Optimal Postprocessing Strategies With LSTM for Global Streamflow Prediction in Ungauged Basins

Water Resources Research, Journal Year: 2023, Volume and Issue: 59(7)

Published: July 1, 2023

Abstract Streamflow prediction in ungauged basins (PUB) is challenging, and Long Short‐Term Memory (LSTM) widely used to for such predictions, owing its excellent migration performance. Traditional LSTM forced by meteorological data catchment attribute barely highlight the optimum integration strategy from data‐rich ones. In this study, we experimented with 1,897 global catchments found that LSTM‐corrected Global Hydrological Models (GHMs) outperformed uncorrected GHMs, improving median Nash‐Sutcliff efficiency (NSE) 0.03 0.66. Notably, there was a large gap between traditional modeling autoregressive basins, GHM‐forced were an effective way close basins. The spatial heterogeneity of performance mainly influenced three metrics (dryness, leaf area index latitude), which described hydrological similarity among catchments. Weaker continental results larger variability LSTM, best Siberia (NSE, 0.54) worst North America 0.10). However, significantly improved 0.63) when considered. This study stressed advantages due significance should be attached similarities improve

Language: Английский

---

Restoring Historic Forest Disturbance Frequency Would Partially Mitigate Droughts in the Central Sierra Nevada Mountains

Water Resources Research, Journal Year: 2025, Volume and Issue: 61(4)

Published: April 1, 2025

Abstract Forest thinning and prescribed fire are expected to improve the climate resilience water security of forests in western U.S., but few studies have directly modeled hydrological effects multi‐decadal landscape‐scale forest disturbance. By updating a distributed process‐based model (DHSVM) with vegetation maps from ecosystem (LANDIS‐II), we simulate resource impacts management scenarios targeting partial or full restoration pre‐colonial disturbance return interval central Sierra Nevada mountains. In fully restored regime that includes fire, thinning, insect mortality, reservoir inflow increases by 4%–9% total 8%–14% dry years. At sub‐watershed scales (10–100 km 2 ), dense can increase streamflow >20% thinner forest, increased understory transpiration compensates for decreased overstory transpiration. Consequentially, 73% gains attributable rain snow interception loss. Thinner headwater peak flows, reservoir‐scale flows almost exclusively influenced climate. Uncertainty future precipitation causes high uncertainty yield, additional yield is about five times less sensitive annual uncertainty. This decoupling response makes especially valuable supply during Our study confidence benefits restoring historic frequencies mountains, our modeling framework widely applicable other forested mountain landscapes.

Language: Английский

---

Bark beetle impacts on forest evapotranspiration and its partitioning

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 880, P. 163260 - 163260

Published: April 6, 2023

Insect outbreaks affect forest structure and function represent a major category of disturbance globally. However, the resulting impacts on evapotranspiration (ET), especially hydrological partitioning between abiotic (evaporation) biotic (transpiration) components total ET, are not well constrained. As result, we combined remote sensing, eddy covariance, modeling approaches to determine effects bark beetle outbreak ET its at multiple scales throughout Southern Rocky Mountain Ecoregion (SRME), USA. At covariance measurement scale, 85 % was affected by beetles, water year as fraction precipitation (P) decreased 30 relative control site, with 31 greater reductions in growing season transpiration ET. ecoregion satellite sensing masked areas >80 tree mortality showed corresponding ET/P 9-15 that occurred 6-8 years post-disturbance, indicated majority reduction during season; Variable Infiltration Capacity model an associated 9-18 increase runoff ratio. Long-term (16-18 year) vegetation datasets extend length previously published analyses allowed for clear characterization recovery period. During time, outpaced recovery, which lagged part due persistently reduced winter sublimation, there evidence increasing late summer moisture stress. Overall, comparison three independent methods two demonstrated net negative impact beetles relatively transpiration, following SRME.

