| Abstract [eng] |
Satellite data, with its potential for frequent and comprehensive water body monitoring, remains underutilized in water quality assessment due to the lack of a unified methodology. This dissertation introduces a methodology focusing on chlorophyll α concentration extraction from optical satellite data, aiming to enhance the utilization of satellite data for water quality assessment in Lithuania. The methodology involves data processing and model construction to derive chlorophyll α concentration, a key indicator of algal blooms, from satellite data. The dissertation evaluates the uncertainty of atmospheric correction algorithms pertaining to chlorophyll α concentration retrieval. Furthermore, it introduces a biophysical classification of lakes and develops models for determining chlorophyll α concentration from Sentinel-2 data, utilizing data from Lithuanian lakes larger than 50 hectares and random forest machine learning algorithm. The study applies the developed biophysical classification and chlorophyll α concentration models to 357 Lithuanian lakes and ponds spanning the period from 2015 to 2021. The analysis encompasses seasonal and interannual variations, spatial distribution of chlorophyll α concentration as well as identification of potentially problematic water bodies based on their chlorophyll α concentration. This proposed methodology serves as a valuable tool for assessing the ecological status of mid-latitude lakes. |
