Climate change rapidly warms and acidifies Australian estuaries

Abstract

Climate change is impacting ecosystems worldwide. Estuaries are diverse and important aquatic ecosystems; and yet until now we have lacked information on the response of estuaries to climate change. Here we present data from a twelve-year monitoring program, involving 6200 observations of 166 estuaries along >1100 kilometres of the Australian coastline encompassing all estuary morphologies. Estuary temperatures increased by 2.16 °C on average over 12 years, at a rate of 0.2 °C year^-1, with waters acidifying at a rate of 0.09 pH units and freshening at 0.086 PSU year^-1. The response of estuaries to climate change is dependent on their morphology. Lagoons and rivers are warming and acidifying at the fastest rate because of shallow average depths and limited oceanic exchange. The changes measured are an order of magnitude faster than predicted by global ocean and atmospheric models, indicating that existing global models may not be useful to predict change in estuaries.

Fig. 1. Observed change in estuary temperature, pH and salinity since 2007.

Summer temperature pH and salinity measured over the 12- (temperature and salinity) and 6-year (pH) estuary monitoring programme from December 2007 to January 2019; grey dots indicate each data point, darker dots indicate multiple data on that point. Estuaries are divided into the five estuary types; each estuary is represented by a satellite image of an estuary that is typical of the type, with the body of the estuary outlined in pink. White scale bar indicates 1000 m. Sattelite images of estuaries are sourced from Google Earth (map data: SIO, NOAA, US Navy, NGA and GEBCO). All estuaries are represented on a map of Australia showing the sample sites as black dots. The map of Australia is sourced from Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL.

Fig. 2. Relative importance of variables in predicting estuary temperature, pH and salinity.

Variable importance plots (% MSE as an indicator of importance^,) for a temperature, b pH and c salinity generated from random forest models. Predictor variable categories are colour coded.

Fig. 3. Estuary temperature modelled using average depth and flushing time as predictors.

Temperature as modelled by RF over a range of average depths and flushing times. Each estuary type is shown using its mean average depth and flushing time to indicate their general location in relation to these predictors.