The influence of climate on water chemistry states and dynamics in rivers across Australia

Abstract

For effective water quality management and policy development, spatial variability in the mean concentrations and dynamics of riverine water quality needs to be understood. Using water chemistry (calcium, electrical conductivity, nitrate‐nitrite, soluble reactive phosphorus, total nitrogen, total phosphorus and total suspended solids) data for up to 578 locations across the Australian continent, we assessed the impact of climate zones (arid, Mediterranean, temperate, subtropical, tropical) on (i) inter‐annual mean concentration and (ii) water chemistry dynamics as represented by constituent export regimes (ratio of the coefficients of variation of concentration and discharge) and export patterns (slope of the concentration‐discharge relationship). We found that inter‐annual mean concentrations vary significantly by climate zones and that spatial variability in water chemistry generally exceeds temporal variability. However, export regimes and patterns are generally consistent across climate zones. This suggests that intrinsic properties of individual constituents rather than catchment properties determine export regimes and patterns. The spatially consistent water chemistry dynamics highlights the potential to predict riverine water quality across the Australian continent, which can support national riverine water quality management and policy development.

Synthesizing the impacts of baseflow contribution on concentration–discharge (<i>C</i>–<i>Q</i>) relationships across Australia using a Bayesian hierarchical model

“…The current knowledge gap in understanding catchment export regimes for regions other than Europe and North America was partially addressed in Lintern et al (2021), which focused on differences in water quality status and the C-Q relationships across different climate zones in the Australian continent. One remaining question that Lintern et al (2021) highlighted is our lack of understanding of the substantial variations in C-Q relationships within each climate zone.…”

“…The model is a powerful approach to capture water quality variability across catchments of varying conditions and record lengths, which is the case for Australian water quality data (Guo et al, 2019(Guo et al, , 2020Liu et al, 2021). We use a subset of the grand dataset that Lintern et al (2021) used, which enables us to focus on representative catchments with water quality records captured under a wider range of flow conditions. As such, by analysing the impacts of baseflow contributions on C-Q relationships, this study will (i) explain the variations in C-Q relationships within individual climate zones and (ii) broaden the existing knowledge of how baseflow contribution impacts C-Q relationships to a wider range of climate conditions and thus infer key constituent transport pathways in different climate zones.…”

“…As such, by analysing the impacts of baseflow contributions on C-Q relationships, this study will (i) explain the variations in C-Q relationships within individual climate zones and (ii) broaden the existing knowledge of how baseflow contribution impacts C-Q relationships to a wider range of climate conditions and thus infer key constituent transport pathways in different climate zones. All available water quality data were obtained from all seven state agencies in late 2019 and collated into a single national-scale database (see more details in Lintern et al, 2021). Quality control of the data was performed using quality codes, flags and detection limits provided by individual state agencies (as detailed in Table S1 in the Supplement).…”

“…This study focuses on six water quality variables: total suspended solids (TSS), total phosphorus (TP), soluble reactive phosphorus (SRP), total nitrogen (TN), the sum of nitrate and nitrite (NO x ) and electrical conductivity (EC). These six variables have been included because they are of key concern for Australian riverine water quality and are well monitored across Australia both spatially and temporally, as illustrated in Lintern et al (2021).…”

“…1. There are over 50 pairs of corresponding concentration and flow data points; this ensures that the C-Q relationships observed are more robust against outliers (Lintern et al, 2021).…”