Regional climate change projections for the Tully sugar region

Sexton, J., Everingham, Y., and Skocaj, D. (2013) Regional climate change projections for the Tully sugar region. In: Proceedings of the 20th International Congress on Modelling and Simulation. pp. 2792-2798. From: MODSIM2013: 20th International Congress on Modelling and Simulation, 1-6 December 2013, Adelaide, SA, Australia.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: http://www.mssanz.org.au/modsim2013/L11/...
 
3


Abstract

Faced with the challenges of a changing climate, it is imperative that primary industry decision makers have access to climate projections on a local scale. In the Australian sugar industry, changes in maximum and minimum temperature, radiation and rainfall can significantly affect future economic and environmental sustainability. In northern regions an increase in rainfall during the harvest season can lead to crop standover and lower economic returns. Modelling the effects of climate change at a local scale will help develop regional management strategies.

General Circulation Models (GCMs) allow researchers to explore projections of climate variables under a range of possible future emission scenarios. However, current outputs from GCMs are often only available at a coarse resolution (up to 300 km by 300 km). As a result many small scale conditions that can affect climate variables are often not represented in GCM outputs. Sugarcane in Australia is primarily grown in a narrow band on the eastern coast. The Tully region of northern Queensland is situated between the World Heritage Listed Wet Tropics Rainforest and Great Barrier Reef. Topographical changes across the Tully sugarcanegrowing region result in climatology varying spatially. Downscaled GCMs can provide a bridge between large scale climate projections and the need for local climate variables. This paper explores climate change projections for maximum and minimum temperature, radiation and rainfall in the Tully sugarcane-growing region at a high spatial resolution.

Temperature and rainfall data were obtained from 11 GCMs for the period 1961 to 2000. Projections for this period were based on 20th century forcings (20C3M) as described by the International Panel of Climate Change (IPCC). GCM projections for the period 2046 to 2065 were also obtained, based on a high emissions scenario (A2). A statistical downscaling methodology was used to downscale daily temperature and rainfall data on a 0.05 by 0.05 decimal degree grid (approximately 5 km by 5 km). Data were downscaled for grid locations known to grow sugarcane within the Tully region. Daily radiation data were not available using the downscaling process. Instead, daily radiation data were generated from downscaled rainfall and temperature data. Equations for daily radiation were parameterised using temperature and rainfall data from a nearby high quality weather station and calculated total solar flux. Estimates were bias corrected to replicate weather station records of radiation.

The projected change in each of the four climate variables was assessed on a regional level for the 11 GCMs and spatial variations within the region were identified. For temperature and radiation variables, the absolute projected change was calculated. For rainfall the projected relative change was calculated. The relative change was defined as the percent change from the baseline period (1961 to 2000). Projected changes were analysed at each grid point for summer, autumn, winter and spring. The regional mean change was calculated for each GCM and a 95% bootstrapped confidence interval was produced for the 25th, 50th and 75th percentiles of the paired differences. The percentiles of projected change across the range of GCMs were used to capture the uncertainty between model projections. If the 95% confidence interval of the 50th and 75th (25th) percentile captured only positive (negative) values, an increase (decrease) was considered plausible. An increase (decrease) was considered highly plausible if the confidence intervals for the 25th 50th and 75th all captured positive (negative) values.

For the Tully region, an increase in temperature was considered highly plausible for all seasons under A2 simulated forcings. A projected decrease in radiation was considered highly plausible for winter and spring and plausible for autumn. A relative increase in radiation was plausible for summer and a relative increase in rainfall was considered plausible for spring. However, projected changes in seasonal rainfall varied spatially within the Tully region. An increase in seasonal temperature and rainfall may require changes to traditional management strategies. High resolution climate change projections can help industry decision makers develop localised, robust adaptation strategies. Local adaptations are vital for an economically and environmentally sustainable future.

Item ID: 32642
Item Type: Conference Item (Research - E1)
ISBN: 978-0-9872143-3-1
Keywords: temperature; radiation; rainfall; GCM; downscale
Related URLs:
Funders: Sugar Research and Development Corporation (SRDC)
Projects and Grants: Sugar Research and Development Corporation (JCU032).
Date Deposited: 16 Apr 2014 02:17
FoR Codes: 01 MATHEMATICAL SCIENCES > 0104 Statistics > 010401 Applied Statistics @ 100%
SEO Codes: 82 PLANT PRODUCTION AND PLANT PRIMARY PRODUCTS > 8203 Industrial Crops > 820304 Sugar @ 100%
Downloads: Total: 3
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page