Arab J Geosci DOI 10.1007/s12517-012-0707-2

ORIGINAL PAPER

Drought risk assessment using remote sensing and GIS techniques Abdel-Aziz Belal & Hassan R. El-Ramady & Elsayed S. Mohamed & Ahmed M. Saleh

Received: 23 April 2012 / Accepted: 12 October 2012 # Saudi Society for Geosciences 2012

Abstract Beginning with a discussion of drought definitions, this review paper attempts to provide a review of fundamental concepts of drought, classification of droughts, drought indices, and the role of remote sensing and geographic information systems for drought evaluation. Owing to the rise in water demand and looming climate change, recent years have witnessed much focus on global drought scenarios. As a natural hazard, drought is best characterized by multiple climatological and hydrological parameters. An understanding of the relationships between these two sets of parameters is necessary to develop measures for mitigating the impacts of droughts. Droughts are recognized as an environmental disaster and have attracted the attention of environmentalists, ecologists, hydrologists, meteorologists, geologists, and agricultural scientists. Temperatures; high winds; low relative humidity; and timing and characteristics of rains, including distribution of rainy days during crop growing seasons, intensity, and duration of rain, and onset and termination, play a significant role in the occurrence of droughts. In contrast to aridity, which is a permanent feature of climate and is restricted to low rainfall areas, a drought is a temporary aberration. Often, there is confusion between a heat wave and a drought, and the distinction is emphasized between heat wave and drought, noting that a typical time scale associated with a heat wave is on the order of a week, while a drought may persist for months or even years. The combination of a heat wave and a drought has dire socioA.-A. Belal (*) : E. S. Mohamed : A. M. Saleh National Authority for Remote Sensing and Space Sciences (NARSS), 23, Joseph Brows Tito St. Nozha, El-Gedida, P.O. Box 1564 Alf-Maskan, Cairo, Egypt e-mail: [email protected] H. R. El-Ramady Soil Science Department, Faculty of Agriculture, Kafrelsheikh University, 33516, Kafr El-Sheikh, Egypt

economic consequences. Drought risk is a product of a region’s exposure to the natural hazard and its vulnerability to extended periods of water shortage. If nations and regions are to make progress in reducing the serious consequences of drought, they must improve their understanding of the hazard and the factors that influence vulnerability. It is critical for drought-prone regions to better understand their drought climatology (i.e., the probability of drought at different levels of intensity and duration) and establish comprehensive and integrated drought information system that incorporates climate, soil, and water supply factors such as precipitation, temperature, soil moisture, snow pack, reservoir and lake levels, ground water levels, and stream flow. All drought-prone nations should develop national drought policies and preparedness plans that place emphasis on risk management rather than following the traditional approach of crisis management, where the emphasis is on reactive, emergency response measures. Crisis management decreases self-reliance and increases dependence on government and donors. Keywords Agricultural drought . Meteorological drought . MODIS . NDVI . GIS and remote sensing . Drought indices

Remote sensing of droughts Introduction Water scarcity has been frequently occurring these days in many parts of the world, partly because water demand has increased manifold due to the growth in population and expansion of agricultural, energy and industrial sectors, and partly because of climate change and contamination of water supplies. The water scarcity is being further compounded by droughts which affect both surface water and

Arab J Geosci

groundwater resources and can lead to reduced water supply, deteriorated water quality, crop failure, and disturbed riparian habitats. Therefore, understanding drought and modeling its components have drawn attention of ecologists, hydrologists, meteorologists, and agricultural scientists. Droughts are of great importance in water resources planning and management, and for a review of drought concepts, the reader is referred to Mishra and Singh (2010). One third of the world’s population lives in area with water shortages, and 1.1 billion people lack access to safe drinking water. Globally, drought (7.5 %) is the second-most geographically extensive hazard after floods (11 %) of the earth’s land area. The percent of area affected by serious drought has doubled from 1970s to the early 2000 (Nagarajan 2009). In recent years, droughts have been occurring frequently, and their impacts are being aggravated by the rise in water demand and the variability in hydro-meteorological variables due to climate change. As a result, drought hydrology has been receiving much attention. A variety of concepts have been applied to modeling droughts, ranging from simplistic approaches to more complex models. It is important to understand different modeling approaches as well as their advantages and limitations. It is found that there have been significant improvements in modeling droughts over the past three decades. Hybrid models, incorporating largescale climate indices, seem to be promising for long leadtime drought forecasting. Further research is needed to understand the spatio-temporal complexity of droughts under climate change due to changes in spatio-temporal variability of precipitation. Applications of copula-based models for multivariate drought characterization seem to be promising for better drought characterization. Research on decision support systems should be advanced for issuing warnings, assessing risk, and taking precautionary measures, and the effective ways for the flow of information from decision makers to users need to be developed. Finally, some remarks are made regarding the future outlook for drought research (Mishra and Singh 2011). Prolonged multiyear drought has caused significant damages both in the natural environment as well as in the development of the human societies. The annual estimate for the cost of drought in the United States ranges from 6 to 8 billion dollars (Schubert et al. 2007). As the remote sensing technology makes more and more process, with the development of geographic information system (GIS) and Global Positioning System, the real-time monitoring of drought over the large areas can be achieved. Remote sensing and GIS techniques are increasingly being regarded as a useful drought detection techniques, as evidenced by its use across many parts of the world. Some indexes developed from the remote sensing data, such as the Normalized Difference Vegetation Index

(NDVI), Land Surface Temperature (LST), etc., have been used to monitor the agricultural droughts in the relation to the plant growth (Han et al. 2010). Wang et al. (2003) developed the Vegetation Temperature Condition Index (VTCI) based on the information of the LST versus the NDVI scatter plot falling into a triangular shape. It is testified that the VTCI index can be used to effectively monitor the drought of an area in the real time. The autoregressive integrated moving average (ARIMA) model is one of the most widely used time series models. The ARIMA model has several advantages over others, such as moving average, exponential smoothing, neural network, and, in particular, its forecasting capability and its richer information on time-related changes. The objective of this review paper is to provide the reader with useful information about impacts of drought, drought risk evaluation methods using GIS and remote sensing techniques, and finally, drought indices (PDSI, CMI, SPI, SWSI, NDVI, VCI, etc). To accomplish that, the paper reviews the existing literature, summarizes the methods and major findings, investigates the problems and limitations of previous models, and discusses using of remote sensing and GIS for drought risk assessment. Drought definitions Differences in hydrometeorological variables and socioeconomic factors as well as the stochastic nature of water demands in different regions around the world have become an obstacle to having a precise definition of drought. Yevjevich (1967) stated that widely diverse views of drought definitions are one of the principal obstacles to investigations of droughts. When defining a drought, it is important to distinguish between conceptual and operational definitions (Wilhite and Glantz 1987). Conceptual definitions—those stated in relative terms (e.g., a drought is a long, dry period), where as operational definitions, on the other hand, attempt to identify the onset, severity, and termination of drought periods. Generally, operationally defined droughts can be used to analyze drought frequency, severity, and duration for a given return period (Table 1) (Mishra and Singh 2010). A drought is a complex phenomenon that can be defined from several perspectives, and drought definitions are categorized into conceptual (definitions formulated in general terms) and operational (Hisdal and Tallaksena 2003). By studying the above definitions, it can be understood that drought is mainly concerned with the shortage of water which in turn affects availability of food and fodder, thereby leading to displacement and loss to economies as a whole. On the other hand, drought is a normal, recurrent feature of climate, although often erroneously considered an unexpected and extraordinary event. It occurs in virtually all climatic

