A hierarchical Bayesian model for regionalized seasonal forecasts: Application to low flows in the northeastern United States

Abstract

This study presents a regional, probabilistic framework for seasonal forecasts of extreme low summer flows in the northeastern United States conditioned on antecedent climate and hydrologic conditions. The model is developed to explore three innovations in hierarchical modeling for seasonal forecasting at ungaged sites: (1) predictive climate teleconnections are inferred directly from ocean fields instead of predefined climate indices, (2) a parsimonious modeling structure is introduced to allow climate teleconnections to vary spatially across streamflow gages, and (3) climate teleconnections and antecedent hydrologic conditions are considered jointly for regional forecast development. The proposed model is developed and calibrated in a hierarchical Bayesian framework to pool regional information across sites and enhance regionalization skill. The model is validated in a cross‐validation framework along with five simpler nested formulations to test specific hypotheses embedded in the full model structure. Results indicate that each of the three innovations improve out‐of‐sample summer low‐flow forecasts, with the greatest benefits derived from the spatially heterogeneous effect of climate teleconnections. We conclude with a discussion of possible model improvements from a better representation of antecedent hydrologic conditions at ungaged sites.

Revealing Hidden Climate Indices from the Occurrence of Hydrologic Extremes

“…While this parsimonious model allows considering larger, moderately heterogeneous regions, it is still not applicable at a large spatial scale (national, continental or even global). Indeed, climate effects are unlikely to follow the deterministic pattern used by Ahn et al (2017) at such a large spatial scale. Moreover, a single HCI may not be sufficient to explain the variability of hydrologic extremes.…”

“…The model proposed by Renard and Lall () was based on the assumption that the HCI to be identified has a common effect for all target sites: this restricts its application to small regions where the effect of climate variability can be considered as homogenous. Ahn et al () relaxed this assumption by means of a parsimonious description of the variation of the HCI effect in space. They assumed that the HCI effect is maximal at some unknown location of the study area, and then decreases as a deterministic function of the distance to this location.…”

Statistical Models to Predict and Assess Spatial and Temporal Low‐Flow Variability in New England Rivers and Streams

“…The NAO has been shown to be correlated with New England peak flows in winter and early spring, which in turn influences baseflow at the end of the growing season (Steinschneider and Brown , ; Coleman and Budikova ; Ahn et al. ).…”

A Multivariate Approach to Generate Synthetic Short‐To‐Medium Range Hydro‐Meteorological Forecasts Across Locations, Variables, and Lead Times

“…Synthetic forecasts can enable this type of procedure in the forecast informed setting, aiding in robust policy selection frameworks that emulate model selection procedures in machine learning (Shalev-Shwartz & Ben-David, 2013). Synthetic forecasts can be beneficial at all timescales, including seasonal forecasts that are used to inform water supply based decisions (Ahn et al, 2017;Anghileri et al, 2016;Denaro et al, 2017;Giuliani et al, 2019;Turner et al, 2017;Yuan et al, 2015), and shorter range forecasts that are used for hazard management (Valeriano et al, 2010;You & Cai, 2008). However, uncertainty in seasonal forecasts surpasses that of short-to-medium range forecasts (Giuliani et al, 2019).…”