Operational Drought Monitoring in Australia

Several levels of drought monitoring exist in Australia. The Bureau of Meteorology (BoM) monitors drought across the nation from a meteorolog ical perspective. The climate monitoring network, which is the basis for BoM's drought assessment, also serves as input to a national monitoring system operated by several state government agencies. These systems are described later in this chapter. Individual state governments also operate official drought declaration schemes that involve monitoring emerging drought situations at both individual property and district scales. Such monitoring is largely subjective and based, in the first instance, on the applications for government assistance by individual property owners. The process used in Queensland since 1964 is outlined below as an example.

Drought Monitoring at a State Level: The Queensland Example

The Queensland drought scheme primarily aims to assist primary producers with livestock management during drought. The process for declaring drought in Queensland begins with an individual primary producer approaching the local stock inspector for an "individual droughted property" (IDP) declaration. Following a property inspection by the stock inspector, a local drought committee assesses the application. If drought conditions appear widespread and a significant number of IDPs have been issued, the local drought committee may recommend that the entire district (shire or part thereof) be declared drought stricken. The state government reviews the application and, until policy changes in the early 1990s, would declare drought if summer (October-March) rainfall at official recording stations in the district was 40% below the long-term average rainfall or, for southern Queensland, if winter (April-September) rainfall was 50% below the long-term average (Daly, 1994). Since the early 1990s, drought has been defined as a once in 10- to 15-year event. From this definition, a threshold of annual rainfall below the 7th to 10th percentile has been adopted as a declaration criterion. This threshold equates to an annual rainfall of approximately 40% below the long-term average and is therefore similar to thresholds used before 1964. Drought status is revoked once it is ascertained that stock numbers can return to normal levels. Because the revocation threshold is higher than the declaration threshold, official droughts have occurred more than 10% of the time.

Rainfall and Climate Monitoring

BoM monitors rainfall across Australia through a network of approximately 7000 stations operated largely by volunteers. The National Climate Center (NCC) receives digital data via phone lines each day from approximately 300 automatic weather stations and 1500 observers. This telegraphic network provides the basis for near-real-time assessment of rainfall and climate. The remaining stations report via regular mail on a monthly basis. These written records are checked and digitized at the NCC before being placed on a national database. Data from most stations are available in digital format within three months. Maps of rainfall anomalies for various periods from 24 h to more than 12 months are updated in near-real time (http://www.bom.gov.au/cgi-bin/climate/rainmaps.cgi). Rainfall anomalies are mapped as percentiles, as absolute or percentage departures from the long-term average, and as drought severity levels.

An accepted approach to monitoring meteorological drought in Australia is to rank, as a percentile, current rainfall against historical rainfall:

Percentile (0-100) = [(r — 1)/(n — 1)] 100 [29.1]

where r is a rank in a series of n values. BoM's drought analysis indicates regions in which rainfall has been below the 10th percentile (a serious rainfall deficiency) or the fifth percentile (a severe rainfall deficiency) for three months or more. The rainfall deficiency is considered removed if, for the past month alone, rainfall exceeds the 30th percentile for the three-month period commencing that month or if, for the past three months, rainfall is above the 70th percentile. This information is presented both on the Web (above) and as a monthly publication (Drought Review-Australia).

Use of Agronomic Models to Monitor Droughts

Anecdotal experience of graziers and the results from experimental trials show that, for different years, different amounts of pasture growth are experienced with similar amounts of seasonal rainfall (Stafford Smith and McKeon, 1998). Likewise, the relationship between crop yield and rainfall during the growing season is generally poor because yield also depends greatly on the timing of rain as shown, for example, by Nix and Fitzpatrick (1969). Thus, agronomic models are probably the best indicators of drought severity and the effectiveness of rain (White et al., 1998). The following section describes the implementation of a national-scale framework for modeling pasture and crop production in near-real time and the adaptation of model output for drought monitoring and early warning.

0 0

Post a comment