TY - Generic T1 - Mapping Snow Grain Size Using LiDAR Intensity (Extended Abstract) T2 - 88th Annual Western Snow Conference Y1 - 2021 A1 - Chelsea Ackroyd A1 - S. McKenzie Skiles KW - intensity KW - LiDAR KW - radiometric correction KW - remote sensing KW - snow grain size AB -

Net solar radiation is the primary driver of snowmelt, which is mainly determined by snow albedo. Controls on snow albedo vary spectrally: in the visible wavelengths it is controlled by light-absorbing particles, including dust and black carbon. For clean snow, snow albedo is dependent upon ice absorption in the near infrared wavelengths, typically characterized using the effective grain size (Warren, 1982). Grain size is currently estimated using radiative transfer inversion methods that leverage reflectance data from passive optical remote sensing imagery. Theoretically, it may also be possible to relate lidar intensity to grain size when the wavelength of the lidar is in the near-infrared wavelengths: smaller grains would reflect more light back to the sensor, larger grains would reflect less. This indicates that lidar could be used to retrieve surface optical properties. Here, we evaluate how well aerial lidar intensity at 1064 nm can be related to the grain size of snow at the basin-scale.

JF - 88th Annual Western Snow Conference CY - Bozeman, MT UR - /files/PDFs/2021Ackroyd.pdf ER - TY - Generic T1 - Snowmelt Detection from Sentinel-1 Synthetic Aperture Radar in the Lajoie Basin, British Columbia T2 - 88th Annual Western Snow Conference Y1 - 2021 A1 - Sara Darychuk A1 - Joseph Shea A1 - Anna Chesnokova A1 - Brian Menounos A1 - Frank Weber A1 - Georg Jost KW - Google Earth Engine KW - remote sensing KW - snowmelt KW - snowpack dynamics KW - Synthetic Aperture Radar AB -

Snowmelt runoff supplements streamflow and soil moisture during warm summer months in Western North America. As direct snowpack measurements are sparse, many models exist to predict the release of runoff in alpine regions. An increase in spatially distributed observational data of seasonal snow may help to refine and improve these efforts going forward. Synthetic Aperture Radar (SAR) is sensitive to the liquid water content of snow and has been successfully used to map wet snow in alpine regions. We employ SAR and multispectral data to estimate the onset and duration of snowmelt in 2018 in the Lajoie Basin, British Columbia. We collate and process Sentinel-1, Sentinel-2 and Landsat-8 images in Google Earth Engine. A backscatter threshold is used to define the inferred period at which the snowpack is saturated and begins to generate runoff. Multispectral imagery is used to estimate snow-free dates across the basin to define the end of the snowmelt period. These methods are most effective on moderate to low slopes (< 30°) in open areas. This approach has high potential for adaptability to other alpine basins or regions and can be used for future model calibration.

JF - 88th Annual Western Snow Conference CY - Bozeman, MT UR - /files/PDFs/2021Darychuk.pdf ER - TY - Generic T1 - Measuring Snow Depth Using RPAS Photogrammetry in a Subalpine Coastal Region of British Columbia (Extended Abstract) T2 - 88th Annual Western Snow Conference Y1 - 2021 A1 - Alexandre Landry A1 - William Floyd A1 - Will McInnes A1 - Keith Holmes A1 - Santiago Gonzalez Arriola A1 - Alex Cebulski A1 - Trevor Dickinson A1 - Stewart Butler A1 - Derek Heathfield A1 - Brian Menounos KW - point cloud filterin KW - remote sensing KW - RPAS KW - snow depth KW - structure from motion photogrammetry KW - vegetation AB -

