What a '"drought" is probably depends on where you live. I am told that in western Tasmania a drought occurs when there has been no rain for three days. Our recent unusual (continuing?) drought arose from a record low rainfall over a 6 month period but there was some rain. In northern Northern territory a 6 month drought with no rain is commonplace even though the average yearly rainfall can be similar to ours.
Graziers value pasture which can be a "fuel" for animals or for fires. The growth and curing of pastures, however, depends on soil moisture which is affected by rainfall, but not only by rainfall - the focus of many a drought definition. Evaporation, soil storage capacity, intercepting vegetation and drainage all play a part in affecting the water supply of the pasture. In fact, the performance of the pasture is affected by air-water (humidity) as well as soil-water. We get around problems of the prediction of the effects of soil moisture or "drought' on pasture by measuring curing and using the value directly in our predictions of fire rates of spread. Predictions of curing are useful if we have few archived data but want to go back in time to compare seasonal conditions.
In forests, we are unable to see the effects of drought as easily as we can in pastures. Therefore, people have used a Drought Index (and Drought Factor) to gauge the effect of drought on fires. The drought indexes commonly used in Australia are the Keetch-Byram Drought Index (Keetch and Byram 1968) and the Soil Dryness Index (Mount 1972). These are basically predictors of the deficiency of soil moisture in a hypothetical soil profile having 200mm (8 inch) capacity. The Soil Dryness Index seems to be the better of the two predictors of soil moisture but, even so, requires calibration to be at its most accurate.
Predicting soil moisture is all very well but how does soil moisture affects fires? It doesn't, but it may affect litter fall, little moisture, large-fuel moisture and live-fuel moisture. Various tests have been conducted, I recall, to see what the relationship is between soil-dryness indexes (or soil moistures) and fuel moistures (e.g Pook & Gill 1993) but the results have been mixed.
Perhaps the main effect indicated by Drought Indexes is nothing to do with fuel moisture at a point above a soil profile but is related to the continuity of dry fuels across slopes and valleys? In a severe drought the whole landscape is connected as far as a fire is concerned whereas in a moderate drought there are some 'green' drainages that stop fires.
When we conducted a study on the effects of drought on weed movement in fodder some years ago the definition of drought used by the NSW Department of Agriculture was "drought is deemed to exist when paddock conditions are such that there is insufficient grazing or water available to sustain sheep or cattle." (Thomas et al. 1984). Ironically, this definition means that a "drought" could occur when a property was subject to a fire, or even a flood, to the extent that there was insufficient grazing available to sustain stock. This definition is now "old hat" and "gathering dust" as far as significant declarations of drought are concerned.
Definition of drought is difficult and important. What the definition is depends on the data available, the values at stake and the financial implications of any "declaration". The more direct the definition is in relation to the problem of concern the better it is likely to be. Untangling the mechanisms behind the effects of drought on fire behaviour may take some time.Literature cited:
Division of Plant Industry, CSIRO
23 April 1998