An official website of the United States government
Abstract. Seasonal snowpack is an essential component in the hydrological cycle and plays a significant role in supplying water resources to downstream users. Yet the snow water equivalent (SWE) in seasonal snowpacks, and its space–time variation, remains highly uncertain, especially over mountainous areas with complex terrain and sparse observations, such as in High Mountain Asia (HMA). In this work, we assessed the spatiotemporal distribution of seasonal SWE, obtained from a new 18-year HMA Snow Reanalysis (HMASR) dataset, as part of the recent NASA High Mountain Asia Team (HiMAT) effort. A Bayesian snow reanalysis scheme previously developed to assimilate satellite-derived fractional snow-covered area (fSCA) products from Landsat and MODIS platforms has been applied to develop the HMASR dataset (at a spatial resolution of 16 arcsec (∼500 m) and daily temporal resolution) over the joint Landsat–MODIS period covering water years (WYs) 2000–2017. Based on the results, the HMA-wide total SWE volume is found to be around163 km3 on average and ranges from 114 km3 (WY2001) to 227 km3 (WY2005) when assessed over 18 WYs. The most abundant snowpacks are found in the northwestern basins (e.g., Indus, Syr Darya and Amu Darya) that are mainly affected by the westerlies, accounting for around 66 % of total seasonal SWE volume. Seasonal snowpack in HMA is depicted by snow accumulating through October to March and April, typically peaking around April and depleting in July–October, with variations across basins and WYs. When examining the elevational distribution over the HMA domain, seasonal SWE volume peaks at mid-elevations (around 3500 m), with over 50 % of the volume stored above 3500 m. Above-average amounts of precipitation causes significant overall increase in SWE volumes across all elevations, while an increase in air temperature (∼1.5 K) from cooler to normal conditions leads to an redistribution in snow storage from lower elevations to mid-elevations. This work brings new insight into understanding the climatology and variability of seasonal snowpack over HMA, with the regional snow reanalysis constrained by remote-sensing data, providing a new reference dataset for future studies of seasonal snow and how it contributes to the water cycle and climate over the HMA region.
more »
« less
A Spatially‐Distributed Machine Learning Approach for Fractional Snow Covered Area Estimation
Abstract Snowpack in mountainous areas often provides water storage for summer and fall, especially in the Western United States. In situ observations of snow properties in mountainous terrain are limited by cost and effort, impacting both temporal and spatial sampling, while remote sensing estimates provide more complete spacetime coverage. Spatial estimates of fractional snow covered area (fSCA) at 30m are available every 16 days from the series of multispectral scanning instruments on Landsat platforms. Daily estimates at 463m spatial resolution are also available from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument on the Terra satellite. Fusing Landsat and MODIS fSCA images creates high resolution daily spatial estimates of fSCA that are needed for various uses: to support scientists and managers interested in energy and water budgets for water resources and to understand the movement of animals in a changing climate. Here, we propose a new machine learning approach conditioned on MODIS fSCA, as well as a set of physiographic features, and fit to Landsat fSCA over a portion of the Sierra Nevada USA. The predictions are daily 30m fSCA. The approach relies on two stages of spatially‐varying models. The first classifies fSCA into three categories and the second yields estimates within (0, 100) percent fSCA. Separate models are applied and fitted within sub‐regions of the study domain. Compared with a recently‐published machine learning model (Rittger, Krock, et al., 2021), this approach uses spatially local (rather than global) random forests, and improves the classification error of fSCA by 16%, and fractionally‐covered pixel estimates by 18%.
