An assessment of a daily rainfall erosivity model for New South Wales

A rainfall erosivity model using daily rainfall amounts to estimate rainfall erosivity was tested for 29 sites in New South Wales to see whether such a model could adequately describe the temporal variation and seasonal distribution of rainfall erosivity. The coefficient of determination varied from 0.57 to 0.97 and the average discrepancy between actual and estimated seasonal distribution was no more than 3%. A set of parameter values for sites without pluviograph data was recommended for New South Wales. With this set of recommended parameter values, the percentage of total variance explained was decreased to 44%–89% for the 29 sites. Large errors, however, can occur when estimating extreme storm erosivity with large return periods. The daily erosivity model could be used for determining the seasonal distribution of rainfall erosivity or for simulating changes to rainfall erosivity as part of climate change impacts assessment.

Download Full-text

Modelling and mapping rainfall erosivity in New South Wales, Australia

Soil Research ◽

10.1071/sr14188 ◽

2015 ◽

Vol 53 (2) ◽

pp. 178 ◽

Cited By ~ 22

Author(s):

Xihua Yang ◽

Bofu Yu

Keyword(s):

New South ◽

New South Wales ◽

Daily Rainfall ◽

Rainfall Amount ◽

Annual Rainfall ◽

Model Parameters ◽

Rainfall Erosivity ◽

Erosion Risk ◽

South Wales ◽

The Impact

Considerable seasonal and inter-annual changes exist in rainfall amount and intensity in New South Wales (NSW), Australia. These changes are expected to have significant effect on rainfall erosivity and soil erosion by water, but the magnitude of the impact is not well quantified because of the non-linear and dynamic nature of the relationship between rainfall amount and rainfall erosivity. The primary aim of this study was to model spatial and temporal variations in rainfall erosivity and impacts on hillslope erosion across NSW. We developed a daily rainfall erosivity model for NSW to calculate monthly and annual rainfall erosivity values by using gridded daily rainfall data for a continuous 53-year period including a baseline period (1961–90) and a recent period (2000–12). Model parameters were improved based on their geographic locations and elevations to be truly geo-referenced and representative of the regional relationships. Monthly and annual hillslope erosion risk for the same periods was estimated with the Revised Universal Soil Loss Equation. We produced finer scale (100-m) maps of rainfall erosivity and hillslope erosion through spatial interpolation techniques, and implemented the calculation of rainfall erosivity and hillslope erosion in a geographic information system by using automated scripts so that it is fast, repeatable and portable. The modelled rainfall erosivity values were compared with pluviograph calculations and previous studies, and the Nash–Sutcliffe coefficient of efficiency is >0.90. Outcomes from this study provide not only baseline information but also continuous estimates of rainfall erosivity and hillslope erosions allowing better monitoring and mitigation of hillslope erosion risk in NSW.

Download Full-text