Deep Learning for Simulating Harmful Algal Blooms Using Ocean Numerical Model

Abstract

In several countries, the public health and fishery industries have suffered from harmful algal blooms (HABs) that have escalated to become a global issue. Though computational modeling offers an effective means to understand and mitigate the adverse effects of HABs, it is challenging to design models that adequately reflect the complexity of HAB dynamics. This paper presents a method involving the application of deep learning to an ocean model for simulating blooms of Alexandrium catenella. The classification and regression convolutional neural network (CNN) models are used for simulating the blooms. The classification CNN determines the bloom initiation while the regression CNN estimates the bloom density. GoogleNet and Resnet 101 are identified as the best structures for the classification and regression CNNs, respectively. The corresponding accuracy and root means square error values are determined as 96.8% and 1.20 [log(cells L–1)], respectively. The results obtained in this study reveal the simulated distribution to follow the Alexandrium catenella bloom. Moreover, Grad-CAM identifies that the salinity and temperature contributed to the initiation of the bloom whereas NH4-N influenced the growth of the bloom.

Spatiotemporal estimation of groundwater and surface water conditions by integrating deep learning and physics-based watershed models

“…These findings highlight the significance of utilizing spatiotemporal input data measuring 600 m × 600 m (3 × 3 grids) and encompassing the preceding 12 months to estimate the hydrological conditions in our study. In a previous study using a CNN, spatial information of less than 5 × 5 grids and temporal data spanning over 30 d was observed to be suitable for simulating harmful algal blooms within the study site ( Baek et al, 2021 ).…”

Incorporating Microbial Species Interaction in Management of Freshwater Toxic Cyanobacteria: A Systems Science Challenge

“…Considering such ecological complexities will reduce the potential for non-target effects and potentially increase the efficiency of these methods by helping to determine the appropriate combination and sequence for methods targeting the respective cyanobacterial and fungal threats. Furthermore, some of the underlying complexities could potentially be teased out via the use of mesocosm experiments [183] and ecological network models parameterized with field observations [184], as has been achieved to some extent in ocean systems.…”

Modeling- and Simulation-Driven Methodology for the Deployment of an Inland Water Monitoring System

“…These EWSs must be capable of autonomous water analyses: processing informative data through modeling and simulation (M&S) techniques to predict the temporal and spatial dynamics of HACBs. For instance, machine learning models have proven effective for simulating HACBs in water systems [9]. It can be seen from previous works how obtaining knowledge from complex aquatic environment data sources is a challenge that requires extensive software engineering knowledge [10].…”