Local and Regional Dust simulation - From Emission, Transport to Physical Effects

Mineral dust is an important source of atmospheric ice nuclei (AIN) (Isono et al., 1959) and plays an important role on atmospheric radiation (Twomey, 1974) and the balance of ecosystem (Jickells et al., 2005). For Taiwan, most of the desert dust came from Northern China and Mongolia during spring. However, judging from the seasonal variation of aerosol chemical compositions observed here, a great portion of mineral dust aerosols in Taiwan should be locally produced. In this study, we build up a dust deflation module appropriate for use in the Taiwan area to provide explicit information on local dust emission. Simulations were performed for the period of 2000/09-2001/08 using the MM5 and TAQM/kosa model. In the regional-scale version of TAQM/kosa, the three predictors used for determining the dust mobilization condition are the friction velocity (or surface wind speed), the relative humidity and dominant weather system (Wang et al., 2000). Through sensitivity tests, we designed new emission thresholds appropriate for the Taiwan area. According to the new analysis, we also try to build up modified thresholds for dust emission that can be generally used for various regions. We found that soil moisture can be used to replace the “relative humidity” criteria, provided that the meteorology model can produce reasonable soil moistures. Comparison between locally produced and long-range transported dust concentrations shows that annually local dust contributes to about 71% of the airborne mineral dust in Taiwan. Local dust concentration is the highest in winter due to strong winds associated with the northeast Winter Monsoon in East Asia. But percentage-wise, autumn has the highest local dust proportion. Spatially, local dust is produced mainly in western Taiwan, with average dust mass concentration of about 6.6 μg/m3. The number concentration of dust varied from 105 to 10 per liter and the number concentration of potential ice nuclei is about 100 per liter. Surface radiative forcing caused by local dust is -1.4 W/m2, whereas air heating rate may reach about 0.034 K/day with max. occur mainly in the boundary layer.Abstract ii Acknowledgement iv List of Tables v List of Figures vi Chapter 1. Introduction 1 Chapter 2. Model description 6 2.1 MM5 6 2.2 TAQM/kosa 9 2.3 Radiation model 10 Chapter 3. Dust Emission 14 3.1 Dust emission setup 14 3.2 Emission threshold tests 17 3.2.1 Original thresholds and modification factor 20 3.2.2 No relative humidity threshold 20 3.2.3 Use soil moisture threshold 21 3.2.4 Use soil moisture threshold but without the RH modification factor 21 3.2.5 Use soil moisture threshold and soil moisture modification factor 22 Chapter 4. Local and regional dust simulation 24 4.1 Detailed analysis of local dust events 24 4.1.1 Case 1: Daliau dust event (August 16~17, 2005) 24 4.1.2 Case 2: Southern Taiwan high pollution event (December 21~23, 2005) 26 4.1.3 Case 3: Local dust event accompanied with long-range dust transport (March 19~21, 2006) 29 4.2. Regional dust simulation and verification 30 4.3 Consolidate local and regional dust deflation scheme 34 Chapter 5. Dust effects 37 5.1 Potential ice nuclei 37 5.2 Radiative forcing 38 5.3 Radiative heating rate 40 Chapter 6. Conclusion 42 Reference 45 Appendix A: MM5 Noah Land-Surface Model 49 Appendix B: The comparison of different dust deflation modules 54 Appendix C: Acronym List 5