Arab J Geosci Table 1 Drought and its definitions Definition

Citation

Drought as a sustained period of time without significant rainfall Drought as the smallest annual value of daily stream flow Drought as a significant deviation from the normal hydrologic conditions of an area FAO defines a drought hazard as “the percentage of years when crops fail from the lack of moisture.” Drought means a sustained, extended deficiency in precipitation. Drought means the naturally occurring phenomenon that exists when precipitation has been significantly below normal recorded levels, causing serious hydrological imbalances that adversely affect land resource production systems. An extended period—a season, a year, or several years—of deficient rainfall relative to the statistical multi-year mean for a region. Drought is a normal part of climate, rather than a departure from normal climate. Drought is not a word with a precise definition. A drought is simply a period during which rainfall is markedly lower than the average for that time of year in that place, and consequently, water is in such short supply that domestic and industrial users, farmers, and wildlife are affected. Drought is an insidious natural hazard that results from a deficiency of precipitation from expected or “normal” that, when extended over a season or longer, is insufficient to meet the demands of human activities and the environment. Drought is a normal, recurrent feature of climate, although often erroneously considered an unexpected and extraordinary event. It is a temporary aberration within the natural variability and can be considered an insidious hazard of nature; it differs from aridity which is a long-term, average feature of climate. Drought is a period of drier-than-normal conditions that results in water-related problems. It is the period when rainfall is less than normal for several weeks, months or years, the flow of streams and rivers declines and water levels in lakes and reservoirs descent and the depth of water in wells increase. Drought is a recurring extreme climate event over land characterized by below-normal precipitation over a period of months to years. Drought is a temporary dry period, in contrast to the permanent aridity in arid areas. Drought is by far the most important environmental stress in agriculture, causing important crop losses every year.

Linsely et al. (1959) Gumbel (1963) Palmer (1965) FAO (1983) WMO (1986) UNa Secretariat General (1994)

a

Schneider (1996) Glantz (2003) Allaby (2003)

Wilhite and Buchanan-Smith (2005) MWD (2007)

Nagarajan (2009)

Dai (2010)

Mastrangelo et al. (2012)

UN Secretariat General, the UN Convention to Combat Drought and Desertification

zones, but its characteristics vary significantly from one region to another. Droughts generally result from a combination of natural factors that can be enhanced by anthropogenic influences. The primary cause of any drought is a deficiency in rainfall and, in particular, the timing, distribution, and intensity of this deficiency in relation to the existing water storage, demand, and use. This deficiency can result in a shortage of water necessary for the functioning of a natural (eco-) system and/or necessary for a certain human activities (MWD 2007). Classification of droughts The droughts are generally classified into four categories (Wilhite and Glantz 1985; AMS 2004; Hennessy et al. 2008), which include: 1. Meteorological drought is defined as a lack of precipitation over a region for a period of time. Precipitation has been commonly used for meteorological drought analysis. Considering drought as precipitation deficit with respect to average values, several studies have analyzed droughts using monthly precipitation data.

Other approaches analyze drought duration and intensity in relation to cumulative precipitation shortages. Otherwise, it is a period of months to years when atmospheric conditions result in low rainfall. This can be exacerbated by high temperatures and high evaporation, low humidity, and desiccating winds. 2. Hydrological drought is related to a period with inadequate surface and subsurface water resources for established water uses of a given water resources management system. Streamflow data have been widely applied for hydrologic drought analysis. From regression analyses relating droughts in stream flow to catchment properties, it is found that geology is one of the main factors influencing hydrological droughts. Otherwise, it is prolonged moisture deficits that affect surface or subsurface water supply, thereby reducing stream flow, groundwater, dam, and lake levels. This may persist long after a meteorological drought has ended. 3. Agricultural drought, usually, refers to a period with declining soil moisture and consequent crop failure without any reference to surface water resources. A

Arab J Geosci

decline of soil moisture depends on several factors which affected by meteorological and hydrological droughts along with differences between actual evapotranspiration and potential evapotranspiration. Several drought indices, based on a combination of precipitation, temperature, and soil moisture, have been derived to study agricultural droughts. Otherwise, it is a short-term dryness in the surface soil layers (root-zone) at a critical time in the growing season. Plant water demand depends on prevailing weather conditions, biological characteristics of the specific plant and stage of growth, and the physical and biological properties of soil. The start and end may lag that of a meteorological drought, depending on the preceding soil moisture status. 4. Socio-economic drought is associated with failure of water resources systems to meet water demands and thus associating droughts with supply of and demand for an economic good (water). Socio-economic drought occurs when the demand for an economic good exceeds supply as a result of a weather-related shortfall in water supply. Otherwise, it is the effect of elements of the above droughts on supply and demand of economic goods and human well-being. A relationship between the meteorological, agricultural, and hydrological droughts can be analyzed from the figure below (Fig. 1). Impacts of drought The impacts of a drought can be economic, environmental, or social. Drought produces a complex web of impacts that spans many sectors of the economy and reaches well beyond the area experiencing physical drought. This complexity exists because water is integral to society’s ability to produce goods and provide services. Impacts are commonly referred to as direct and indirect. Direct impacts include reduced crop, rangeland, and forest productivity; increased fire hazard; reduced water levels; and increased livestock and wildlife mortality rates, and damage to wildlife and fish habitat. The consequences of these direct impacts illustrate indirect impacts. For example, a reduction in crop, rangeland, and forest productivity may result in reduced income for farmers and agribusiness, increased prices for food and timber, unemployment; and reduced tax revenues because of reduced expenditures, foreclosures on bank loans to farmers and businesses; migration; and disaster relief programs. According to MWD (2007) and Hadish (2010), the impacts of drought can be categorized as follows. Economic impacts Drought-induced water deficiency affects production, sales, and business operations in a variety of industries (Ding et al.

