Drought conditions of some magnitude are present almost every year in some part of Australia because of its vast size and semiarid to arid climate. Such occurrences are a part of normal life and are not of major concern at the national level. Sustained droughts, usually lasting one to two years, possibly for three years, and extending across large tracts of the country have created great disasters. These are of relatively less frequent occurrence, and each of them has different spatial, duration, and intensity characteristics. When drought conditions are so intense and protracted that they are beyond those that can reasonably be factored into normal risk management strategies, they are termed drought exceptional circumstances (Lembit, 1995). In practice, this is a drought of such rarity and severity that it occurs no more than once in every 20 to 25 years and is more than 12 months in duration (Clark et al., 2000; Dixon, 1995).
The National Drought Policy has laid out a process with a framework for the determination of drought exceptional circumstances and a set of six core criteria to be taken into account by both the commonwealth and the states in consideration of drought exceptional circumstances declarations (Lembit, 1995). The six core criteria are
1. meteorological conditions,
2. agronomic and stock conditions,
3. water supplies,
4. environmental impacts,
5. farm income levels, and
Drought exceptional circumstances are indicated when the combined impact on farmers of the core criteria is a rare and severe occurrence. Meteorological conditions are the threshold or primary condition for exceptional circumstances but should be assessed in terms of "effective rainfall." The threshold conditions would involve a "rare and severe event"; rare being one in twenty years, and severe being either more than twelve months or at least three consecutive failed seasons depending on the nature of the production system being considered (Queensland Department of Primary Industries, 1995).
Assessment will go further if the criterion of meteorological conditions is satisfied. The remaining criteria should collectively indicate drought exceptional circumstances. The criteria are used together to form an overall judgment on exceptional drought circumstances. A similar process must be followed for the revocation of drought exceptional circumstances.
White (1997) has given a summary of the indices that are presently used for the assessment of drought exceptional circumstances by the commonwealth and state and territory governments in Australia. Although none of the major indices is superior to the rest in all circumstances, some indices are better suited than others are for certain uses. Some of the methods used or with potential for use are summarized here.
Rainfall is the main criterion for assessing drought exceptional circumstances. Analysis based on rainfall data averaged over a meteorological division has a drawback of not necessarily coinciding with the area of interest with respect to a drought event. Analysis of individual rainfall stations selected to represent the region of interest is more useful, as it could give a quick indication of the most affected areas and where boundaries might lie. Several indices measure how much rainfall for a given period of time has deviated from a historically established normal.
Statistical techniques for the analysis of drought events based on historical rainfall records of individual stations using commonly available spreadsheet packages operating on desktop computers are suggested (Bedo, 1997). These analyses complement more sophisticated approaches available only to specialists. A series of Microsoft Excel macros, which analyze rainfall in several ways to test the meteorological criteria, are used. These macros provide three techniques of rainfall analysis to identify an exceptional circumstances event.
1. For visual checking of individual monthly rainfall values and confirming patterns in the past, decile and percentile values for every month of the historical record are calculated and presented in a tabular form.
2. The cumulative rainfall anomaly is calculated and plotted for any period within the historical records of a rainfall station. For example, anomalies can be calculated for periods restricted to agriculturally important seasons. Plotting the cumulative rainfall anomaly for the en tire historical record provides useful indications of past exceptional events.
3. Analysis of rainfall for seasonal periods appropriate to different agricultural regions and farming systems is possible. The monthly periods can be selected to correspond to winter or summer rainfall climates or split to represent autumn and spring within a calendar year or a cool and warm season extending over two calendar years. Rainfall totals are calculated for the selected seasons over the historical record, and the percentile (or decile) ranking is determined. The macro then scans the result table and marks those seasons that qualify under the current criterion of three consecutive seasons at or below a critical percentile. The critical percentile value is adjusted so that the number of events over the historical record occurs about one in 20 to 25 years.
