A comparative assessment of Australia's Lower Lakes water quality under extreme drought and post-drought conditions using multivariate statistical techniques

Abstract

Abstract Drought generally results in a decline to freshwater quality, but the spatial nature of these impacts and recovery processes in large lakes systems remain poorly understood. This study applied multiple statistical methods such as cluster analysis (CA), discriminant analysis (DA), principal component analysis (PCA) and factor analysis (FA), to assess spatial and temporal variations of water quality in the Lower Lakes (Australia) during drought (April 2008–September 2010) and post-drought (October 2010–October 2013) periods. The comprehensive analysis of water quality from 22 locations and including 22 key parameters showed that Lower Lakes were eutrophic in both drought and post-drought periods with higher nutrient and algae concentrations than guideline levels for aquatic ecosystem. The Lower Lakes were identified three distinct spatial zones, i.e., (1) low eutrophication for the southeast of Lower Lakes (SE), (2) moderate eutrophication for northeast of Lower Lakes (NE), and (3) high eutrophication for northwest of Lower Lakes (NW) in the drought as well as low eutrophication for NW and high eutrophication for SE in the post-drought. DA allowed a better reduction in the dimensionality of the large dataset during post-drought than during the drought period with better results for spatial analysis rather than for temporal analysis regardless of hydrological periods. PCA/FA reflected three major factors of mineral dissolution, erosion and anthropogenic sources accounting for water constituents. Our results demonstrate the powerful utility of multivariate statistical techniques for revealing the persistent and spatially complex nature of drought-induced impacts on lake water quality and highlight that optimal utilization of water resources in the upper catchment of Lower Lakes are urgently needed for the sustainable lake ecosystems.