Variability of Rainfall Areal Reduction Factors for a Coastal City: A Case Study of Shenzhen, China

Areal reduction factors (ARFs) are widely used to transform the point rainfall intensity to the areal rainfall intensity in engineering practice. Inappropriate ARFs may result in an overestimate or underestimate of the areal rainfall and consequently lead to the inappropriate design of infrastructure. This study aims to explore the differences in ARFs estimated by four empirical methods and quantitatively analyze the effect of rainfall duration, area, return period, local topography, and rain gauge density on ARFs in the coastal city Shenzhen, China. The results indicate that the original fixed-area method yields more conservative (higher) ARF estimates than the other three methods, which also consider the return period with the coefficient of variation ranging from 0.014 to 0.054. Bell's method and its modified versions produced modest discrepancies in ARFs, with coefficient of variation (COV) values ranging from 0.008 to 0.023. A declining trend of ARFs with increasing return period was observed for six durations (1, 2, 3, 6, 36, and 48 h), whereas ARFs tended to increase with increasing return period for 12- and 24-h durations. Meanwhile, ARFs in mountainous areas (the east part of Shenzhen) were lower than that in the flat terrain in the west part with a maximum reduction of 0.13, which might be associated with the higher spatial variability of rainfall caused by the terrain effect. In addition, ARFs derived from the sparse rain gauge network may be overestimated compared with that from the dense network (maximum overestimation of 0.041). This study provides new insights into the relationship between ARFs and return periods, and highlights that ARFs should be further studied based on the up-to-date rainfall data to tackle the changing climate.