Objectives: the proposed work will quantify the temporal and spatial variability of snow and ICE in high mountain asia (HMA). We will use satellite remote sensing from 1985 to present to create a…

This federal contract award to the University of California, Santa Barbara (UCSB) aims to quantify the temporal and spatial variability of snow and ice in High Mountain Asia (HMA) using satellite remote sensing from 1985 to the present. The $185,877.84 grant will create a 35+ year daily record of snow/ice surface properties and mass estimates, including snow water equivalent (SWE), snow/ice melt, and sublimation, at 500 m spatial resolution. This extensive data set will serve as a unique calibration and validation tool for regional climate models. The research will involve combining measurements from various satellite sensors and using reconstruction and machine learning techniques. The work is not associated with a set-aside contract and is being funded by a civilian federal agency, likely focused on climate research and modeling. As a prominent public research university, UCSB has significant experience delivering specialized scientific research and innovative solutions to federal agencies across diverse domains.

Objectives: the proposed work will quantify the temporal and spatial variability of snow and ICE in high mountain asia (HMA). We will use satellite remote sensing from 1985 to present to create a daily 35+ year record of snow/ICE surface properties and mass estimates that include snow water equivalent (SWE), snow/ICE melt, and sublimation, all at 500 m spatial resolution. A daily record of this length is useful for regional climate models (rcms) as IT captures both oceanic and atmosphere oscillations that occur on decadal or longer time scales, in addition to inter-annual variability. Furthermore, our approach does not rely on total precipitation estimates from reanalysis or observational datasets, making IT a unique calibration and validation tool. Through satellite observations and energy balance modeling efforts, we will contribute to an improved understanding of the processes controlling change in HMA. Methods: we will combine areal measurements of snow surface properties with reconstruction and machine learning techniques to estimate SWE and daily snow/ICE loss. We will rely on overlapping periods for sensors from landsat (30 m, 1985 to present), modis (500 m, 2000 to present), viirs (1 KM, 2012 to present), sentinel (15 m, 2017 to present), and a series of passive microwave satellites (enhanced to 3-6 KM, 1985 to present) to produce optimal estimates. We will validate our remotely-sensed products using measurements from the following higher resolution sensors/platforms: aviris, aviris-NG, and worldview 2/3. We will estimate SWE and loss from maximum accumulation to melt out using a reconstruction technique. For the period prior to maximum accumulation, we will use a machine learning method called random forests trained on variables available in near-real time, for prediction. We will validate our SWE estimates and snow surface properties with existing and planned in situ measurements that span the various climate regions. To better understand where most of the water originates in HMA, we will classify sources into: snow on land (SOL), snow on ICE (SOI), exposed glacier ICE (EGI), and supraglacial debris, taking into account that the size, melt volume, and timing of melt vary for these cryospheric reservoirs. Our partitioning also allows direct estimation of regional snow line altitudes (SLA) and the equilibrium line altitude (ELA) that separates the accumulation zone of a glacier from the ablation zone, while ground observations can only observe this for individual glaciers. We will investigate the inter-annual variability of snow and ICE across the entire region as well as long-term trends in SOL, SOI, EGI, and ELA. These modeling efforts will expand knowledge of processes controlling snow/ICE changes in HMA. Benefit to NASA: the proposed work incorporates data from a number of NASA sensors (e.g. modis, ceres, aviris-NG) into snow/ICE surface and volume estimates that will be provided to the himat team and incorporated into the gmelt toolbox. Improved confidence in modeling the historical period using remote-sensing datasets increases our certainty for projections of future climate and water availability. We expect climate modelers to use our data to better simulate present day snow/ICE melt, snowfall, snow/ICE albedo, and precipitation phase, requisite to project future changes. We will provide software to the gmelt toolbox that summarizes our products in time and space. The proposed research addresses NASA earth science focus areas in: climate change and variability, water and energy cycle, and societal benefits.