Climate Exchange

[

] 46

Supporting decision-making in the

sugar industry with integrated

seasonal climate forecasting

Roger C. Stone, Neil Cliffe, Shahbaz Mushtaq, University of Southern Queensland;

and Yvette Everingham, James Cook University, Australia

n integrated approach has been developed linking seasonal

climate forecasting models to sugar yield, production and

sugar content models in order to improve predictability of

the size of the Queensland sugar crop in any year.

Queensland produces 90 per cent of the Australian sugar crop, most

of it destined for export. In this respect, approximately 32 million

tons of cane and 5 million tons of raw sugar is produced in a ‘normal

climate year’. To produce this amount of sugar there are approxi-

mately 4,000 cane farms, 24 sugar mills and six bulk storage ports,

making export agency Queensland Sugar Limited (QSL) the third-

largest sugar supplier in the world. Climate extremes, especially

excessive rain during the harvesting period (June to November)

can result in massive losses for the entire industry – from the farm

production component through to the milling and especially the

marketing and export components. For example, in 2010, one

component alone of the sugar industry suffered a loss of Au$500m

following excessive rain through the entire harvest period, due to

the development and continuation of the major La Niña event. This

resulted in yield downgrading and inability to harvest many crops

due to wet weather and flooding. Appropriate climate forecasting

systems, especially those that have the capability to be integrated

into sugar yield models and core decision systems, are urgently

required to be developed and included at various stages of the

management cycle in the sugar industry.

QSL requires precise forecasts of total yield, the likelihood of

‘standover cane’ (cane that cannot be harvested) and other likely

disruptions due to weather and climate, especially excessive rain

and lack of potential for dry spells. Other industry sectors are also

closely involved, particularly all the sugar mill owners and opera-

tors in Queensland but also the cane growers themselves and their

farming organizations such as the Queensland Cane Growers’

Council. Direct knowledge of seasonal climate forecasting oppor-

tunities will allow farmers to make better decisions for the coming

seasons about:

• Planting and harvesting

• Farm equipment purchases, which need to be more aligned to

the season ahead – such as the purchase of irrigation equipment

in potentially excessively dry seasons compared with purchase of

tractors with wide tyres in potentially excessively wet seasons

• Scheduling of harvesting operations in potentially wet seasons to

harvest the wet blocks first.

For farming and mill production, the following types

of climate information are used:

• Southern Oscillation Index (SOI)-derived seasonal

forecasting (SOI phases)

for regional-scale forecasts

updated each month on a rolling three-month basis

• Outputs from the new generation of seasonal

forecast outputs locally and internationally, such as

Bureau of Meteorology general statements of likely

El Niño Southern Oscillation (ENSO) conditions,

amd US Climate Prediction Center outputs on

potential for El Niño and La Niña events

• Specific web-based services of the University of

Southern Queensland (USQ), Bureau of Meteorology

and other bodies

• Regular production of targeted climate forecast

newsletters that provide reviews of the various

climate forecast products currently available,

specifically written for local regions and

incorporating ‘farmer jargon’ where possible.

Export and marketing agencies use a fully inte-

grated yield production model incorporating crop

simulation model input (such as the Agricultural

Production Systems Simulator (APSIM) and

Canegrow) integrated with statistical climate fore-

cast systems (SOI phases etc.), but also likely soon

to incorporate aspects of the model code associated

with Global Climate Model seasonal forecast systems

such as Bureau of Meteorology POAMA/ACCESS; UK

Met Office and European Centre for Medium-Range

Weather Forecasts (ECMWF) seasonal forecast

output downscaled to local regions. This type of

output is in research mode only at this stage. Once

deemed suitable for operational use, a direct opera-

tional system may be developed emanating directly

from the climate agency itself.

The climate forecast information issued for

growers and millers is tailored to the extent that

the timing of issue of key aspects of the forecast

is closely aligned with the major decisions being

made across the state both ‘on-farm’ and at the mill

production level:

griculture