Talk on “Large-Sample Hydrology to Improve Prediction in Ungauged Basins using Information Theory Techniques”
- The venue was CWPRS Pune
- Talk of the date is
20 December 2024
I talked about
- Prediction in Ungauged Basins (PUB): PUB addresses the challenge of estimating hydrological behaviors, such as streamflow and runoff, in river basins where direct measurements are unavailable.
- Trading Space for Time (TSFT) Approach: TSFT is a novel method utilizing data from multiple catchments to mitigate the impact of low parameter or discharge data availability.
- Large-Sample Hydrology (LSH): LSH leverages extensive datasets across spatial and temporal scales to improve prediction models for ungauged basins.
- Recent datasets like CAMELS and CARAVAN provide critical data for hypothesis testing and advancing regionalization techniques.
Key points of talk
- Proposed the framework to group catchments to imporve the TSFT for better prediction.
- We use Purity, NMI methods to compare the clusterings.
#LSH #PUB #TSFT #Hydro2024