Potential Impacts Of Dust Acting As Cloud Nucleating Aerosol On Water Resources In The Colorado River Basin

Potential Impacts Of Dust and Pollution Aerosol Acting As Cloud-Nucleating Aerosol On Water Resources In The Colorado River Basin

Vandana Jha, W. R. Cotton, and G. G. Carrió Colorado State University

The Southwest United States has huge demands on water resources. Those in the Colorado River Basin (CRB) are potentially affected by anthropogenic aerosol pollution and dust acting as cloud-nucleating aerosol as well as impacting snowpack albedo.

The specific objectives of this research is to quantify the impacts of both dust and pollution aerosols on wintertime precipitation in the Colorado Mountains. No one to our knowledge has examined the combined effects of dust serving as ice nuclei (IN), giant cloud condensation nuclei (GCCN), and cloud condensation nuclei (CCN) on precipitation, in combination of anthropogenic pollution aerosol and, in particular, on water resources in the CRB. This will be done for entire winter season of 2005. We hypothesize that dust will enhance precipitation, while aerosol pollution will reduce water resources in the CRB via the “spill-over” effect. Since dust is more episodic and aerosol pollution is more pervasive throughout the winter season, we anticipate that the combined response to dust and aerosol pollution is a net reduction of water resources in the CRB. The question is by how much are those water resources affected.

Dust can affect precipitation processes. Moreover, it is well known that dust can serve as IN; (eg: Schaefer 1949, 1954; Roberts and Hallett, 1968; Sassen et al., 2003; DeMott et al., 2003, 2009). It is expected that dust serving as IN will enhance precipitation in wintertime orographic clouds. However, if the dust is coated with sulfates or originates over dry lake beds, it can serve as giant cloud condensation nuclei GCCN which when wetted can result in larger cloud droplets and thereby enhance the warm-rain collision and coalescence process and ice particle riming. Dust serving as GCCN should enhance precipitation, (Levin et al., 1996), thus acting to support the activity of IN. But smaller dust particles coated with sulfates can enhance droplet concentrations, leading to numerous smaller droplets and decreasing collision and coalescence and ice particle riming. Thus dust functioning as CCN may work in opposition to its activity as GCCN and IN and suppressing precipitation. It all depends on the actual sizes and chemistry of the dust particles.

The Colorado State University (CSU) Regional Atmospheric Modeling System (RAMS) version 6.0 is used for this study. Homogeneous ice nucleation of cloud and haze droplets is parameterized using the DeMott et al. (1994), scheme, while heterogeneous ice nucleation is now parameterized using the IN-based scheme of DeMott et al. (2010). RAMS was modified to ingest GEOS-CHEM output data and periodically update aerosol fields (dust and pollution aerosols). GEOS-CHEM is a chemical transport model (Bey et al. 2001), which uses assimilated meteorological data from the NASA Goddard Earth Observation System (GEOS), including wind, convective mass fluxes, mixed layer depths, temperature, clouds, precipitation and surface properties. The smallest size bin of dust, DST1 (0.1-1 microns) is used for dust nudging into the RAMS code. GEOS-Chem is mainly used to estimate long-range transported dust. For local dust sources(N Arizona, Nevada, and New Mexico) we use the dust source and transport model of Dave Lerach's dissertation (2012).

The fields of dust for the two coarsest grids are updated every 12 hours whereas the finest (grid 3) has been kept entirely independent. The aerosol that has been taken from the GEOS-CHEM model consists of 13 different species. The aerosol data comprise of a sum of hydrophobic and hydrophilic black carbon and organic aerosol, hydrophilic SOAs (Lump of aerosol products of first 3 (ALPH + LIMO + TERP) hydrocarbon oxidation, Aerosol product of ALCO oxidation, Aerosol product of SESQ oxidation, Aerosol product of ISOP oxidation, Aerosol product of aromatics oxidation) and inorganic aerosols (nitrate, sulfate and ammonium).

We examine the cumulative effect of dust acting as CCN, GCCN, and IN as well as pollution aerosols CCN over the entire Colorado Rocky Mountains for the period between 1 October and 1 May for 2005, Owing to year-to-year variability we plan to also examine the 2006, 2007 and 2008 winter seasons in the future(Saleeby et al 2009,2010 ). Preliminary results for selected precipitation events (documented SNOTEL data) that corresponded to high dust concentrations are described below. We examined two ways to compute equivalent kappa values. The mass-weighted case gave unphysically low cloud droplet number concentrations and an expected over-estimation in precipitation. Weighing by mass, produces a bias towards the low dust value (0.03). The larger size (and density) of dust overwhelms the impact of the more numerous background aerosols even though they have a greater chemical affinity for water. Weighing kappa by number concentrations produced more reasonable results. Simulations correspond to two 48-hour and 96-hour periods, starting on April 12 and March 23 2005. Dust serving as CCN acts to enhance ice particle riming along with dust acting as GCCN, and since dust enhances IN, total precipitation is enhanced in a dustier atmosphere. The first set of comparisons has been made for April 28. Case 1 ingested both dust and pollution aerosol fields predicted by the GEOS model. Cases 2 considered the predicted dust concentrations but the concentration of pollution aerosols were affected by a factor of 3. Case 3 did not consider dust, only ingesting the predicted concentrations of pollution aerosols. And Case 4 ingested the predicted pollution aerosols but dust concentrations were three times higher. It was found that out of the 4 cases dust increases precipitation the most when increased by a factor of 3. However, it was also observed that when aerosol is increased a factor of three, it leads to a decrease in the precipitation. Similar results are observed for March 23 as well.