Research
Radiative Transfer Modeling
2-stream radiative transfer modeling for snow and sea ice
We evaluate three 2-stream algorithms (SNICAR, Icepack/CICE/MPAS-seaice, and 2-stream discrete ordinate (2SD)) on their performances in modeling snow solar properties. Based on the results, we propose a unified solar radiative transfer model for all cryosphere components in Earth System Models. A parameterization is also developed to adjust the underestimated direct-IR albedo and overestimate near-IR heating persistent across all 2-stream models when solar zenith angles is larger than 75°.
Parameterization on broadband and narrowband albedo of pure snow and snow containing LAPs
Using radiative transfer model DISORT, we computed the spectral albedo of pure snow and snow containing BC and MD. Broadband albedo is shown for mass fractions covering the full range from pure snow to pure BC and pure dust, and for snow grain radii from 5 μm to 2500 μm, to cover the range of possible grain sizes on planetary surfaces. Parameterizations are developed for opaque homogeneous snowpacks for three broad bands used in general circulation models and several narrower bands. For more details, please see paper.
Effect of snow grain shape on snow albedo
Radiative transfer models of snow albedo have usually assumed spherical snow grains, using Mie theory to compute single-scattering quantities. The scattering by realistic nonspherical snow grains is less in the forward direction and more to the sides, resulting in a smaller asymmetry factor g. Compared to a snowpack of spherical grains with the same area-to-mass ratio, a snowpack of nonspherical grains will have a higher albedo, thin snowpacks of nonspherical grains will more effectively hide the underlying surface, and light-absorbing impurities in the snowpack will be exposed to less sunlight. These effects are examined for a variety of snow grain shapes, with aspect ratio from 0.1 to 10. For more details, please see paper.
Radiative effects of light-absorbing particles in snow: China, North America and the Arctic
Using field observations, we perform radiative transfer calculations on snowpacks in the Arctic, China, and North America to quantify the impact of light-absorbing particles (LAPs) on snow albedo and its sensitivity to different factors. For new snow, the regional-averaged albedo reductions caused by all LAPs in the Arctic, North America, and China are 0.009, 0.012, and 0.077, respectively, of which the albedo reductions caused by black carbon (BC) alone are 0.005, 0.005, and 0.031, corresponding to a positive radiative forcing of 0.06, 0.3, and 3 W m2. For more details, please see paper.
Field Campaign
Field campaign in western North America
From January to March of year 2013, I participated a large area survey of light-absorbing particles in snow, led by Steve Warren and Sarah Doherty. We collected more than 600 snow samples from 67 sites at 13 US states and 3 Canadian provinces. At each site, vertical profiles of snow samples were collected; snow depth and vertical profiles of snow density were measured. We processed these samples in temporary labs along the trip, and collected the light-absorbing particles by nuclepore filters and GHP filters. These LAPs samples were then brought back to the lab at the University of Washington for optical measurement and chemical experiments.
Laboratory Analysis
Optical analysis on light-absorbing particle (LAPs) samples
We measured the nuclepore samples using ISSW spectrophotometer (Grenfell et al., 2011) developed earlier by our group to quantify the light absorption by LAPs. Since BC and non-BC LAPs have different light-absorbing properties (i.e. absorbing Angstrom exponent), we can separate the light absorption due to BC and non-BC LAPs. This method was developed by Grenfell et al., 2011 and has been used in this field campaign as well. The maximum amounts of BC for all sampling layers and sites see Figure 3. The regional median concentration of BC in western North America varies from 25-111ng/g, much smaller than what have been found in China: 340 to 1220 ng/g (Wang et al., 2013, Table 1). To learn more details about field campaign, optical analysis, and chemical analysis on sources of LAPs, please see paper.
Chemical analysis on light-absorbing particle (LAPs) samples
Besides using optical measurements to separate the light absorption due to BC and non-BC LAPs, we also applied serial chemical extractions on LAPs samples acquired from snow samples, to study the light absorption by different types of OC. The results suggest that humic-like substances (sodium hydroxide (NaOH)-soluble), polar OCs (methanol-soluble), and iron oxides are responsible for 9%, 4%, and 14% of the total light absorption, respectively. The total light absorption due to non-BC LAPs estimated by chemical methods is lower than that estimated by optical method by about 10% in all sampling regions. The reasons caused such discrepancy and more about this work are discussed in paper