Mountain snow in British Columbia is primarily monitored using automated remote weather stations, which measure snow depth (SD), snow water equivalent (SWE), and multiple weather parameters, and a manual snow survey program, with records dating back to the 1930’s The network provides excellent information to track change through time and data are effectively used to provide regional forecasts for both flooding and drought, however there is still high uncertainty when scaling these measurements to the watershed level. In addition, there is potential for the current network of stations and snow courses to have more frequent snow free years due to climate change, and gaining a better understanding of how they represent snow that occurs above them is needed. Advances in remote sensing technologies present important opportunities to accurately measure the high degree of spatial variation in SD, and use existing methods (Hill et. al., 2019; Strum et. al., 2010) or develop new approaches to improve estimates of snow density and SWE at finer spatial resolutions. To help address the limitations of relying on discrete, in-situ SD sampling and reduce the uncertainties in overall snowpack measurements, small remotely piloted aircraft systems (RPAS) based structure-from-motion photogrammetry (SFM) has emerged as an effective technology. While they are not capable of measuring areas at the same scale as full-size aircraft or satellites, certain RPAS can be used to capture data across several square kilometres, are much less cost prohibitive and their potential to produce 3D models at high accuracies and very fine spatial resolutions is becoming increasingly well established.

JF - 88th Annual Western Snow Conference CY - Bozeman, MT UR - /files/PDFs/2021Landry.pdf ER - TY - Generic T1 - Snow Pillows, LiDAR, and Streamgauges: Incorporating Snow and Streamflow Observations in the Basin Water Balance T2 - 83rd Annual Western Snow Conference Y1 - 2015 A1 - Brian Henn A1 - Martyn P. Clark A1 - Dmitri Kavetski A1 - Bruce McGurk A1 - Thomas H. Painter A1 - Jessica Lundquist KW - basin hydrology KW - hydrologic modeling KW - orographic precipitation KW - remote sensing AB -

Prior studies have suggested that point measurements of SWE may not be spatially representative of the basin snowpack. The development of remotely-sensed snow observations, such as the Airborne Snow Observatory (ASO) and MODIS-based snow cover estimates, allows for relating point and distributed estimates of snow. We examine how SWE at courses and pillows compares with distributed ASO SWE in water year 2014 in a highelevation basin in Yosemite National Park. We find that peak basin-mean ASO SWE was less than that found by averaging regional snow pillows, but that it melted at a slower rate than pillow SWE during the ablation season. Based on this, we develop an approach for bias-correcting historical snow pillow indices of SWE to better represent the basin mean. We then calibrate an ensemble of lumped hydrologic models to infer basin-mean precipitation from streamflow and SWE. Models calibrated to only streamflow observations have substantial uncertainty in inferred precipitation. Including appropriate SWE observations in the calibration is found to reduce this uncertainty. However, calibrating to fractional snow cover in addition to streamflow did not improve the consistency of the models. We suggest that this approach can improve understanding of water balance components in high-elevation, sparsely-measured basins.

 

Presentation in PDF

JF - 83rd Annual Western Snow Conference T3 - Proceedings of the Western Snow Conference CY - Grass Valley, California UR - /files/PDFs/2015Henn.pdf ER - TY - Generic T1 - Use of Snow Data From Remote Sensing in Operational Streamflow Prediction T2 - 82nd Annual Western Snow Conference Y1 - 2014 A1 - Stacie Bender A1 - Paul Miller A1 - Brent Bernard A1 - John Lhotak KW - operational hydrology KW - remote sensing KW - snow observations KW - snowmelt forecasting KW - Streamflow forecasting AB -

The Colorado Basin River Forecast Center (CBRFC) issues operational forecasts of streamflow for the Colorado River Basin and eastern Great Basin. As part of a multi-year collaborative effort, CBRFC has partnered with the research-oriented Jet Propulsion Laboratory (JPL) under funding from NASA to incorporate remotely sensed snow data from NASA’s MODIS instrument into operational hydrologic forecasting and modeling at CBRFC. Snowpackconditions indicated by MODIS informCBRFC forecasters when determining causes of divergence between modeled and recently observed streamflow. The first two years of the collaborative partnership have yielded improved forecasts at select locations, in select cases, using information from remotely sensed snow data. CBRFC and JPL also retrospectively analyzed relationships between the MODIS-derived snow datasets and streamflow patterns for several watersheds within CBRFC region. The collaboration is expected to continue over the next several years as CBRFC and JPL work to further improve modeling of snowmelt and prediction of snowmelt-driven streamflow in operational hydrologic forecasting.

