Optimized Microcystis Prediction Model Using EFDC-NIER and LH-OAT Method

• Published in: KSCE journal of civil engineering Vol. 27; no. 3; pp. 1066 - 1076
• Main Authors: Ahn, Jung Min, Kim, Jungwook, Kwak, Sunghyun, Kang, Taegu
• Format: Journal Article
• Language: English
• Published: Seoul Korean Society of Civil Engineers 01.03.2023; Springer Nature B.V; 대한토목학회
• Subjects: Algae; Calibration; Civil Engineering; Cyanobacteria; Engineering; Environmental research; Functional groups; Geotechnical Engineering & Applied Earth Sciences; Graphical user interface; Hydraulic Engineering; Hypercubes; Industrial Pollution Prevention; Labour; Mathematical models; Microcystis; Modelling; Parameters; Phytoplankton; Prediction models; Source code; Vertical migration; Vertical migrations; Water quality; 토목공학; EFDC-NIER; Water quality modeling; LH-OAT; Automatic parameter calibration
• ISSN: 1226-7988; 1976-3808
• DOI: 10.1007/s12205-023-1886-y

The National Institute of Environmental Research (NIER) has improved the Environmental Fluid Dynamic Code (EFDC) source code (version 20100328) released by Dynamic Solutions International ( https://www.ds-intl.biz ). Based on this, a new source code, known as EFDC-NIER, can simulate the vertical migration mechanisms of multiple algal species and cyanobacteria. Along with the vertical migration mechanisms of multiple algal species and cyanobacteria, new parameters are added to the EFDC-NIER model. As the phytoplankton functional group (PFG) concept is adopted, a method to predict Microcystis spp. cell count and parameter calibrations is required. The existing parameter calibration processes for water quality modeling are labor-intensive and time consuming, as the calibration depends on each user’s experience, and require repeated manual changes to the parameters of the model. Such manual calibration tends to aggravate the uncertainty in the modeling results. Therefore, this study adopted the Latin-Hypercube One-factor-At-a-Time (LH-OAT) method to construct model sets for parameter calibration. A graphical user interface (GUI) system was also developed for automatic model parameter set configuration, model construction, and modeling, with suggestions for optimal parameters based on the simulation results. The proposed method and tools reported in this study will be useful to determine an optimal parameter interval and values to predict the cell counts of harmful cyanobacteria, such as Microcystis spp.