Stephens (1997) proposed a Drought Exceptional Circumstances Index (DECI) as a criterion for defining exceptional drought in cropping areas. The index is based on long-term rainfall records of stations spread across the wheat belt, representing the major agrometeorological zones. For each region, he defined a cropping year, which covered the essential period of soil moisture accumulation and crop growth. This began on October 1 in the previous year and ended when rainfall stopped contributing to wheat yield. A five-step approach is used to derive the index:
1. Long-term wheat yields were calculated assuming no change in technology for all years of available rainfall data. He used a yield forecasting model (STIN) based on a "moisture stress index" (Stephens, 1996).
2. Growing season rainfall was added and ranked as percentiles. Abnormal years were discarded.
3. Relative winter wheat yields were combined with relative mean soil moisture to form a Yearly Productivity Index (YPI). This index ranges between 0 and 1.
4. Individual yearly yields must be in the lowest 30 percent of values for conditions to be exceptional. For a two-year interval, both yields must be below this cutoff point, whereas three- or four-year intervals should allow for one year with yields in the 30 to 40 percent range, before conditions fall back into exceptional circumstances again.
5. Individual years that qualify were ranked in ascending order of severity on the basis of four-, three-, and two-year mean yields to identify the worst years. Drought duration and severity were integrated with a summation of yield (YPI) deviations below a 40 percent value. This summation is called the Drought Exceptional Circumstance Index.
This method is claimed to be more responsive to the ground situation than many other empirical, agroclimatic, and simulation models (Brook, 1996).
The primary argument for simulation of system performance is that meteorological conditions alone do not easily capture the true state of the agricultural system. Rainfall at one time of the year can be carried over under fallows to be used at other times of the year. Failure of planting rains at a critical time may downgrade otherwise average seasonal rainfall conditions in terms of production potential. A crop-soil management system simulation model has the potential to integrate the meteorological and agricultural dimensions of the production system.
Pasture simulation models: McKeon (1997) has advocated the use of the National Drought Alert Strategic Information System for assessing the events of drought exceptional circumstances. The National Drought Alert Strategic Information System is a good combination of rainfall analysis, seasonal climate forecasts, satellite and terrestrial monitoring, and simulation models of meaningful biological processes.
The core simulation model used in the National Drought Alert Strategic System is GRASP (GRASs Production), which has been thoroughly validated in Queensland (Carter and Brook, 1996). GRASP produces estimates of pasture growth, biomass in green and dead pools, green cover, soil moisture, animal liveweight gain, and pasture utilization on a daily basis and can be run forward up to 180 days into the future. When pasture production is combined with stock estimates, calculations of the degree of pasture utilization can be made and displayed as maps of feed availability and land condition, with a resolution of a quarter to half a shire. These maps form a core product of the strategic information system for assessing drought exceptional circumstances.
The National Drought Alert Strategic Information System produces per-centile views of meaningful biological and agricultural variables. So it is possible to construct a percentile view of grassland production and condition that is more aligned with the actual extent and severity of drought than are rainfall percentile maps. A particular month's or season's grass biomass can be compared to the last 30 years or 100 years of biomass that would have existed at that location.
Smith and McKeon (1997) used the simulation models for assessing the historical frequency of drought events on rangelands. They analyzed the results in terms of various measures that could be used to identify an exceptional circumstances drought event on the basis of those occurring once in 20 years in the long term.
Another pasture simulation model that has been advocated to identify drought is GrassGro (Donnelly and Freer, 1997). GrassGro models pasture and animal production which in turn identify severe drought conditions. With the GrassGro model, simulation results are tabulated for monthly rainfall, weight of green herbage available, and weight of supplementary feed required to maintain the stock. In the pasturelands of southern Australia, severe droughts identified with GrassGro are more realistic than those identified with rainfall alone.
Crop simulation models: Keating, Meinke, and Dimes (1997) explored the potential role for crop simulation models, such as APSIM (Agricultural Production System Simulation Model), to assist in the objective assessment of drought. The study concluded that the dynamic simulation of agricultural systems has much to offer to the objective identification of drought exceptional circumstances. This does not mean that other, simpler methods which relate crop performance to weather could not achieve similar results. It should be possible to combine the various models on the strengths and weaknesses of the alternative approaches.
