Automated Baseflow Separation for Changing Catchments

Baseflow represents rain water that infiltrates through the ground and takes longer (weeks/months) to rise and decay in streamflow, compared to days for the surface runoff. A long-standing problem in hydrology is how this should be separated given only a streamflow record. Forty five years ago, I invented the conceptually-confounded Lyne-Hollick baseflow algorithm (LH), out of desperation but now commonly used to separate out baseflow from a streamflow record. LH has been widely used globally and for most rivers in Australia including the Murray-Darling. But many rely upon the bogus-default 0.925 parameter. Here, I reinvent LH into a conceptually more meaningful automatic algorithm (AutoVL) that does not rely upon any guessed-parameters. This new algorithm provides not only the separated baseflow but also tells us how fast it is decaying and how fast surface waters are infiltrating into the ground. Sadly, application of AutoVL to the Harvey River, Dingo Road streamflow tells us that the massive declines in rainfall for Western Australia’s southwest is accompanied by drying catchments infiltrating water quicker to baseflow which is also decaying quicker. A double whammy to declining streamflows and water supplies from climate change. Forty five years ago we knew about the long-term rainfall trend (started since the early 1900's) but now catchments are drying up and forest fire risks are increasing, with threats to habitats, species, ecosystems, and recalcitrant humans.

You can find a pre-print of the article at: https://www.researchgate.net/publication/381094299_Automated_Baseflow_Separation_for_Changing_Catchments