Detecting the impact of land cover change on observed rainfall.

Chun Xia Liang,Floris F Van Ogtrop,R Willem Vervoort

doi:10.7717/peerj.7523

Abstract

Analysis of observational data to pinpoint impact of land cover change on local rainfall is difficult due to multiple environmental factors that cannot be strictly controlled. In this study we use a statistical approach to identify the relationship between removal of tree cover and rainfall with data from best available sources for two large areas in Australia. Gridded rainfall data between 1979 and 2015 was used for the areas, while large scale (exogenous) effects were represented by mean rainfall across a much larger area and climatic indicators, such as Southern Oscillation Index and Indian Ocean Dipole. Both generalised additive modelling and step trend tests were used for the analysis. For a region in south central Queensland, the reported change in tree clearing between 2002–2005 did not result in strong statistically significant precipitation changes. On the other hand, results from a bushfire affected region on the border of New South Wales and Victoria suggest significant changes in the rainfall due to changes in tree cover. This indicates the method works better when an abrupt change in the data can be clearly identified. The results from the step trend test also mainly identified a positive relationship between the tree cover and the rainfall at p < 0.1 at the NSW/Victoria region. High rainfall variability and possible regrowth could have impacted the results in the Queensland region.

Highlights

Land use and land cover changes can lead to changes in the local climate
The pixels, where the tree cover change based on the MOD44B data was significant (p ≤ 0.05) in each study region, are shown in Fig. 5 for the QLD region (Fig. 5A) and the New South Wales and Victoria (NSW/VIC) region (Fig. 5B)
In the NSW/VIC region, much of the tree loss between 2002 and 2003 was concentrated in the Snowy Mountains close to the border of NSW and VIC, which is evident in the figure

Summary

Introduction

Land use and land cover changes can lead to changes in the local climate. Empirical and modelling studies have found that cloud types and rainfall are correlated to large scale vegetation cover changes, such as deforestation in the Amazon and in the Sahel (Chagnon & Bras, 2005; Wang et al, 2009; Pinto et al, 2009; Mei & Wang, 2010; Kucharski, Zeng & Kalnay, 2013; Pitman & Lorenz, 2016) and afforestation in south Israel (Otterman et al, 1990; Ben-Gai et al, 1998). Using airborne measurements in Western Australia, Junkermann et al (2009) showed a significantly higher level of aerosols over an agricultural area compared to an adjacent area with natural vegetation. They suggested that a modification of aerosol concentrations due to deforestation could have contributed to a reduction of local rainfall, as more, but smaller rain droplets were observed. Nair et al (2011) reported from the Bunny Fence Experiment in Western Australia that local land use change altered the synoptic west coast trough dynamics and surface roughness, and. Detecting the impact of land cover change on observed rainfall. Maximum temperatures were found to be sensitive to land cover change in eastern Australia (McAlpine et al, 2007)

Objectives

Methods

Results

Discussion

Conclusion