Predicting in-lake responses to short and long-term changes using lake physical models

Moore, Tadhg Nolan (2020) Predicting in-lake responses to short and long-term changes using lake physical models. Doctoral thesis, Dundalk Institute of Technology.

PDF (Final thesis) - Accepted Version Download (10MB)

Abstract

Lakes and reservoirs are under increasing pressure from urbanisation, agricultural intensification, and directional climate change, including an increasing occurrence of extreme climatic events. These pressures can reduce water quality by promoting the occurrence of nuisance algal blooms and higher levels of dissolved organic carbon (DOC), two issues that can cause substantial problems for water treatment, aquatic ecology and recreational users. Thus, there is a need to develop a modelling framework that is flexible, adaptable and provides information that can be utilised to mitigate potential risks to lakes and reservoirs. This thesis describes three specific pieces of work which in combination, further the use of hydrodynamic models for adaptive management. Firstly, the suitability of different meteorological datasets for forcing one-dimensional hydrodynamic models and accurately simulating water temperatures was examined. The European Centre for Medium-Range Weather Forecasts produces freely available gridded meteorological datasets: ERA-Interim, ERA5 and EWEMBI. Lake temperature simulations produced using these three datasets were compared to those based on local meteorological data. Simulations with ERA5 and ERA-Interim simulated water temperatures to a similar degree of accuracy as those forced with local measured data. This highlighted how gridded meteorological datasets can be used to simulate lake thermodynamics in areas where there is no locally measured meteorological data. Secondly, the improvement in short-term model performance when assimilating observed water temperature profile data into model simulations was assessed. Single profiles were inserted into simulations for three lakes that reflected potential monitoring programmes of different temporal frequencies. These monitoring data were compiled by subsetting high frequency data from the sites. Assimilating measured temperature profiles of up to one month prior to the forecast, greatly reduced forecast error. This will allow for short-term forecasting frameworks to be developed for low-frequency monitoring programmes. In the last results section, the effects of different future climate change scenarios on water temperature for a global set of lakes were characterised, using an ensemble of lake models forced with an ensemble of General Circulation Models. The responses in lake temperature and in functional characteristics such as the strength and length of stratification were shown to be highly variable across 46 lakes of varying morphometries. Comprehensively, there was an unequivocal warming of lake water temperature throughout the water column and an extension of the duration of stratification. Such increases in water temperature, heighten the risk of anoxia and the occurrence of algal blooms which are water quality issues which can be actively managed. Overall, this study has found that lake forecasting frameworks (short and long term) can be setup using open access software and data, for sites with low-frequency monitoring data, forced with freely available meteorological data and produce high quality forecasts. These finding will be of increasing importance as we seek to simulate our freshwater ecosystems in a rapidly changing climate to aid in their management.

Item Type:	Thesis (Doctoral)
Subjects:	Science > Biology Science > Physics
Research Centres:	UNSPECIFIED
Depositing User:	Eleanor Jennings
Date Deposited:	18 Nov 2020 18:21
Last Modified:	18 Nov 2020 18:21
License:	Creative Commons: Attribution-Noncommercial-Share Alike 4.0
URI:	https://eprints.dkit.ie/id/eprint/729

Actions (login required)

View Item

Downloads