A Subseasonal Regime Approach for Assessing Intra-annual Variability of Evapotranspiration and Application to the Upper Colorado River Basin

Evapotranspiration (ET) is strongly influenced by gradual climate change and fluctuations in meteorological conditions, such as earlier snowmelt and occurrence of droughts. While numerous studies have investigated how climate change influences the inter-annual variability of ET, very few studies focused on quantifying how subseasonal events control the intra-variability of ET. In this study, we developed the concept of subseasonal regimes, whose timing and duration are determined statistically using Hidden Markov Models (HMM) based on meteorological conditions. We tested the value of subseasonal regimes for quantitatively characterizing the variability of seasonal and subseasonal events, including the onset of snow accumulation, snowmelt, growing season, monsoon, and defoliation. We examined how ET varied as a function of the timing of these events within a year and across six watersheds in the region. Variability of annual ET across these six sites is much less significant than the variability in hydroclimate attributes at the sites. Subseasonal ET, defined as the total ET during a given subseasonal regime, provides a measure of intra-annual variability of ET. Our study suggests that snowmelt and monsoon timing influence regime transitions and duration, such as earlier snowmelt can increase springtime ET rapidly but can trigger long-lasting fore-summer drought conditions that lead to decrease subseasonal ET. Overall, our approach provides an enhanced statistically based framework for quantifying how the timing of subseasonal-event transitions influence ET variability. The improved understanding of subseasonal ET variability is important for predicting the future impact of climate change on water resources from the Upper Colorado River Basin regions.

On the similarity of hillslope hydrologic function: a process-based approach

Hillslope similarity is an active topic in hydrology because of its importance to improve our understanding of hydrologic processes and enable comparisons and paired studies. In this study, we propose a holistic bottom-up hillslope similarity classification based on a region’s integrative hydrodynamic response quantified by the seasonal changes in groundwater levels. The main advantage of the proposed classification is its ability to describe recharge and discharge processes. We test the performance of the proposed classification by comparing it to seven other common hillslope similarity classifications. These include simple classifications based on the aridity index, topographic wetness index, elevation, land cover, and more sophisticated machine-learning classifications that jointly integrate all these data. We assess the ability of these classifications to identify and categorize hillslopes with similar static characteristics, hydroclimatic behaviors, land surface processes, and subsurface dynamics in a mountainous watershed, the East River, located in the headwaters of the Upper Colorado River Basin. The proposed classification is robust as it reasonably identifies and categorizes hillslopes with similar elevation, land cover, hydroclimate, land surface processes, and subsurface hydrodynamics (and hence hillslopes with similar hydrologic function). In general, the other approaches are good in identifying similarity in a single characteristic, which is usually close to the selected variable. We further demonstrate the robustness of the proposed classification by testing its ability to predict hillslope responses to wet and dry hydrologic conditions, of which it performs well when based on average conditions.

Causal effect of the tropical Pacific sea surface temperature on the Upper Colorado River Basin spring precipitation

The Colorado River is one of the most important rivers in the southwestern U.S., with ~ 90% of the total flow originating from the Upper Colorado River Basin (UCRB). The UCRB April–July streamflow is well-correlated to the UCRB spring precipitation. It is known that the UCRB precipitation is linked to an El Niño-like sea surface temperature (SST) pattern, but the causal effect of the tropical Pacific SST on the UCRB spring precipitation is still uncertain. Here, we apply a Granger causality approach to understand the causal effect of the tropical Pacific averaged SST in previous three seasons (winter, fall, and summer) on the UCRB averaged precipitation in spring in observations and two climate models. In observations, only the winter SST has Granger causal effect (with p-value ~ 0.05) on spring precipitation, while historical simulations of the two climate models overestimate the causal effect for winter and fall (with p-value < 0.01 and < 0.05, respectively) due to model biases. Moreover, future projections of the two climate models show divergent causal effects, especially for the scenario with high anthropogenic emissions. The divergent projections indicate that (1) there are large uncertainties in model projections of the causal effect of the tropical Pacific SST on UCRB spring precipitation and (2) it is uncertain whether climate models can reliably capture changes in such causality. These uncertainties may result in large uncertainties in seasonal forecasts of the UCRB hydroclimate under global climate change.

Long-Lead Seasonal Prediction of Streamflow over the Upper Colorado River Basin: The Role of the Pacific Sea Surface Temperature and Beyond

We have developed two statistical models for extended seasonal predictions of the Upper Colorado River Basin (UCRB) natural streamflow during April–July: a stepwise linear regression (reduced to a simple regression with one predictor) and a neural network model. Monthly, basin-averaged soil moisture, snow water equivalent (SWE), precipitation, and the Pacific sea surface temperature (SST) are selected as potential predictors. Pacific SST Predictors (PSPs) are derived from a dipole pattern over the Pacific (30°S–65°N) that is correlated with the lagging streamflow. For both models, the correlation between the hindcasted and observed streamflow exceeds 0.60 for lead times less than four months using soil moisture, SWE, and precipitation as predictors. This correlation is higher than that of an autoregression model (correlation ~0.50). Since these land-surface and atmospheric variables have no statistically significant correlations with the streamflow, PSPs are then incorporated into the models. The two models have a correlation of ~0.50 using PSPs alone for lead times from six to nine months, and such skills are probably associated with stronger correlation between SST and streamflow in recent decades. The similar prediction skills between the two models suggest a largely linear system between SST and streamflow. Four predictors together can further improve short-lead prediction skills (correlation ~0.80). Therefore, our results confirm the advantage of the Pacific SST information in predicting the UCRB streamflow with a long lead time, and can provide useful climate information for water supply planning and decisions.