2010). In this review article, these effects are referred to as the direct economic impacts of drought, while indirect economic impacts of drought stem from the interactions and transactions among industries and sectors. Drought also causes environmental and social impacts, and results in non-market losses. An overview of drought economic impacts is supplied in Fig. 2. Drought impacts are most eye-catching in the agricultural sector. Dried crops, abandoned farmland, and withered and yellow pastureland are the common signs of drought. Prolonged soil moisture deficits due to drought cause damage to crops and pastures. Crop failures and pasture losses are the primary direct economic impact of drought within the agricultural sector. Drought-induced production losses cause negative supply shocks, but the amount of incurred economic impacts and distribution of losses depends on the market structure and interaction between the supply and demand of agricultural products. Drought also causes significant economic impacts in non-agricultural sectors through its effects on water supplies including stream flows, reservoirs, wetlands, and groundwater. These non-agricultural sectors include, but are not limited to, tourism and recreation, public utilities, horticulture and landscaping services, navigation, and other industries/ businesses that have significant water consumption. Environmental impacts Drought also affects the environment in many different ways. Plants and animals depend on water, just like people. When a drought occurs, their food supply can shrink and their habitat can be damaged. Sometimes, the damage is only temporary and their habitat and food supply return to normal when the drought is over. But sometimes, drought’s impact on the environment can last a long time, maybe forever. Also, environmental losses are the result of damages to plant and animal species, wildlife habitat, and air and water quality; forest and range fires; degradation of landscape quality; loss of biodiversity; and soil erosion. Some of the effects are short-term, and conditions quickly return to normal following the end of the drought. Other environmental effects linger for some time or may even become permanent. Wildlife habitat, for example, may be degraded through the loss of wetlands, lakes, and vegetation. However, many species will eventually recover from this temporary aberration. The degradation of landscape quality, including increased soil erosion, may lead to a more permanent loss of biological productivity of the landscape. Although environmental losses are difficult to quantify, growing public awareness and concern for environmental quality has forced public officials to focus greater attention and resources on these effects.

Arab J Geosci Fig. 1 The sequence of drought impacts associated with meteorological, agricultural, and hydrological drought (Source: National Drought Mitigation Centre, http:// enso.unl.edu/ndmc/enigma/ def2.htm)

Social impacts The social dimensions of drought are wide-ranging and typically compound problems that may have already existed Fig. 2 An overview of drought economic impacts (adapted from Ding et al. 2010)

within the community. For example, if a community is experiencing a shortage of health, education, housing, or employment resources, then the effects of drought will place further strain on those limited resources and affect the ability

Arab J Geosci

of providers to deliver effective services (Kenny 2008). A degree of stress is normal in life, and most rural people are experienced in coping with droughts and various other difficulties. However, a prolonged drought represents a time of major change and crisis for many in rural communities. Research indicates that social impacts as a result of drought on individuals, families, and communities may include:

1. Identify the drought hazard with regard to its spatial extends, frequency, and severity. 2. Identify and quantify drought vulnerability, e.g., people, economy, and structure exposed to the drought hazard. 3. Compute drought risk pattern from drought hazard and vulnerability.

&

Drought vulnerability assessment provides a framework for identifying the social, economic, and environmental causes of drought impacts. It is one of the main aspects of drought planning, mitigation, and bridging the gap between impact assessment and policy formulation by directing policy attention to underlying cause of vulnerability rather than the negative effects which follow triggering events such as droughts. With a map of drought vulnerability, decisionmakers can visualize the hazard risk and convey vulnerability information to other sectors to ensure that they will act in a timely and effectively way to tackle drought-related losses. Vulnerability to drought is dynamic and is influenced by a multitude of factors, including increases and regional shifts in population, urbanization, technology, government policies, land use and other natural resource management practices, desertification processes, water-use trends, and increasing environmental awareness. To determine drought vulnerability, the most important and most difficult task is to select the factors and to determine the weighting of those factors, which are commonly subjective and may vary between regions. The main reason for this problem is that the mechanism for determining how agricultural drought vulnerability is produced is still unclear (Wu et al. 2011). Drought risk assessment in a particular area plays an important role in water resources management (Bordi et al. 2006). From the traditional “drought disaster crisis management” (Wilhite and Glantz 1985) to the modern “drought disaster risk management,” it is the international research trend of drought disaster (AMS 1997). Drought risk, without universal definition, is usually assessed in terms of its impacts on human activities, economic, social, and environmental systems. The purpose of drought risk assessment is to identify appropriate actions that can be taken to reduce the loss of the potential damage. Based on the drought risk assessment results, decision makers can visualize the hazard and appreciate the loss of agricultural producers, natural resource managers, and others (Zhang et al. 2011a, b). Most methods of drought risk assessment currently focus on the drought dangerousness and vulnerability. As important parts of agricultural drought risk, exposure and drought-resistibility also need to be considered. Exposure and drought-resistibility are considered into the agricultural drought risk assessment model. Using variable fuzzy set model, dangerousness, vulnerability, exposure, and drought-resistibility of drought and comprehensive agricultural drought risk are assessed, and through the spatial

& & & & & & & & & &

People being reluctant to get involved in community activities, A decline in traditional industries, Volunteer stress or burnout, or an inability to even have a volunteering effort, The need to and or ability to seek off-farm work, Increased financial pressures, A decline in the health (both physical and mental health) of individuals and their families, Dealing with questions of whether to leave the farm and/ or problems associated with succession planning, A loss of local farm labor, An inability to leave the property because of the demands of feeding and water regimes, The local economy impact from a postponement of capital purchases as a result of drought, and A general increase of working hours with little opportunity for recreation and family time.

Drought risk assessment Risk is the probability of harmful consequences, or expected losses resulting from interactions between hazards and vulnerable conditions. Therefore, a conceptual approach to risk assessment can be broken down into a combination of the hazard and vulnerability. Similar to other natural hazard risks, drought risk depends on a combination of the physical nature of drought and the degree to which a population or activity is vulnerable to the effects of drought (Shahid and Behrawan 2008). Drought risk assessment is a challenging task, because drought is a very complex and least understood phenomenon lacking a universal definition and onset criteria (Wilhite 2000). Drought is not only affected by the natural factors such as meteorology and hydrology but also by crop planting structures and drought resistance capacities (Huo et al. 2003). It is usually estimated in terms of drought dangerousness, vulnerability, exposure, and drought-resistibility. Therefore, to study the risk of drought, it is essential to study the frequency, severity, and spatial extent of drought as well as the infrastructural and socioeconomic ability of the region to anticipate and cope with the drought. The following steps can be used to identify the drought risk assessment:

