Great Lakes Runoff Intercomparison Project Phase 3: Lake Erie (GRIP-E).

Juliane Mai,Amin Haghnegahdar,Luis Samaniego,Nicolas Gasset,Emily A Bradley,Jimmy Lin,James R Craig,Shervan Gharari,Meghan Mcleod,Vincent Fortin,Lauren M Fry,Rohini Kumar,Étienne Gaborit,Sabine Attinger,Xiaojing Ni,Tricia A Stadnyk,Hervé Awoye,Sungwook Wi,Saman Razavi,Tirthankar Roy,Alain Pietroniro,Frank Seglenieks,Yongping Yuan,Prasad Daggupati,Martin Gauch,Bryan A Tolson,Mohamed Elshamy ,Helen C Shen ,Daniel Princz ,N B Basu ,André Guy Tranquille Temgoua ,Narayan Kumar Shrestha ,Oldřich Rakovec ,Timothy Hunter

doi:10.1061/(asce)he.1943-5584.0002097

Abstract

Hydrologic model intercomparison studies help to evaluate the agility of models to simulate variables such as streamflow, evaporation, and soil moisture. This study is the third in a sequence of the Great Lakes Runoff Intercomparison Projects. The densely populated Lake Erie watershed studied here is an important international lake that has experienced recent flooding and shoreline erosion alongside excessive nutrient loads that have contributed to lake eutrophication. Understanding the sources and pathways of flows is critical to solve the complex issues facing this watershed. Seventeen hydrologic and land-surface models of different complexity are set up over this domain using the same meteorological forcings, and their simulated streamflows at 46 calibration and seven independent validation stations are compared. Results show that: (1) the good performance of Machine Learning models during calibration decreases significantly in validation due to the limited amount of training data; (2) models calibrated at individual stations perform equally well in validation; and (3) most distributed models calibrated over the entire domain have problems in simulating urban areas but outperform the other models in validation.

Highlights

Hydrologic modeling efforts in the Great Lakes region of North America are a fundamental component for the understanding and management of water resources impacting more than 30 million people in Canada and the United States (USEPA, n.d.)
The reason for this is that the stations for objective 1 are chosen purely based on the regulation type as “natural” and “reference” provided by Water Survey Canada (WSC) and USGS
The stations initially selected as low human impact should have been manually double-checked to ensure that they are not affected by catchment management

Summary

Introduction

Hydrologic modeling efforts in the Great Lakes region of North America are a fundamental component for the understanding and management of water resources impacting more than 30 million people in Canada and the United States (USEPA, n.d.). These models are used for operational forecasting of Great Lakes water levels or for evaluating the impacts of climate change. At the smaller or regional scales, these models can be used for reservoir management, water supply, floodplain mapping, and a multitude of other applications. Multimodel development and careful model intercomparisons can be a critical part of improving simulation or forecasting skills, as well as capturing the uncertainty in any future climate assessments (Wada et al 2013; Warszawski et al 2014; McSweeney and Jones 2016; Rosenzweig et al 2017; Frieler et al 2017; Huang et al 2017). The following review of past hydrologic model intercomparison studies is focused mainly on studies that (1) compare models in watersheds of at least hundreds of km with a sample size of at least 10 watersheds; and (2) include the participation of multiple independent modeling groups

Objectives

Methods

Results

Conclusion