Towards a Multi-Variate, Multi-Sensor Assimilation Framework over Snow-Covered Terrain in Western Colorado

Snow is an important component of global freshwater storage. It provides a freshwater supply for more than 1 billion people who lived in the high-latitude and mountainous environments. Therefore, accurate snow measurements, including snow water equivalent (SWE) and snow depth (SD), are essential for use in snowmelt-driven runoff predictions and water resources management applications. A number of different remote sensing techniques exist to detect snow mass, including passive microwave (PMW) radiometry, active microwave (AMW) radar, and LIDAR. For example, a spectral difference of brightness temperatures derived from PMW sensors is sensitive to snow volume scattering and provides a rapid revisit interval, but the signal-to-noise ratio is severely limited in regions of deep and/or wet snow. Compared to PMW radiometry, LIDAR can detect subtle snow depth variations, but with a typically low revisit rate. Each technique has its own strengths and weaknesses that must be weighed relative to one another in order to maximize the information content from a suite of feasible sensor configurations.

In this study, we assess the added value of observations from sensors in a hypothetical constellation consisting of one (or more) PMW radiometers, a wide-swath LIDAR, a narrow-swath LIDAR, and a synthetic aperture radar (SAR) sensor versus the observations from the PMW, LIDAR, or SAR sensor only. We employ an observing system simulation experiment (OSSE) using geophysical states and fluxes (with a focus on snow) derived from the Noah-MP v3.6 land surface model in the NASA Land Information System (LIS). In our experiment, the "nature run" is forced by meteorological fields from MERRA-2. The open loop (OL) case was similar in design, but forced with TRMM precipitation as to represent plausible precipitation errors relative to MERRA2. Assimilation experiments were then conducted using synthetic observations from a suite of different sensor configurations (i.e., different permutations of PMW radiometers, LIDAR, and/or SAR). Results were evaluated against a common benchmark. Evaluation of performance metrics from the OSSE will help inform future mission design for use in characterizing global snow mass.