From CAMELS to Caravan: Community datasets for catchment hydrology - Gemma Coxon & Nans Addor

Recording of BHS National Webinar "From CAMELS to Caravan: Community datasets for catchment hydrology over regional to global domains" from Friday 7th October 2022 presented by Gemma Coxon (University of Bristol) and Nans Addor (Fathom).

The last few years have seen the release of several large-sample hydrology (LSH) datasets. These datasets make available an easy-to-use combination of high-quality hydroclimatic time series (atmospheric and river flow data) and socio-biogeophysical attributes (e.g. topography, land cover, soil, geology and level of human interventions) for large samples (typically hundreds) of catchments. These datasets have been widely adopted in academia and by the industry, enabling the systematic exploration of hydrological processes across hydroclimatic regions, the benchmarking of hydrological models, and they have accelerated the emergence of machine learning approaches for hydrology.

In this talk, we will first introduce the CAMELS-GB dataset (Catchment Attributes and MEteorology for Large-sample Studies for Great Britain), which collates a suite of hydro-meteorological timeseries and catchment attributes for 671 catchments across Great Britain, and we will demonstrate how the dataset is being used to shed new light on environmental change. We will then discuss challenges faced by regional LSH datasets, and introduce the Caravan initiative, which brings together these regional LSH datasets to create a global community LSH dataset on the cloud. These datasets and tools constitute a global open source community resource to facilitate data sharing, collaborations and advances in hydrology.

The meeting was hosted by Lucy Barker, BHS Honorary Secretary.

Find out more about BHS: https://www.hydrology.org.uk/
Find out more about Camels-GB: https://doi.org/10.5194/essd-12-2459-...
Download the Camels-GB data: https://catalogue.ceh.ac.uk/documents...
Find out more about Caravan: https://eartharxiv.org/repository/vie...
Find out more about large-sample hydrology: https://doi.org/10.1080/02626667.2019...