Satellite data can be directly related to land cover (vegetation and soil) status or functioning. Remotely sensed data are unsurpassed in supporting the formulation of drought indicators because they are actual observations of landscape status and its performance. Obtaining a time series of remotely sensed images allows information to be extracted regarding the location and duration of below-average biomass and below-average soil moisture (Smith, 1996). Normalized Difference Vegetation Index (NDVI) data are used to monitor vegetation health and to fine tune regional differences. Remotely sensed data (visible, thermal, etc.), geographic information system (GIS) data layers (soils, geology, etc.), and point-based measurements (climate, biomass, etc.) all have space and time dimensions and can be integrated for a better appreciation of the environment. This information can then be combined with other necessary information, such as agronomic, economic and social data, which allow drought exceptional circumstances to be determined objectively (McVicar, 1997).
Several federal and state departments and organizations use remotely sensed data to assess the seasonal quality and quantity of agricultural production for temporal comparisons of environmental conditions for their specific uses (Graetz, 1997; Cridland, 1997; McVicar et al., 1997). The topic is further discussed in Chapter 7.
OVERVIEW OF DROUGHT ASSESSMENT METHODS Rainfall Data Analysis
Drought is a consequence of rainfall deficiency in relation to potential water loss through evaporation. Analysis of rainfall is therefore the primary basis of identification of drought. However, a number of pitfalls have to be kept in mind while depending on rainfall data for assessing drought.
Rainfall measurements are at points that are often widely separated. Maps of rainfall deficiency or surplus drawn on the basis of point measurements are frequently far from the reality of the conditions some distance from these points. Rainfall occurring at one time of the year can be carried over under fallow to be used at other times of the year, making average monthly rainfall values irrelevant. Results from various studies suggest that ranking the year according to rainfall may be quite different to ranking the year from simulated pasture growth. Failure of rain at the optimum time may downgrade the otherwise average rainfall conditions in terms of production potential. Average rainfall conditions mask the influence of rainfall intensity and rainy-spell duration on the actual performance of crops and pastures.
An examination of temporal rainfall records in Australia at some locations has shown a tendency toward higher rainfall in the second half of this century. If this observation turns out to be true, then the emerging pattern is likely to be strengthened further under the global warming scenario and will have a significant impact on the identification and temporal comparisons of droughts on the basis of severity and impacts. By the present definition, there may not be any drought exceptional circumstances event in the near future to be compared with those that occurred in first half of the twentieth century.
Simulation models hold great promise in objective identification of drought and drought exceptional circumstances because they hold the potential to integrate the climate and farming dimensions of production sys tems. At the same time, simulation models are not capable now, nor will they be in the near future, of replacing other measures of drought identification.
Any minor wrong information about plant characteristics or soil parameters included in running the simulation can greatly influence the output of the model. Efforts to overcome this shortcoming need to be balanced against the expected gains over the simpler models.
The majority of the models can easily identify a major drought event, but they differ considerably in highlighting marginal events. Some models have a tendency to amplify minor events, while at other times major events are presented in a suppressed form. The differing results can create a confusing situation.
Modeling living systems is very complex. Models available at present are capable of giving outputs only approximating reality.
Satellite-derived images are useful in broad-scale assessment of greenness of ground cover, especially the vegetation response to a rainfall event. Another major value of remote sensing information is for spatial and temporal validation of the simulation models. However, remote sensing has inherent limitations in providing a total solution to drought monitoring.
Images from satellites give poor information on biomass, and tree cover confounds the signals. Interpretations of the images do not consider the effects of vegetation structure and cover. Remotely sensed data products currently available contain a lot of noise from the instruments onboard and the atmosphere. This noise is of sufficient magnitude to disqualify remotely sensed products for use in drought analysis, and if used unhindered, the state-of-the-art products may not stand up to the test of law courts.