Arab J Geosci

distribution maps of drought risk, decision makers could find out the drought situation and make decisions on resisting drought reasonably (Zhang et al. 2011a, b). Zhang et al. (2011a, b) mentioned that the major methods of drought risk assessment are as follows: 1. The principal component analysis (Che et al. 2010), 2. The analytic hierarchy process method (Ni and Gu 2005), 3. The fuzzy evaluation method (Zhou 2005), and 4. The gray model evaluation method and artificial neural network models (Feng et al. 2000). Most of these assessments methods deal with the evaluation standard as point value, and the evaluation results lack scientific basis. A variable fuzzy set model, based on a relative difference function, is a good solution which divides evaluation results with interval. Through varying parameters, a variable fuzzy sets model could locate the assessment results in the intervals, which makes assessment results more believable. Role of remote sensing and GIS in drought studies The detection, monitoring, and mitigation of disasters require gathering of rapid and continuous relevant information that are not effectively collected by conventional methods. Remote sensing tools and techniques make it possible to obtain and distribute continuous information rapidly over large areas by means of sensors operating in several spectral bands, mounted on aircraft or satellites. A satellite, which orbits the Earth, is able to explore the whole surface in a few days and repeat the survey of the same area at regular intervals while an aircraft can give a more detailed analysis of a smaller area, if a specific need occurs. The spectral bands used by these sensors cover the whole range between visible and microwaves (Hadish 2010). The remote sensing monitoring of drought can get frequent and sustained information on the surface characteristics of planar with full using information of ground surface spectrum of time, space, and direction. It can provide macro, dynamic, and real-time monitor data sources for real-time and dynamic monitoring of drought (Zhang et al. 2011a, b). For the last three decades, advancements in the fields of GIS and remote sensing (RS) have greatly facilitated the operation of drought risk assessment. Most data required for drought risk assessment have a spatial component and also change over time. Therefore, the use of GIS and RS has become essential. It is evident that GIS has a great role to play in drought risk assessment because natural hazards are multi-dimensional. The main advantage of using GIS for drought risk assessment is that it not only generates a visualization of hazard but also creates potential to further analyze this product to estimate probable damage due to drought hazard. Drought risk

assessment requires up-to-date and accurate information on the terrain topography and the use of the land. The remotely sensed images from satellites and aircrafts are often the only source that can provide this information for large areas at acceptable costs (Wipulanusat et al. 2009). A meteorological station can connect to GIS and keep receiving meteorological information directly entered into GIS, and then these data will managed and analyzed uniformly by the system database. GIS transformed the model to its language and analyzes the data by powerful analysis function, and then adds drought assessment early warning function into drought assessment system (Tao et al. 2011). The technical procedure for early warning and drought risk assessment is shown in Fig. 3. Meteorological drought indices and drought detection Drought indices are commonly used to detect the potential risk of occurrence and severity of drought, and to study spatial–temporal reasoning. Many of these indices have been developed for detecting temporal variability and magnitude of the drought actions in interesting regions. There are several indices that measure how much precipitation for a given period of time has deviated from historically established norms. Some of the widely used drought indices include Palmer Drought Severity Index (PDSI), Crop Moisture Index (CMI), Standardized Precipitation Index (SPI), and Surface Water Supply Index (SWSI). Available drought indices are tabulated in Table 2. Palmer Drought Severity Index (PDSI) The PDSI was originally developed by Palmer (1965) with the intent to measure the cumulative departure in surface water balance. It incorporates antecedent and current moisture supply (precipitation, P) and demand (potential evapotranspiration, PE) into a hydrological accounting system, which includes a two-layer bucket-type model for soil moisture calculations. The PDSI is a standardized measure, ranging from about −10 (dry) to +10 (wet) with values below −3 representing severe to extreme drought. Also, the PDSI has been widely used to study aridity changes in modern and past climates. Efforts to address its major problems have led to new variants of the PDSI, such as the self calibrating PDSI (sc_PDSI) and PDSI using improved formulations for potential evapotranspiration (PE), such as the Penman– Monteith equation (PE_pm) instead of the Thornthwaite equation (PE_th). To compare and evaluate the data, four forms of the PDSI are used, namely, the PDSI with PE_th (PDSI_th), PE_pm (PDSI_pm), the sc_PDSI with PE_th (sc_PDSI_th), and PE_pm (sc_PDSI_pm), calculated using available climate data from 1850 to 2008 (Dai 2011).

Arab J Geosci Fig. 3 Information system implementing procedure for early warning drought risk assessment (adapted from Tao et al. 2011)

For meteorological drought, precipitation is the primary variable in computing the indices, with secondary contributions from surface air temperature to account for the effect of evaporation in some of the indices, such as the PDSI. For agricultural drought, soil moisture content (not always measured) is often used, whereas stream flow is commonly used in measuring hydrological drought. Table 3 shows a classification system linking PDSI s for dry and wet periods (Palmer 1965). Crop Moisture Index (CMI) Three years after the introduction of his drought index, Palmer (1968) introduced a new drought index based on weekly mean temperature and precipitation known as Crop Moisture Index (CMI). CMI reflects moisture supply in the short-term across major crop-producing regions and is not intended to assess long-term drought. It identifies potential areas for agricultural droughts. CMI defines drought in terms of the magnitude of computed abnormal ET deficit which is the difference between actual and expected weekly ET. The expected weekly ET is the normal value adjusted up or down according to the departure of the week’s temperature from normal. It is a location-based estimate and differs from place to place indicating the moisture condition. As CMI is designed for short-term soil moisture demand of the crops; it is not effective for long-term drought monitoring. CMI is not useful for crop initiation periods when it differs from place/plot to place-seed germination and uses a meteorological approach to monitor week-to-week crop conditions and evaluate moisture conditions across major crop producing regions. It responds rapidly to changing conditions and location and time. Meanwhile, it may not be

applicable during seed germination or specific cropgrowing season. The continuous soil moisture measurement using tensiometers is essential in monitoring agriculture drought. Water deficit affects crop growth and development, directly or indirectly. Crop water status is highly dynamic and influenced by soil and atmospheric microenvironment and regulated by physiological factors of species. The crop may be tolerant for moderate drought at flowering stage, and the yield reduction would result in decline in crop yield, when the drought is prevalent during grain and filling stage (Nagarajan 2010). Standardized Precipitation Index (SPI) Jain et al. (2010) reported that there are a number of indices to quantify drought using meteorological data; however, the SPI is most widely used index. The SPI was formulated by Tom McKee, Nolan Doesken, and John Kleist of the Colorado Climatic Centre in 1993. SPI can be calculated at different time scales and hence can quantify water deficits of different duration. SPI was designed to show that it is possible to simultaneously experience wet conditions on one or more time scales and dry conditions at another time scale. SPI is computed by fitting historical precipitation data to a Gamma probability distribution function for a specific time period and location, and transforming the Gamma distribution to normal distribution with a mean of zero and a standard deviation of 1. SPI for given rainfall amount is then given by the precipitation deviation from the mean of an equivalent normally distributed function with a zero mean and a standard deviation of 1. The main premise of the current effort is that the use of a drought index, such

Arab J Geosci Table 2 Comparison among available drought indices (adapted from Hayes 2012) Indices

Method

Application Pros

Cons

Percent of normal

Percent of normal is a simple method to detect drought. It is calculated by dividing actual precipitation by normal precipitation—typically a 30-year mean and multiplying it by 100 % for each location. Data are not normalized.

Percent of normal is effective in single region or season,

Percent of normal cannot determine the frequency of the departures from normal or compare with different locations. Also, it cannot identify specific impact of drought or the inhibition factor for drought risk mitigation plans.