Language: Английский

---

Revisiting the tension water storage capacity distribution in conceptual rainfall-runoff modeling: A large-sample approach

Journal of Hydrology, Journal Year: 2024, Volume and Issue: 641, P. 131834 - 131834

Published: Aug. 13, 2024

Accurately characterizing the spatial variability of tension water storage capacity (TWC) within a catchment is challenging due to limited in-situ hydrologic data availability. Conventional conceptual rainfall-runoff models typically rely on an empirically specified TWC distribution. However, this empirical distribution lacks physical foundation and fails effectively redistribute critical components, such as local contributing area, real-world contexts. To overcome limitation, topographic wetness index (TWI) its generalized form (GTWI) are introduced bridge information with components. Four curves contrived based parabolic distribution, linear TWI, GTWI, respectively. The effects these alternate distributions streamflow investigated framework HYdrologic MODel (HYMOD) across 460 Australian catchments. results illustrate that GTWI-based HYMOD (GTHYMOD) outperforms other in terms daily streamflow, high Kling-Gupta Efficiency (KGE) attained 74.8% study catchments during validation period. eastern coast presents superior performance compared western coast. GTHYMOD demonstrates superiority variability, aspect lacked. This has potential refine from perspective advance our comprehension underlying behaviors.

Language: Английский

---

Evaluating Noah‐MP Simulated Runoff and Snowpack in Heavily Burned Pacific‐Northwest Snow‐Dominated Catchments

Journal of Geophysical Research Atmospheres, Journal Year: 2024, Volume and Issue: 129(9)

Published: April 25, 2024

Abstract Terrestrial hydrology is altered by fires, particularly in snow‐dominated catchments. However, fire impacts on catchment are often neglected from land surface model (LSM) simulations. Western U.S. wildfire activity has been increasing recent decades and projected to continue over at least the next three decades, thus it important evaluate if neglecting operational models (LSMs) a significant error source that noticeable signal among other sources of uncertainty. We widely used state‐of‐the‐art LSM (Noah‐MP) runoff snowpack simulations two representative fire‐affected catchments Pacific Northwest: Andrew's Creek Washington Johnson Idaho. These selected across all western because they experienced more than 50% burning single event with minimal outside this event, which allows analyses distinct pre‐ post‐fire periods. There statistically shifts skills pre‐to years simulating snowpack. At both study catchments, miss enhancements early‐spring annual efficiency during years, resulting persistent underestimates anomalies throughout 12‐year analysis Enhanced snow accumulation melt contributes observed but unmodeled increases spring these Informing satellite cover classifications, leaf area index, green fraction do not consistently improve ability simulate hydrologic responses disturbances.

Language: Английский

---

A global streamflow indices time series dataset for large-sample hydrological analyses on streamflow regime (until 2021)

Published: March 6, 2023

Abstract. With the booming big data techniques, large-sample hydrological analysis on streamflow regime is becoming feasible, which could derive robust conclusions processes from a big-picture perspective. However, there not comprehensive global dataset for components of yet. This paper presents new time series indices calculated daily records after quality control. The contains 79 over seven major (i.e., magnitude, frequency, duration, changing rate, timing, variability, and recession) 5548 river reaches globally. in are available until 2021, lengths vary 30 to 215 years with an average around 66 years. Restricted-access typical basins China included dataset. Compared existing datasets, this covers more indices, especially those characterizing recession regime. dataset, research will become easier without spending handling raw records. be valuable resource hydrology community facilitate wide range studies, such as studies behaviour catchment, prediction data-scarce regions, well variations

Language: Английский

---

A global streamflow indices time series dataset for large-sample hydrological analyses on streamflow regime (until 2022)

Earth system science data, Journal Year: 2023, Volume and Issue: 15(10), P. 4463 - 4479

