Data and Code for "Automated Adaptive Baseflow Separation Across Australian Catchments Using AutoVL"

Data and Code for Automated Adaptive Baseflow Separation Across Australian Catchments Using AutoVL Authors Vincent Lyne, Ratnasingham Srikanthan, Thomas A. McMahon 1. Overview This repository contains the derived data products and code supporting the analysis presented in: Lyne, V., Srikanthan, R., McMahon, T.A. (2026). Automated Adaptive Baseflow Separation Across Australian Catchments Using AutoVL. Water Resources Research. The dataset comprises AutoVL-derived baseflow separation results and associated hydrological statistics for 467 Australian catchments spanning diverse hydroclimatic regimes. 2. Data provenance The underlying streamflow data were obtained from the Bureau of Meteorology Hydrologic Reference Stations (HRS) dataset: https://www.bom.gov.au/water/hrs/#panel=data-download These data are publicly available and maintained by the Bureau of Meteorology.They are accessed via an interactive portal and are not distributed as fixed individual datasets. Accordingly, the present repository contains derived data products only. 3. Repository contents 3.1. Dataset SriTom_merged_corrected.csv A catchment-scale dataset (n = 467) containing: Station identifiers and geographic attributes Record characteristics (start year, end year, record length) Hydrological statistics (mean flow, coefficient of variation, skewness) AutoVL-derived parameters Baseflow indices and flow-partitioning metrics Cluster classifications 3.2. Code AutoVL_single_pass_standalone.R Standalone implementation of the AutoVL algorithm, including: Variable-leak flow separation ARX/ARMA-based parameter estimation Iterative parameter convergence Diagnostic outputs (including BFI) The implementation uses base R only and reproduces the methodology described in the manuscript. 4. Dataset structure Each row represents one catchment. Key variables include: Identification StationID, StationName Region, StationCode Latitude, Longitude, CatchmentArea Record characteristics StartYear, EndYear, RecordLength Hydrological statistics MeanFlow Cv (coefficient of variation) Skewness Flow components TotalFlow FastFlow TransientFlow BaseFlow (defined as the slow-flow component) AutoVL parameters InitialAlphaF AlphaF (fast-flow AR parameter) AlphaB (baseflow AR parameter) Beta (moving-average term) Derived indices BFI_VL (AutoVL Baseflow Index) BFI_LH (Lyne–Hollick Baseflow Index) 5. Definitions In this study, baseflow is defined explicitly as the slow-flow component estimated by AutoVL.Transient or intermediate flow is treated as part of the fast-flow domain. 6. Reproducibility Results can be reproduced by: Downloading streamflow data from the BOM HRS portal Applying the provided AutoVL code to each station Aggregating outputs to derive summary statistics and indices 7. Licensing Data: CC-BY 4.0 Code: MIT License (or CC-BY 4.0 if preferred for consistency) 8. Notes This repository contains derived data only Original streamflow data must be obtained from the Bureau of Meteorology The dataset is designed to support reproducibility and further analysis