Standardized Precipitation Index (SPI)

SPI is a simple index which is calculated from the long-term record of precipi tation in each location (at least 30 years). The data will be fitted to normal distribution and be normalized to a flexible multiple time scale such as 3-,6-,12-,24- 48-, etc. PDSI complexity is calculated from precipitation, temperature and soil moisture data. Soil moisture data has been calibrated to the homogeneous climate zone. PDSI has an inherent time scale of 9 months. PDSI treats all forms of precipitation as rain.

SPI can provide early warning of drought and its severity because it can specify for each location and is well-suited for risk management.

The data can be changed from the longterm precipitation record. The long time scale up to 24 month is not reliable.

PDSI has been widely used to trigger agricultural drought. PDSI can be used to identify the abnormality of drought in a region and show the historical aspects of current conditions.

The PDSI may lag in the detection of drought over several months because the data depend on soil moisture and its properties which have been simplified to one value in each climate division. The PDSI will not present accurate results in winter and spring due to the effects of frozen ground and snow. PDSI also tends to underestimate runoff conditions.

The PHDI has been officially used by NCDC to determine the precipitation needed for drought termination and amelioration which has a PHDI equal to −0.5 and −2.0 consecutively. It has been used Indiana for drought monitoring. CMI is used to monitor crop condition. It is effective for the detection of short-term agricultural drought while the Z index determines drought on a monthly scale. It can detect drought sooner than PDSI and PHDI.

The PHDI is developed from precipitation, outflow, and storage. PHDI may change more slowly than PDSI and it has sluggish response for drought.

Palmer Drought Severity Index (PDSI)

Palmer Hydrological Drought Index

PHDI has been derived from the PDSI index to quantify the long term impact from hydrological drought.

Crop Moisture Index (CMI)

CMI is a derivative of PDSI which was developed from moisture accounting procedures as the function of the evapotranspiration anomaly and the moisture excesses in the soil. It also can be present as the monthly moisture anomaly or Z index (ZNDX) as a product from PDSI calculation. CMI looks at the top 5 ft of the soil layer. SWSI is used for frequency analysis to normalize long-term data such as precipitation, snow pack, stream flow, and reservoir level.

Surface Water Supply Index (SWSI) Reclamation Drought Index (RDI)

Deciles

The RDI index is similar to the SWSI index. It combines the functions of supply, demand, and duration. RDI also combines temperature features and duration in the index. Deciles have been developed to use instead of percent of normal. Deciles are calculated from the number of occurrences distributed from 1 to 10. The lowest value indicates conditions drier than normal and the higher value indicate conditions wetter than normal.

CMI is limited to use only in the growing season; it cannot determine the long term period of drought.

The SWSI is very useful for indicating snow pack conditions in mountain areas to measure the water supplied for community.

The index of different basins cannot be compared with each other and has been computed seasonally. States such as Colorado, Oregon, Montana, Idaho, and Utah have used SWSI.

The RDI is used as the trigger to evaluate drought reclamation plans and to release drought emergency funds.

The disadvantage of RDI is the same as the SWSI index. The state such as Oklahoma has used RDI.

The deciles index has been used in Australia; it provides accurate precipitation data for drought response.

However, its use requires a long climatology record to accurately calculate the deciles index.

Hayes (2012). Drought Indices, National Drought Mitigation Center (http://www.drought.unl.edu/whatis/indices.htm). With modifications by Dev Niyogi and Umarporn Charusambot, Indiana State Climate Office, Purdue University (http://iclimate.org/2.2.2012)

as SPI, may lead to a more appropriate understanding of drought duration, magnitude, and spatial extent in semi-

arid areas (Karavitis et al. 2011). Table 4 shows the classification values for SPI.

Arab J Geosci Table 3 PDSI classifications for dry and wet periods (Palmer 1965)

PDSI value

Drought category

4.00 or more 3.00 to 3.99 2.00 to 2.99 1.00 to 1.99 0.50 to 0.99 0.49 to −0.49 −0.50 to −0.99 −1.00 to −1.99 −2.00 to −2.99 −3.00 to −3.99 −4.00 or less

Extremely wet Very wet Moderately wet Slightly wet Incipient wet spell Near normal Incipient dry spell Mild drought Moderate drought Severe drought Extreme drought

The SPI is a way of measuring drought that is different from the PDI. Like the PDI, this index is negative for drought and positive for wet conditions. But, the SPI is a probability index that considers only precipitation, while Palmer’s indices are water balance indices that consider water supply (precipitation), demand (evapotranspiration), and loss (runoff). Figure 4 shows the SPI in the northwest coast of Egypt. Egypt has drought when the SPI value less than zero and no drought when SPI is more than zero.

Satellite-based drought indices for drought characterization Drought indicators assimilate information on rainfall, stored soil moisture, or water supply but do not express much local spatial detail. Also, drought indices calculated at one location is only valid for single location. Thus, a major drawback of climate-based drought indicators is their lack of spatial detail, as well as their dependence on data collected at weather stations, which are sometimes sparsely distributed, affecting the reliability of the drought indices. Satellitederived drought indicators calculated from satellite-derived surface parameters have been widely used to study droughts. NDVI, VCI, and Temperature Condition Index (TCI) are some of the extensively used vegetation indices. Normalized Difference Vegetation Index (NDVI) The NDVI is a measure of the “greenness,” or vigor of vegetation. It is derived based on the known radiometric properties of plants, using visible (red) and near-infrared (NIR) radiation. NDVI is defined as: NDVI ¼ ðNIR  REDÞ=ðNIR þ REDÞ

Standardized Water Supply Index (SWSI) The SWSI is a hydrological drought index that was developed to replace the PDSI in areas where local precipitation is not the sole (or primary) source of stream flow. SWSI was designed for mountainous locations with significant snowfall because of the delayed contribution of snowmelt runoff to surface water supplies. The SWSI is calculated based on the monthly non-exceedance probability which is determined using available historical records of reservoir storage, stream flow, precipitation, and snowpack. Using a basin-calibrated SWSI algorithm, weights are assigned to each hydrological component based on its typical contribution to the water supply. SWSI is a particularly good measure of surface water supply conditions in the Western United States because it accounts

Table 4 The classification values for SPI (McKee et al. 1993)

for the major hydrological variables that contribute to surface water supply there (Quiring et al. 2007).

