Blowing dust and highway safety in the southwestern United States: Characteristics of dust emission "hotspots" and management implications.

Despite the widespread media attention of chain-reaction traffic incidents and property damage caused by windblown dust in the U.S. and elsewhere in the world, very few studies have provided in-depth analysis on this issue. Remote sensing and field observations reveal that wind erosion in the southwestern U.S. typically occurs in localized source areas, characterized as "hotspots", while most of the landscape is not eroding. In this study, we identified the spatial and temporal distribution patterns of hotspots that may contribute dust blowing onto highways in the southwestern U.S. We further classified the hotspots for the potential of blowing dust production based upon field observations and wind erosion modeling. Results of land use and land cover show that shrubland, grassland, and cropland accounted for 42%, 31%, and 21% of the overall study area, respectively. However, of the 620 total hotspots identified, 164 (26%), 141 (22%), and 234 (38%) are located on shrubland, grassland, and cropland, respectively. Barren land represented 0.9% of the land area but 8% of the dust hotspots. While a majority of these hotspots are located close to highways, we focused on 55 of them, which are located <1km to adjacent highways and accessible via non-private roads. Field investigations and laboratory analysis showed that soils at these hotspot sites are dominated by sand and silt particles with threshold shear velocities ranging from 0.17-0.78m s-1, largely depending on the land use of the hotspot sites. Dust emission modeling showed that 13 hotspot sites could produce annual emissions >3.79kg m-2, yielding highly hazardous dust emissions to ground transportation with visibility <200m. Results of location, timing, and magnitude of the dust production at the hotspots are critical information for highway authorities to make informed and timely management decisions when wind events strike.