Data availability from remote sensing satellites is not guaranteed because the majority of the countries in the world purchase data from foreign sources. Furthermore, a failure of systems onboard a satellite at a critical time during the drought season may render the previous satellite information completely useless. The remotely sensed data are currently available for less than two decades. This is too short a period to identify an exceptional circumstance event on a temporal scale. No future projection of the intensity and magnitude of the event is possible through remote sensing measurements. These limitations of remotely sensed information suggest that it can not be used exclusively in defining a drought or drought exceptional circumstances. It is a supplement to other measures of the event.
This leads to the conclusion that a combination of methods—rainfall analysis, crop and pasture simulation, and monitoring the health of vegetation through remote sensing—supported with field surveys is the most realistic approach to assess the extent and intensity of drought.
MEETING THE CHALLENGE: A DROUGHT MITIGATION PLAN
Effective drought mitigation should be based on a comprehensive view of drought, because drought is not simply a deficiency of rainfall but is a more complex phenomenon that influences the whole society. Strategies to minimize the impact of drought at a farm scale are different from those needed at the state or national level. Strategies normally adopted at the farm level are based on local experience. Some of these are discussed in Chapter 10. Combating drought at the national or state level is a three-stage process. The first stage is monitoring the drought development in terms of spread and intensity as realistically as possible. In the second stage, the monitored information is used as an early warning system. Activation of a readily available drought mitigation plan is the third step of the process (Anonymous, 2000; National Drought Mitigation Center, 1996d).
Three groups of people are the key players in tackling a drought situation. In the first group are climatologists and others who monitor how much water is available now and in the foreseeable future. The second group includes natural resource managers and others who determine how the lack of water is affecting various interests, such as agriculture, municipal supplies, and recreation. The third group of people is comprised of high-level decision makers who have the authority to act on information they receive about water availability and the drought's effects. The major challenge in successful drought planning is bringing together all these groups on a platform to communicate effectively with one another.
In the United States, a systematic plan is suggested for drought management. The plan is referred as "10 Steps to Drought Preparedness" (National Drought Mitigation Center, 1996a). Some salient points of this plan described in this section can serve as a model and could be adopted by other countries/regions with modifications and alterations as deemed necessary (Wilhite, 2001).
Creating a task force is the first step of the drought mitigation plan. The task force has two purposes. First, during plan development, it will super vise and coordinate the development of the plan. Second, after the plan is implemented and activated during times of drought, the task force will assume the role of policy coordinator, reviewing and recommending alternative policy options.
The task force includes representatives from the most relevant agencies within government and from universities. The composition of the task force recognizes the multidisciplinary nature of drought and its impacts. It may also include a representative of the media in an advisory capacity to ensure public awareness of drought severity and the actions implemented by government.
Drought Policy and the Plan's Purpose and Objectives
The drought task force develops a drought policy that specifies the general purpose for the drought plan. State officials consider many questions as they define the purpose of the plan. These include the purpose and role of state government in drought mitigation efforts; the scope of the plan; the most drought-prone areas of the state; and the most vulnerable sectors of the state's economy. It also includes the role of the plan in resolving conflict between water users during periods of shortage; the resources (human and economic) that the state is willing to commit to the planning process; the legal and social implications of the plan; and the principal environmental concerns caused by drought. Answers to these and other questions help to determine the objectives of drought policy.
Political, social, and economic values are bound to clash as competition for scarce water resources intensifies during a drought. To lessen conflict and develop satisfactory solutions, it is essential that the views of citizens, the public, and environmental interest groups be considered early in the drought planning process. In fact, these groups are likely to impede progress in the development of plans if they are not included in the process. Local groups could be set up to bring neighbors together to discuss their water-use problems and seek cooperative solutions.
The drought task force should prepare an inventory of resources and constraints that might enhance or inhibit fulfilment of the objectives of the plan ning process. Resources include natural resources, human expertise, infrastructure, and capital available to the government.