SPI value

Drought category

2.00 and above 1.50 to 1.99 1.00 to 1.49 −0.99 to 0.99 −1.00 to −1.49 −1.50 to −1.99 −2.00 and less

Extremely wet Very wet Moderately wet Near normal Moderately dry Severely dry Extremely dry

Where NIR and RED are the reflectance in the nearinfrared and red bands. NDVI is a good indicator of green biomass, leaf area index, and patterns of production because, when sunlight strikes a plant, most of the red wavelengths in the visible portion of the spectrum (0.4–0.7 mm) are absorbed by chlorophyll in the leaves, while the cell structure of leaves reflects the majority of NIR radiation (0.7–1.1 mm). Healthy plants absorb much of the red light and reflect most NIR radiation. In general, if there is more reflected radiation in the NIR wavelengths than in the visible wavelengths, the vegetation is likely to be healthy (dense). If there is very little difference between the amount of reflected radiation in the visible and infrared wavelengths, the vegetation is probably unhealthy (sparse). However, this can also result from partially or non-vegetated surfaces. NDVI values range from −1 to +1, with values near zero indicating no green vegetation and values near +1 indicating the highest possible density of vegetation. Areas of barren rock, sand, and snow produce NDVI values of <0.1, while shrub and grassland typically produces NDVI values of 0.2–0.3, and temperate and tropical rainforests produce values in the 0.6–0.8 range. Figure 5 shows the Egypt NDVI in March 31, 2012. Comparing the NDVI time series for a number of years at the same location provides information about the relative health of the vegetation in a given year. Interannual variations

Arab J Geosci Fig. 4 Standardized Precipitation Index (SPI) northwest coast of Egypt

in the magnitude and evolution of the NDVI for a particular location are mainly governed by meteorological variables such as precipitation, temperature, and relative humidity; however, changes in land use and land cover can also cause interannual variations and trends in the NDVI. It can be inferred that low productivity (lack of “greenness” or vigor) is caused, in part, by poor weather conditions, and that high productivity is due, in part, to favorable weather conditions. It should be noted that the interpretation of NDVI values Fig. 5 Egypt-Normalized Difference Vegetation Index, March 31, 2012. (http:// www.star.nesdis.noaa.gov/ smcd/emb/vci/VH/vh_ imageloopByCountry.php)

is spatially dependent. This is because more productive ecosystems have different radiometric properties than less productive ones due to differences in climate, soil, and topography (Quiring and Ganesh 2010). Vegetation Condition Index (VCI) The VCI is a pixel-wise normalization of NDVI that is useful for making relative assessments (e.g., pixel-specific) of changes in the NDVI signal by filtering out the

Arab J Geosci

contribution of local geographic resources to the spatial variability of NDVI (Quiring and Ganesh 2010). Jain et al. (2010) stated that VCI is an indicator of the status of vegetation cover as a function of NDVI minima and maxima encountered for a given ecosystem over many years. It normalizes NDVI (or any other vegetation index) and allows for a comparison of different ecosystems. It is an attempt to separate the short-term climate signal from the long-term ecological signal, and, in this sense, it is a better indicator of water stress condition than NDVI. The significance of VCI is strongly related to the relation between the vegetation index and the vitality of the vegetation cover under investigation. VCI is defined as:

distance, the stronger the drought and the lower the soil moisture or vice versa. Therefore, for a random, mixed pixel E (RRed, RNIR) in the NIR-Red spectral space, the vertical distance from E (RRed, RNIR) to line L (PDI), can be written as the following:

VCIj ¼ ðNDVIj  NDVIminÞ=ðNDVImax  NDVIminÞ

Temperature Condition Index (TCI)

 100 Where NDVI, NDVImax, and NDVImin are monthly NDVI, multi-year maximum NDVI, and multi-year minimum NDVI, respectively, for each grid cell. VCI changes from 0 to100 corresponding to changes in vegetation condition from extremely unfavorable to optimal. In the case of an extremely dry month, the vegetation condition is poor, and VCI is close or equal to zero and reflects an extreme dry month. A VCI of 50 reflects fair vegetation conditions. At optimal condition of vegetation, VCI is close to 100. Also, VCI values indicate how much the vegetation has advanced or deteriorated in response to weather. It was concluded from the above studies that VCI has provided an assessment of spatial characteristics of drought, as well as of its duration and severity, which were in good agreement with precipitation patterns. Figure 6 shows the VCI Egypt in EgyptVegetation Condition Index (VCI) in 31 March 2012. Perpendicular drought index and its inherent limitations For a specific soil type, the soil line can be regarded as the plot that characterizes the spectral behavior of non-vegetated pixels and whose moisture content varies noticeably (Ghulam and Qin 2007). It is not difficult to see from Fig. 7 that the drought severity gradually rises from B to C and reaches its climax at C. Here, BC represents the soil line. A line L, which dissects the coordinate origin and is vertical to the soil line, may serve as a reference to determine how dry the pixel is. For a bare soil, the distance from any point in the NIR-Red spectral space to line L represents the drought severity of a non-vegetated surface. With the increasing amount of vegetation, the plots shift upward along the direction vertical to the soil line, while they do the same along the direction parallel to the soil line and orthogonal to normal line L with decreasing soil moisture. For a vegetated surface, the distance from L to any point in the NIR-Red spectral space may indicate the drought severity of a mixed pixel. That is, the farther the

1 PDI ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðRRed þ MRNIR Þ M2 þ 1 Here, RRed and RNIR refer to the atmospherically corrected reflectance of the Red and NIR bands, respectively, while M refers the slope of the soil line.

According to Owrangi et al. (2011), during the rainy season, in general, it is common for overcast conditions to prevail for up to 3 weeks. When conditions last longer than this, the weekly NDVI values tend to be depressed, giving the false impression of water stress or drought conditions. To remove the effects of contamination in satellite assessment of vegetation conditions, the use of a TCI is suggested. The TCI is calculated much in the same way as the VCI, but its formulation is modified to reflect the vegetation’s response to temperature (i.e., the higher the temperature the more extreme the drought). TCI is based on brightness temperature (BT) and represents the deviation of the current month’s temperature from the recorded maximum. Using meteorological observations, as well as the relationship between ground surface temperature and moisture regimes, drought-affected areas can often be detected before biomass degradation occurs. Hence, TCI plays a key role in drought monitoring. According to Singh et al. (2003), TCI is based on the thermal band (Channel 4) of Advanced Very-HighResolution Radiometer (AVHRR) converted to BT. TCI is used to determine temperature-related vegetation stress and also stress caused by excessive wetness. The TCI algorithm is similar to the VCI algorithm and is given as TCI ¼ ðBT max BTÞ=ðBTmax  BtminÞ  100 Where BT, BTmax, and BTmin are the smoothed weekly brightness temperature, multi-year maximum, and multi-year minimum, respectively, for each grid cell. The conditions are estimated relative to the maximum and minimum temperature envelopes. The above formula reflects different response of vegetation to temperature. High temperatures in the middle of the growing season indicate unfavorable conditions for drought, while low temperatures indicate mostly favorable conditions (Owrangi et al. 2011). Figures 8 and 9 show the TCI for Egypt in March 31, 2012, and Egypt-Smoothed Brightness Temperature, March 31, 2012, respectively.

