FW: SAVE THE DATE || Seminar on using ML for developing hydrologic models
Juras Roman
Vážení kolegové,
Posílám odkaz na zajímavý online seminář.
S pozdravem
Ing. Roman Juras, Ph.D.
Katedra vodního hospodářství a environmentálního modelování
Fakulta životního prostředí
Česká zemědělská univerzita v Praze
From: Francesco Avanzi <francesc...@cimafoundation.org>
Sent: Wednesday, August 4, 2021 7:46 PM
To: Simone Gabellani <simone.g...@cimafoundation.org>
Subject: SAVE THE DATE || Seminar on using ML for developing hydrologic models
Dear colleagues,
I am happy to share with you this invitation to a webinar that we are hosting at CIMA Research Foundation on optimizing spatial distribution of watershed-scale hydrologic models using Gaussian Mixture Models (speaker: Dr. Tessa Maurer, UC Berkeley & Blue Forest Conservation).
You can find information and the registration link below.
Please do not hesitate to forward to anyone interested in your research group or network!
We are particularly interested in having early career researchers and students attending this seminar.
Best,
Francesco
-------------------------------------------------------------
Representative sampling is not only for pollsters: Optimizing spatial distribution of watershed-scale hydrologic models using Gaussian Mixture Models
California has one of the most intricate and complex water delivery systems in the world, with a network of reservoirs, aqueducts, and groundwater pumps that deliver water from the mountain headwaters in the northern and eastern portions of the state to population centers and agricultural land in the west and south over distances of up to 900+ km. This geographic imbalance between water supply and water demand, coupled with a Mediterranean climate delivering nearly all annual precipitation during winter, has led water managers across the state to develop seasonal-forecasting tools to predict freshwater availability and so manage allocations to end users. Recent California droughts have demonstrated that precipitation variability and high temperatures have the potential to make these forecasting methods less accurate by jeopardizing water-supply generation in snow-dominated regions, which provide up to two-thirds of the state’s consumptive water supply. As a result, more accurate, spatially explicit hydrologic models are paramount to support water resources management, especially under a warming and changing climate.
Here, I discuss the use of a statistical learning algorithm, Gaussian Mixture Models (GMMs), for spatial distribution of hydrologic models. GMMs objectively select the set of modeling locations that best represent the distribution of watershed features relevant to the hydrologic cycle, thus favoring the development of objective, easy-to-deploy hydrologic models across the complex and highly variable topography of montane regions. The GMM method was applied in two hydrologically distinct headwater catchments of the Sierra Nevada mountains and met or exceeded the performance of both lumped models and models based on hydrologic response units for multiple metrics across the water balance at a fraction of the time cost. I show how the GMM method allows for more robust, objective, and repeatable models, which are critical for advancing hydrologic research and operational decision making.
Registration link: https://zoom.us/meeting/register/tJYodOihqD4pHtY-APLchQ3auLCGBBC3ESj0
--
----------------------------------------------------------------
Francesco Avanzi, PhD
CIMA Research Foundation
Via Armando Magliotto, 2
17100 Savona (Italy)
https://scholar.google.it/citations?user=Zj-MnugAAAAJ&hl=it
----------------------------------------------------------------