The most obvious natural resource of importance is water and its location, accessibility, and quality. Biological resources refer to the quantity and quality of grasslands, rangelands, forests, and wildlife. Human resources include the labor needed to develop water sources, lay pipeline, haul water and hay, process citizen complaints, provide technical assistance, and direct citizens to available services.
Financial and legal constraints are likely to emerge during a drought. Financial constraints include costs such as hauling water or hay and new program or data collection costs. These costs must be weighed against the losses that may result from not having a drought plan. Legal constraints include water rights, methods available to control usage, the kinds of public trust laws in existence, requirements for contingency plans for water suppliers, and the emergency and other powers of state agencies during water shortages.
A drought plan has three primary organizational tasks: monitoring, impact assessment, and response and mitigation. Each task is assigned to a separate group or a committee, but the groups need to work together well, with established communication channels.
The monitoring committee includes representatives from agencies with responsibilities for forecasting and monitoring the principal meteorological, hydrological, and agricultural indicators. The monitoring committee meets regularly, beginning in advance of the peak demand season. Following each meeting, reports are prepared and disseminated to the state's drought task force, relevant state and federal agencies, and the media. The committee ensures that accurate and frequent news bulletins are issued to the public to explain changing conditions and complex problems.
Drought impacts cut across economic sectors. An impact assessment committee represents those economic sectors most likely to be affected by drought. The impact committee considers both direct and indirect losses as drought effects ripple through the economy. It is responsible for determining impacts by drawing information from all available reliable sources. Unfortunately, the quantification of drought impacts is very complicated, and some impacts may be so subtle that detection is very difficult. Working groups composed of specialists in each impact sector are created for this purpose.
The drought task force, or a similar group of senior-level officials, acts on the information and recommendations of the impact assessment committee and evaluates the state and federal programs available to assist agricultural producers, municipalities, and others during times of emergency. During periods of severe drought, the committee makes recommendations to the government about specific actions that need to be taken.
The policymaker's understanding of the scientific issues and technical constraints involved in addressing problems associated with drought is often negligible. Likewise, scientists generally have a poor understanding of existing policy constraints for responding to the impacts of drought. Communication and understanding between the science and policy communities are poorly developed and must be enhanced if the planning process is to be successful. Direct and extensive contact is required between the two groups to distinguish what is feasible from what is desirable for a broad range of science and policy issues.
The drought plan is unveiled and presented to the public in a way that gives maximum visibility to the program and credit to the agencies and organizations that have a role in its operation. For purposes of gaining publicity and attention, it is a a good idea to announce and implement the plan just before the most drought-sensitive season. The cooperation of the media is essential to publicizing the plan. A representative of the media on the drought task force is a valuable resource in carrying out the publicity.
The drought task force initiates an information program aimed at educating the general population about drought and drought management and what individuals can do to conserve water in the short run. Educational programs are long term in design, concentrating on achieving a better understanding of water conservation issues for all age groups and economic sectors. Without such programs, governmental and public interest in water conservation vanish as soon as the drought is over.
Periodic evaluation and updating of the drought plan is essential to keep the plan responsive to the state's needs. To maximize the effectiveness of the system, two modes of evaluation are in place: (1) An ongoing or operational evaluation keeps track of how social changes such as new technology, new research, new laws, and changes in political leadership may affect the drought plan. (2) A postdrought evaluation of the plan teaches lessons from past successes and mistakes. Postdrought evaluation documents the assessment and response actions of government, nongovernmental organizations, and others and implements recommendations for improving the system. Attention is focused not only on those situations in which coping mechanisms failed but also on the areas in which the success achieved has been exemplary. Evaluations of previous responses to severe drought are a good planning aid.
Research needs and institutional gaps become apparent during drought planning and plan evaluation. The drought task force compiles those deficiencies and makes recommendations on how to remedy them to the relevant state agencies.
Was this article helpful?