JF - 82nd Annual Western Snow Conference T3 - Proceedings of the Western Snow Conference CY - Durango, Colorado UR - sites/westernsnowconference.org/PDFs/2014Bender.pdf ER - TY - Generic T1 - Using the Utah Energy Balance Snow Melt Model to Quantify Snow and Glacier Melt in the Himalayan Region T2 - 81st Annual Western Snow Conference Y1 - 2013 A1 - Avirup Sen Gupta A1 - Tarboton, David G. KW - Energy balance KW - glacier and snow melt KW - model KW - remote sensing AB -

Quantification of the melting of glaciers in the Hindu-Kush Himalayan (HKH) region is important for decision making in water sensitive sectors, and for water resources management and flood protection. Access to and monitoring of the glaciers and their melt outflow is challenging, thus modeling based on remote sensing offers the potential for providing information to improve water resources decision making and management. In this paper we report on a distributed version of the Utah Energy Balance (UEB) snowmelt model, referred to as UEBGrid, which was adapted to quantify the melting of glaciers taking advantage of NASA remote sensing and earth science data products such as, satellite data, reanalysis data and climate model outputs. The representations of surface energy balance fluxes in the UEB snowmelt model have been extended to include the capability to quantify glacier melt. To account for clean and debris covered glaciers, substrate albedo determined from remote sensing, and glacier mapping is taken as an input. Representation of glaciers within the model involves inclusion of glacier ice as a substrate and generation of melt from the ice substrate when seasonal snow has melted. In UEBGrid, a watershed is divided into a mesh of grid cells and the model runs individually for each grid cell. Users have control to provide separate inputs for each grid cell, or spatially constant inputs for the entire domain. Therefore, regional variability in snow and glacier melting is computed. Outflow can be aggregated over subwatersheds defined, for example, from a digital elevation model, and input into other hydrologic models. UEBGrid was tested using weather, climate and hydrologic data at Langtang Khola Watershed, Nepal. UEBGrid is being included into the EPA BASINS software to facilitate linking to other models and to take advantage of BASINS’s capability to manage input data and visualize results. This capability for using gridded NASA Earth Science data, and the associated data model and workflow for storage and processing of data into and out of models linked in BASINS advances hydrologic information science. The capability for estimating the melt from glaciers and snow in a data sparse region will help water managers in decision making and management of water resources in areas impacted by glacier and snow melt.

JF - 81st Annual Western Snow Conference T3 - Proceedings of the Western Snow Conference CY - Jackson Hole, Wyoming UR - sites/westernsnowconference.org/PDFs/2013SenGupta.pdf ER - TY - Generic T1 - Observing the Elusive Intermittent Snow Using Traffic Camera Images T2 - 81st Annual Western Snow Conference Y1 - 2013 A1 - Wayand, Nicholas A1 - Lundquist, Jessica KW - intermittent snow KW - rain-on-snow KW - remote sensing KW - traffic cameras AB -

Intermittent snow covers a majority of the western U.S. and is known to have contributed to rain-on-snow type floods. Quantifying the elevational contributions of melt water during such events is hampered from a critical lack of observations of the lowest snow extent. The Moderate Resolution Imaging Spectroradiometer (MODIS) aboard Terra and Aqua offers the best available compromise of sampling and resolution to capture intermittent snow’s spatio-temporal snow cover variations over mid-latitudes. However, twice daily images from MODIS are significantly limited by cloud cover that may obscure rapidly melting snowcover; a critical model initialization state for flood forecasting models. In this study, we approach the question of quantifying the snowmelt contribution through an idealized simulation, followed by a comparison of the current available observations of intermittent snowcover.

 

JF - 81st Annual Western Snow Conference T3 - Proceedings of the Western Snow Conference CY - Jackson Hole, Wyooming UR - sites/westernsnowconference.org/PDFs/2013Wayand.pdf ER -