Water Cycle Snow

Project Title: The Role of EMIT and Thermal Infrared Remote Sensing in Addressing the Decadal Survey’s Most Important Objectives for Snow in the Water Cycle

Two Most Important objectives for the hydrologic cycle in the National Academies' Thriving on Our Changing Planet (2018) identify hyperspectral imagery in the visible and shortwave infrared and multispectral imagery in the thermal infrared as essential observables:

H-1a. Develop and evaluate an integrated Earth system analysis with sufficient observational input to accurately quantify the components of the water and energy cycles & H-1c. Quantify rates of snow accumulation, snowmelt, ice melt, and sublimation from snow and ice worldwide at scales driven by topographic variability.

A subset of Objective H-1a includes energy balance models of the snowpack, driven by measurements of snow albedo and snow surface temperatures. A subset of Object H-1c requires the ability to measure snow albedo and temperature and understand how and why they vary, especially in the world's mountains, which comprise a quarter of the world's land area but produce at least half of the runoff.

The Decadal Survey's traceability matrix (Table B.1) clarifies the specific need: Spectral albedo of subpixel snow and glaciers at weekly intervals to an accuracy to estimate absorption of solar radiation to 10%. Ice/snow surface temperature to ±1 K. At spatial resolution of 30 to 100 m."

EMIT's temporal and spatial coverage do not enable that mission to meet the Decadal Survey's needs, but its spectral capabilities match proposed designs for the VSWIR spectrometer on NASA's SBG and Europe's CHIME. The EMIT imagery therefore provides the opportunity to demonstrate retrieval of snow properties to meet the 10% criterion for the absorption of solar radiation. Similarly, currently available remotely sensed temperatures lack the combined spectral and spatial capability to estimate subpixel snow temperature at ±1 K accuracy, but the thermal infrared component of SBG will enable estimates of subpixel snow temperature. Current sensors can measure temperatures of fully snow-covered pixels.

The proposed investigation will address the following Questions that will use EMIT imagery to help develop appropriate Level 2 and Level 4 products for SBG and CHIME. Q1. How accurately must we measure snow properties to estimate absorption of solar radiation to 10% accuracy?

Q2. How does sensitivity of imaging spectroscopy retrievals of snow and glacier properties in the mountains vary depending on the retrieval algorithms and the available ancillary data?

Q3. How can knowledge of snow albedo, along with surface temperature at two times in the diurnal cycle, constrain an energy balance model of the snowpack? The corresponding Objectives are:

O1. Calculate the accuracy of snow properties needed to estimate absorption of solar radiation.

O2. Assess the sources of uncertainty in algorithms to retrieve snow properties. O3. Determine whether retrieved snow properties from EMIT meet necessary accuracy, based on surface measurements of albedo that are already available. O4. Examine whether snow energy balance models can be improved by knowledge of remotely sensed surface temperature.

O5. Compare available EMIT data and coincident temperature measurements to drive a snow energy balance model.

In most snow-dominated environments, net solar radiation drives melt and metamorphism of snow. Inaccuracy in the estimation of snow albedo constitutes the major source of uncertainty in calculating the snowpack's energy balance. Meeting the Decadal Survey's objectives requires improvement in spatially distributed models of snow melt and metamorphism, which the EMIT data provide a path to achieve.

Project Team

Jeff Dozier

University of California, Santa Barbara