment of the James River. The method is very efficient and can be applied to other estuaries as well.

Drivers of the spatial phytoplankton gradient in estuarine–coastal systems: generic implications of a case study in a Dutch tidal bay

“…The CMZ is a combined consequence of the optimal light conditions and abundant terrestrial nutrients, and the CMZ location and coverage shift with river discharge and weather (Fisher et al, 1988;Miller and Harding, 2007). In some other estuaries with a CMZ (e.g., the Neuse-Pamlico Estuary and York River), owing to a narrow river channel and high discharge, the flushing rate in the upper estuary can be faster than the phytoplankton turnover rate, which, rather than light, limits phytoplankton accumulation (Sin et al, 1999;Valdes-Weaver et al, 2006). In these systems, the CMZ is always in wider reaches with sufficiently long RTs (Valdes-Weaver et al, 2006).…”

“…Under high river discharge, phytoplankton growth can be promoted by increasing nutrient input, whereas advective loss and high riverine SPM loading may inhibit algal enrichment (Lancelot and Muylaert, 2011;Shen et al, 2019). In tide-dominated systems, tides can resuspend SPM, negatively impacting phytoplankton, and concurrently transport regenerated nutrients into the water column, or they can drive upwelling-induced algal blooms from the coastal ocean into estuaries (Sin et al, 1999;Roegner et al, 2002). Nitrate can support more phytoplankton biomass in microtidal estuaries than in macrotidal estuaries (Monbet, 1992).…”

A Machine‐Learning‐Based Model for Water Quality in Coastal Waters, Taking Dissolved Oxygen and Hypoxia in Chesapeake Bay as an Example

“…Since the model uses external forcings as used by a traditional 3‐D water quality model, the data‐driven approach can be used to predict water quality conditions under future climate scenarios in response to change in environmental forcing conditions, such as changes in temperature, wind, and river flow. The nonlinear influence of forcing conditions on water quality is implicitly inherited inside the data‐driven model (Shen et al, 2019). However, it should be noted that such predictions may only be possible when the training data set has included the variations of specific forcing and that the trained model has correctly resolved the response of target variable to the given forcing.…”

“…With the quickly accumulated observational data and latest advances in machine‐learning techniques, the data‐driven model is a promising approach with high efficiency for water quality modeling and environmental management in the near future. In fact, there is increasing interest in using machine‐learning techniques for water quality simulations (e.g., Ross & Stock, 2019; Shen et al, 2019). There are, however, still some questions that remain to be further explored.…”

Drivers of the spatial phytoplankton gradient in estuarine-coastal systems: generic implications of a case study in a Dutch tidal bay

“…More often, it is the delicate balance of multiple factors that determine phytoplankton gradients. Under high river discharge, phytoplankton growth can be promoted by increasing nutrient input, whereas advective loss and high riverine SPM loading may inhibit algal enrichment (Lancelot and Muylaert, 2011;Shen et al, 2019). In tide-dominated systems, tides can resuspend SPM, negatively impacting phytoplankton, while at the same time bringing regenerated nutrients into the water column, or drive upwelling-induced algal blooms from the coastal ocean into estuaries (Sin et al, 1999;Roegner et al, 2002).…”