Arab J Geosci Fig. 6 Egypt-Vegetation Condition Index (VCI) in 31 March 2012. (http://www.star. nesdis.noaa.gov/smcd/emb/vci/ VH/vh_ imageloopByCountry.php)

NDVI–rainfall relationship as indicator of drought Several studies have been devoted towards drought with the aid of satellite-derived information. Reflectance in the visible,

Fig. 7 Sketch map of NIR-Red space and PDI

near-infrared, and thermal bands were combined into VCI, TCI, and NDVI, which considerably improved early drought detection, watch, and monitoring of drought’s impacts on agriculture. Using National Oceanic and Atmospheric Administration (NOAA) AVHRR data, researchers have successfully extended satellite data analysis to large-area vegetation monitoring and biomass productivity estimates. Bajgiran et al. (2008) mentioned that the significant correlations have been found between NDVI values and precipitation data in individual metrological stations in Iran. Therefore, NDVI values can be used for monitoring and mapping of drought conditions in Iran. In semiarid ecosystems, multi-month precipitation has greater impacts on vegetation condition as compared with the current month precipitation. Also, rainfall characteristics may not be the only factor to influence NDVI values, because other local factors, such as topography, soil characteristics, stress in previous years, and land cover characteristics of the area, should be considered in order to explain climate effects on vegetation cover. They also confirmed that both NDVI and VCI indices could be used to monitor drought at a regional scale. This is a promising finding that could eventually lead to the preparation of drought risk maps if the study is extended over a larger region of the country. However, in order to obtain more reliable results, there is a strong need for wider time span of satellite data.

Arab J Geosci Fig. 8 Egypt-Temperature Condition Index in March 31, 2012. (http:// www.star.nesdis.noaa.gov/ smcd/emb/vci/VH/vh_ imageloopByCountry.php)

The approach will have to be tested in different climates, using combination of different climatic factors (such as temperature and precipitation). The major advantage of the presented methodology is seen in high

Fig. 9 Egypt-Smoothed Brightness Temperature, March 31, 2012. (http:// www.star.nesdis.noaa.gov/ smcd/emb/vci/VH/vh_ imageloopByCountry.php)

spatial information content satellite data, which allows for providing drought risk maps. Finally, it must be noted that, in semi-arid climates, precipitation is extremely rare in the summer. This results in the high

Arab J Geosci

dependence of summer vegetation on the spring precipitation which in turn makes multi-month precipitation more influential. Gebrehiwot et al. (2011) studied the coloration between Rainfall and Vegetation indices in Northern highlands of Ethiopia, and they found the result of vegetative drought analysis further illustrates significant correlations between VCI values and precipitation data for individual stations. Therefore, VCI indices can be used to detect unfavorable environmental conditions, particularly the current drought status, as well as to analyze the characteristics of the drought at a regional scale. The study further indicated that VCI values have strong correlation with precipitation as compared with the NDVI. This indicates that rainfall characteristics may not be the only factor to influence NDVI values. Other local factors, such as soil characteristics, stress in previous years, and land cover characteristics of the area could also have an influence on vegetation.

for all locations studied, various types of drought correlated with mean annual precipitation and temperature. The combined approach of WD and SPI mainly carried out for periods of 1 year, but such studies could also be done for shorter periods like months, quarters, or growing season. The most arid regions did not necessarily coincide with areas of the most severe drought, as no correlations between WD and SPI and no altitude-based SPI zones around the Carpathian Mountains, as is the case for other climate characteristics, soils and vegetation. Water resource problems arise where both SPI values characterize extremely droughty periods, and WD values are greatly below −200 mm/year. This combined use of SPI and WD characterizes the dryness of a region better than one factor alone and should be used for better management of water in agriculture in Romania and also other countries with similar climate characteristics.

SPI-based drought identification

Gregoric and Sušnik (2010) stated that drought is a natural phenomenon which closely connects climate and society. When drought occurs, there is certain risk that the population will suffer social and economic consequences. This risk and depth of such consequences depends on the natural frequency and severity of the drought. As the climate changes, natural hazards are increase, and it would be reasonable to imply that social and economic risks are consequently increasing. However, natural hazards are not the only element determining risk. The other factor is society’s capability to overcome difficulties caused by water shortages, i.e., vulnerability. Vulnerability determines the risk of drought impact now and in the future. Risk may rise independent of climatic trends, due to increased water demands caused by population or economic growth, or both. And the other way around: Natural hazard trends may be neutralized by reducing a society’s vulnerability. Assessment of both—natural hazards and societal vulnerability—are among the core objectives of the drought management center for southeastern Europe. Historical assessment of drought occurrence and establishment of drought monitoring systems are undertaken to establish a method for regional estimation of climatological and actual natural hazards connected to occurrence of drought. Some aspects of vulnerability (mainly in the agriculture sector) have been described for some southeast European countries. Few extreme events are as economically and ecologically disruptive as drought, which affects millions of people in the world each year. Severe drought conditions can profoundly impact agriculture, water resources, tourism, ecosystems, and basic human welfare. As mentioned before, understanding drought and modeling its components have drawn attention of ecologists, hydrologists, meteorologists, and agricultural scientists. Droughts are of great importance in water resources planning and management (Dai 2010).

Livada and Assimakopoulos (2007) used SPI to detect drought events in spatial and temporal basis in Greece. Using monthly precipitation data from 23 stations well spread over Greece and for a period of 51 years, a classification of drought is performed, based on its intensity and duration. From the estimation of the SPI on 3-, 6-, and 12-month time scales, it is evident that the frequency of mild and moderate drought conditions is approximately of the same order of magnitude over the whole Greek territory. Frequencies present a small reduction moving from north to south and from west to east. The small precipitation amounts over the southern part of Greece during the summer period resulted in the highest frequency of SPI values for severe droughts on the 3-month time scale. From the study of persistence of severe or more drought conditions on 6- and 12-month time scales according to Besson’s coefficient of persistence, it was found that, in almost all cases and for both time scales, the persistence is statistically significant. Paltineanu et al. (2009) characterized droughts in Romania using the approach of both the SPI and climatic water deficit (WD). The values of the main climatic factors (rainfall, temperature, reference evapotranspiration, etc.) were obtained from 192 weather stations in various regions of Romania. Penman–Monteith reference evapotranspiration (ETo-PM) was used to calculate WD as the difference between precipitation (P) and ETo-PM. SPI calculated from precipitation values. There is a clear difference between drought and aridity. Drought occurrence determines higher WD values for plains and plateaus and lower climatic excess water (EW) values for high mountains in Romania, depending on the aridity of the specific region considered and drought severity. WD was calculated as mean values for both normal conditions and,

Drought management and forecasting

Arab J Geosci Fig. 10 Different components of drought modeling (adapted from Mishra and Singh 2011)

Drought forecasting is a critical component of drought hydrology which plays a major role in risk management, drought preparedness, and mitigation. There has been considerable work done on modeling various aspects of drought, such as identification and prediction of its duration and severity. However, a major research challenge is to develop suitable techniques for forecasting the onset and termination points of droughts. One of the deficiencies in mitigating the effects of a drought is the inability to predict drought conditions accurately for months or years in advance. For example, the NOAA Fig. 11 Different components for drought forecasting (adapted from Mishra and Singh 2011)

issued forecasts of spring and summer droughts for five Midwestern states (USA) in the 2000s. However, in early June of 2000, heavy rains began to fall across the Midwestern drought area which reveals the rainfall totals for June ranked as the sixth wettest in the past 106 years. This may be due to the spatio-temporal variability of hydro-meteorological variables associated with the intensification of global hydrologic cycle. The different components of drought forecasting are shown in Fig. 10, which include input variables, methodology, and the outputs obtained (Figs. 10 and 11; Mishra and Singh 2011).

