CEMSYS modelled wind erosion

摘要: [Executive Summary]: The Leys report on wind and water erosion (Leys et al. 2009b) recommended that modelling be undertaken to assist in reporting the extent severity of across Australia. The could then used by Australian Government, states Natural Resource Management (NRM) bodies for resource condition reporting. same products identifying areas Caring our Country (C4oC) investments. Modelled monthly annual maps Australia at 50-km resolution period July 2006 June 2008 were compiled using Computational Environmental System model (CEMSYS). CEMSYS comprises an atmospheric model, a land surface transport deposition database. It uses analysis data from National Centre Prediction, USA (NCEP) calculate properties like fields, rainfall, radiation clouds. Geographic Information Systems (GIS) are describe soils vegetation satellite is ground cover levels. In this study, described horizontal soil flux (TQ mg/m/s) output CEMSYS. TQ representative average amount moved within pixel each month. To aid with reporting, modified map NRM regions subregions was developed expert panel status. erosion, expressed as five classes (very low very high), calculated subregion, region, state continent 24 months. In 2007–08 dust-year 11% high classes. This compares 9% 2006–07 dust-year; however, 2% yearly difference not statistically significant. largest (high classes) tend focused arid semi-arid rangelands south-western Queensland, western NSW, north-central north-eastern South Western agricultural lands eastern West also had class erosion. Notably, non-agricultural Australia, northern Northern Territory all have levels. The regions, state, highest amounts (TQ) were:  Desert Channels, West, Border Rivers Maranoa–Balonne, Condamine Queensland  Western, Rivers–Gwydir, Namoi, Lachlan, Murrumbidgee, Murray Lower Murray–Darling NSW  the Arid Lands Yorke Australia  the Pastoral Non-Pastoral Territory, and  the Rangeland Goldfields Nullarbor, Gascoyne Murchison mixed farming Agricultural Avon Australia. The outputs produced study offer greater temporal (monthly) spatial (50 km) better statistical descriptions than previous measure Dust Storm Index (DSI). DSI derived Bureau Meteorology observer 110 sites time steps. Maps created interpolating between sites. Despite limitations DSI, it still remains valuable cross validation has major advantage longer series (1960 present). With exception data, there lack test against 50 10-km scales. Roadside survey NSW one- hectare scale determine therefore model. Future testing planned after compilation DustWatch roadside other (South Australia). Model appear most reliable south-eastern rangelands. Outputs savannas grasslands seem less due accuracy product’s ability detect dead or senescing vegetation. error estimates results over-prediction Under-prediction winter cropping southern noted possibly relates simplification types model’s performance will improved projects underway address these issues. The issue scaling needs appreciated project shows problems comparing measured different scales (10 loss precision km. Larger pixels involve averaging underlying information such likelihood values reduced. Therefore, chance locating levels scalded river margins along Edward Wakool end catchment management area classified ‘High’ ‘Very High’ ‘Low’ ‘Moderate’ resolution. With priority would been identified. This concludes implementation plan proposes improvements modelled through provision (2000 present) increasing 10