Published: Oct. 6, 2023

Abstract. With the booming big data techniques, large-sample hydrological analysis on streamflow regime is becoming feasible, which could derive robust conclusions processes from a big-picture perspective. However, there lack of comprehensive global dataset for components yet. This paper presents new time series indices calculated daily records after quality control. The contains 79 over seven major (i.e., magnitude, frequency, duration, changing rate, timing, variability, and recession) 41 263 river reaches globally yearly multiyear scales. Streamflow values until 2022 are covered in dataset. Time span 1806 to with an average length 36 years. Compared existing datasets, this covers more stations indices, especially those characterizing recession regime. dataset, research will become easier without spending handling raw records. be valuable resource hydrology community facilitate wide range studies, such as studies behaviour catchment, prediction data-scarce regions, well variations can accessed at https://doi.org/10.57760/sciencedb.07227 (Chen et al., 2023a).

Language: Английский

---

Water yield response to forest treatment patterns in a sierra nevada watershed

Journal of Hydrology Regional Studies, Journal Year: 2024, Volume and Issue: 53, P. 101762 - 101762

Published: April 5, 2024

Sierra Nevada Snow dominated forests serve as source water supplies to much of the western United States. In recent years, these have experienced an increase in both drought and wildfire, which threaten critical resources. By reducing forest density, treatments offer a promising solution fuel loads potentially increasing surface yield. However, amount load reduction necessary produce change yield is not well-characterized. The objectives current study are to: 1) calibrate distributed parameter, fully-integrated physically-based hydrologic model available watershed data evaluate predicted changes for range treatment scenarios heavily forested experimental basin (Sagehen Creek near Truckee, California, USA), 2) determine extent (threshold) substantial changes, i.e. 25% increase, runoff. Using DHI's physically-distributed code, MIKESHE, develop Sagehen Basin, twenty with varying canopy density reductions (CDRs) areas were simulated five-year period. Statistical testing showed that significant runoff occurred every developed scenario at annual scale (99% confidence interval). Surface highly correlated precipitation patterns, however, compounding effect over years has more dramatic response higher intensities. Increasing CDR was effective than area. Results suggest managers implement treatments, depends on: degree implemented by either area or CDR, future climate variability including extended periods drought, 3) storage conditions how this might buffer disturbance basin.

Language: Английский

---

Forestry impacts on stream flows and temperatures: A quantitative synthesis of paired catchment studies across the Pacific salmon range

Ecological Solutions and Evidence, Journal Year: 2024, Volume and Issue: 5(2)

Published: April 1, 2024

Abstract Forestry is pervasive across temperate North America and may influence aquatic environmental conditions such as flows temperatures, well important species Pacific salmon ( Oncorhynchus spp.). While there have been many large‐scale forestry experiments using paired catchment designs, these studies yet to be quantitatively synthesized. Thus, it remains unclear whether impacts are consistent, context‐dependent or unpredictable. This study aims synthesize on streamflow temperature, through a systematic review synthesis of the range salmon. Specifically, we investigated generalizable relationships exist between intensity (percent watershed harvested) temperature. We also examined features (climate, hydrology lithology) harvest method mediated impacts. extracted information from 35 unique paired‐catchments California Alaska. had strong peak low maximum summer water but responses were quite variable. Across all catchments, elevated ~20% n = 31 catchments), reduced ~25% 13 catchments) increased temperatures ~15% average. However, variable not predictable based intensity, thus broader stressor–response supported. varied spatially. Peak flow with northward latitude temperature decreased eastward longitude. magnitude unrelated other attributes, which included climate (precipitation aridity), rain versus snow hydrology, elevation bedrock lithology. Harvest riparian buffer presence no detected effects statistical models explained proportion variation overall. Collectively, our results indicate that can substantial key conditions; however, impact was could clearly linked easily measured characteristics. implies broadly predictable. Probabilistic risk distributions potential therefore more useful for management in data‐poor situations.

Language: Английский

---

Landscape metrics as predictors of water-related ecosystem services: Insights from hydrological modeling and data-based approaches applied on the Arno River Basin, Italy

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 954, P. 176567 - 176567

Published: Sept. 30, 2024

Language: Английский

---