more »
« less
- PAR ID:
- 10576667
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Water Resources Research
- Volume:
- 60
- Issue:
- 11
- ISSN:
- 0043-1397
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
WaterSmart project is an NSF funded projected seeks water consumption reduction using satellite observations. In order to fit the fine temporal resolution requirement, satellites are required to have a high revisit cycle. MODIS is an ideal platform for monitoring the ground thanks to its daily coverage while the spatial resolution is too coarse. Research has demonstrated the possibility to improve the spatial resolution of MODIS using the Landsat 8 images. This research is aimed to establish a workflow to adapt the data fusion algorithm to achieve automatically processing at real-time in order to support short-term decision making.more » « less
-
Understanding and monitoring nearshore environments is essential, given that these fine-scaled ecosystems are integral to human well-being. While satellites offer an opportunity to gain synchronous and spatially extensive data of coastal areas, off-the-shelf calibrated satellite sea surface temperature (SST) measurements have only been available at coarse resolutions of 1 km or larger. In this study, we develop a novel methodology to create a simple linear equation to calibrate fine-scale Landsat thermal infrared radiation brightness temperatures (calibrated for land sensing) to derive SST at a resolution of 100 m. The constants of this equation are derived from correlations of coincident MODIS SST and Landsat data, which we filter to find optimal pairs. Validation against in situ sensor data at varying distances from the shore in Northern California shows that our SST estimates are more accurate than prior off-the-shelf Landsat data calibrated for land surfaces. These fine-scale SST estimates also demonstrate superior accuracy compared with coincident MODIS SST estimates. The root mean square error for our minimally filtered dataset (n = 557 images) ranges from 0.76 to 1.20 °C with correlation coefficients from r = 0.73 to 0.92, and for our optimal dataset (n = 229 images), the error is from 0.62 to 0.98 °C with correlations from r = 0.83 to 0.92. Potential error sources related to stratification and seasonality are examined and we conclude that Landsat data represent skin temperatures with an error between 0.62 and 0.73 °C. We discuss the utility of our methodology for enhancing coastal monitoring efforts and capturing previously unseen spatial complexity. Testing the calibration methodology on Landsat images before and after the temporal bounds of accurate MODIS SST measurements shows successful calibration with lower errors than the off-the-shelf, land-calibrated Landsat product, extending the applicability of our approach. This new approach for obtaining high-resolution SST data in nearshore waters may be applied to other upwelling regions globally, contributing to improved coastal monitoring, management, and research.more » « less
-
Abstract With increasing demands for precise water resource management, there is a growing need for advanced techniques in mapping water bodies. The currently deployed satellites provide complementary data that are either of high spatial or high temporal resolutions. As a result, there is a clear trade‐off between space and time when considering a single data source. For the efficient monitoring of multiple environmental resources, various Earth science applications need data at high spatial and temporal resolutions. To address this need, many data fusion methods have been described in the literature, that rely on combining data snapshots from multiple sources. Traditional methods face limitations due to sensitivity to atmospheric disturbances and other environmental factors, resulting in noise, outliers, and missing data. This paper introduces Hydrological Generative Adversarial Network (Hydro‐GAN), a novel machine learning‐based method that utilizes modified GANs to enhance boundary accuracy when mapping low‐resolution MODIS data to high‐resolution Landsat‐8 images. We propose a new non‐saturating loss function for the Hydro‐GAN generator, which maximizes the log of discriminator probabilities to promote stable updates and aid convergence. By focusing on reducing squared differences between real and synthetic images, our approach enhances training stability and overall performance. We specifically focus on mapping water bodies using MODIS and Landsat‐8 imagery due to their relevance in water resource management tasks. Our experimental results demonstrate the effectiveness of Hydro‐GAN in generating high‐resolution water body maps, outperforming traditional methods in terms of boundary accuracy and overall quality.more » « less
-
Hybrid Physically Based and Deep Learning Modeling of a Snow Dominated, Mountainous, Karst WatershedAbstract Snow dominated mountainous karst watersheds are the primary source of water supply in many areas in the western U.S. and worldwide. These watersheds are typically characterized by complex terrain, spatiotemporally varying snow accumulation and melt processes, and duality of flow and storage dynamics because of the juxtaposition of matrix (micropores and small fissures) and karst conduits. As a result, predicting streamflow from meteorological inputs has been challenging due to the inability of physically based or conceptual hydrologic models to represent these unique characteristics. We present a hybrid modeling approach that integrates a physically based, spatially distributed, snow model with a deep learning karst model. More specifically, the high‐resolution snow model captures spatiotemporal variability in snowmelt, and the deep learning model simulates the corresponding response of streamflow as influenced by complex surface and subsurface properties. The deep learning model is based on the Convolutional Long Short‐Term Memory (ConvLSTM) architecture capable of handling spatiotemporal recharge patterns and watershed storage dynamics. The hybrid modeling approach is tested on a watershed in northern Utah with seasonal snow cover and variably karstified carbonate bedrock. The hybrid models were able to simulate streamflow at the watershed outlet with high accuracy. The spatial and temporal recharge and discharge patterns learned by the ConvLSTM model were then examined and compared with known hydrogeologic information. Results suggest that ConvLSTM simulates streamflow with higher accuracy than reference models for the study area and provides insight into spatially influenced hydrologic responses that are unavailable within lumped modeling approaches.more » « less