Arab J Geosci

Conclusions It is anticipated that the future will witness increased dynamics in hydro-meteorological variables around the world which will lead to frequent droughts whose impacts will be compounded by growing water demands. Although significant progress has been achieved in drought modeling, much work remains ahead. This contribution provides a review of the methods used for modeling different components of droughts, which will be useful for different sectors dealing with water resources directly or indirectly. A drought is a multivariate event. Therefore, a better approach for describing drought characteristics is to derive the joint distribution of a drought based on its characteristics. The limitations in earlier approaches for deriving joint probability distributions are based on the sameness of marginal distribution for each drought variable. However, recent developments in the application of copulas offer a great deal of flexibility for multivariate drought characterization. However, the type of copulas differs, based on the characteristics of time series. Therefore, identifying a suitable copula will be required for multivariate drought characterization. The copula method can be further explored to characterize droughts based on the combination of different types of droughts for extracting better information with respect to different return periods. This combined approach can be a combination of the meteorological, agricultural, and hydrological droughts for multivariate drought characterizations. There is also a possibility for deriving different drought indices based on multiple types of droughts (Mishra and Singh 2011). Spatio-temporal drought analysis based on the combination of duration, severity, area, and interarrival time are critical for short- and long-term water management. There is much work done on this aspect; however, the gauged data used on spatial scale are unable to produce accurate results due to missing values as well as large distances between gauging stations. Therefore, the availability of remote sensing data will play a crucial role in overcoming these problems. Hence, regionalization of droughts based on remote sensing data needs to be explored. The linkage between large-scale atmospheric patterns and regional droughts can be another way for exploring space–time variability of droughts from local to regional scale, which needs to be investigated as future work. So far, the regionalization of droughts is based on hydro-meteorological variables; however, the major factor affecting droughts is growing water demand with limited natural source of water supply, hence there is scope for regionalization of droughts based on the spatial and temporal water demands. Even though substantial work has been done on different aspects of droughts, a proper approach to convey the results of research to decision makers is not as well articulated. There is a need to develop decision support systems under

climate change scenarios as well to quantify uncertainties for issuing warnings, assessing risk, and taking precautionary measures. This study has examined a number of literature sources; however, it seems to be virtually impossible to include in a review all publications. This review paper highlights an overall approach for drought risk assessment using remote sensing and GIS techniques. It is expected that these gaps could be filled by subsequent contributions and that there is scope for further discussion about drought research possibly in the broader context of future development of the entire hydrological science.

References Allaby M (2003) Draughts. Book, ISBN-13:7980816047932 AMS (American Meteorological Society) (1997) Meteorological drought policy statement. Bull Am Meteorol Soc 78:847–849 AMS (American Meteorological Society) (2004) Statement on meteorological drought. Bull Am Meteorol Soc 85:771–773 Bajgiran PR, Darvishsefat AA, Khalilic A, Makhdoum MF (2008) Using AVHRR-based vegetation indices for drought monitoring in the Northwest of Iran. J Arid Environ 72:1086–1096 Bordi I, Fraedrich K, Petitta M et al (2006) Large-scale assessment of drought variability based on NCEP/NCAR and ERA-40 reanalyses. Water Resour Manag 20(6):899–915. doi:10.1007/ s11269-005-9013-z Che S, Li C, Shen S (2010) Analysis of drought-flood spatial–temporal characteristics based on standard precipitation index in Hebei Province. Chin J Agrometeorol 31(1):137–143 (in Chinese) Dai A (2010) Drought under global warming: a review. WIREs Clim Chang 2:45–65. doi:10.1002/wcc.81 Dai A (2011) Characteristics and trends in various forms of the Palmer Drought Severity Index during 1900–2008. J Geophys Res 116: D12115. doi:10.1029/2010JD015541 Ding Y, Hayes MJ, Widhalm M (2010) Measuring economic impacts of drought: a review and discussion. Papers in Natural Resources. Paper 196. http://digitalcommons.unl.edu/natrespapers/196 Food and Agriculture Organization (1983) Guidelines: land evaluation for rainfed agriculture. FAO Soils Bulletin 52, Rome Feng P, Zhong X, Zhang B (2000) A drought degree evaluation method based on the artificial neural network. Syst Eng Theory Pract 20 (3):141–144 (in Chinese) Gebrehiwot T, Veen AVD, Maathuis B (2011) Spatial and temporal assessment of drought in the Northern highlands of Ethiopia. Int J Appl Earth Obs Geoinf 13:309–321 Ghulam A, Qin Q (2007) Modified perpendicular drought index (MPDI): a real-time drought monitoring method. ISPRS J Photogramm Remote Sens 62(2):150–164 Glantz MH (2003) Usable science: early warning systems: do’s and don’ts. Report of workshop, 20–23 October, Shangai, China Gregoric G, Sušnik A (2010) Drought Management Centre for South Eastern Europe. In: Global environmental change: challenges to science and society in southeastern Europe. doi:10.1007/978-90481-8695-2_20. Springer Science+Business Media B.V. 2010 and published in International Conference “Global Environmental Change: Challenges for Science and Society in South-Eastern Europe” held 19–21 May 2008 Gumbel EJ (1963) Statistical Forecast of Droughts. Bull Int Assoc Sci Hydrol 8(1):5–23

Arab J Geosci Hadish L (2010) Drought risk assessment using remote sensing and GIS: a case study in southern zones, Tigray Region, Ethiopia. Msc., Thesis Addis Ababa University, Addis Ababa, Ethiopia Han P, Wang PX, Zhang SY, Zhu DH (2010) Drought forecasting based on the remote sensing data using ARIMA models. Math Comput Model 51:1398–1403 Hayes MJ (2012) Drought indices, National Drought Mitigation Center (http://www.drought.unl.edu/whatis/indices.htm). With modifications by Dev Niyogi and Umarporn Charusambot, Indiana State Climate Office, Purdue University (http://iclimate.org/ 2.2.2012). Hennessy K, Fawcett R, Kirono D, Mpelasoka F, Jonesb D, Bathols J, Whetton P, Stafford Smith M, Howden M, Mitchell C, Plummer N (2008) An assessment of the impact of climate change on the nature and frequency of exceptional climatic events. Report by Climate and Ocean Services Bureau of Meteorology, CSIRO Climate Adaptation Flagship and Drought Policy Review Climate Change Division Department of Agriculture Fisheries and Forestry, Australian Hisdal H, Tallaksena L (2003) Estimation of regional meteorological and hydrological drought characteristics: a case study for Denmark. J Hydrol 281:230–247 Huo Z, Li S, Wang S et al (2003) Study on the major agriculture meteorological disaster risk assessment. J Nat Disasters 18 (6):692–702 (in Chinese) Jain SK, Keshri R, Goswami A, Sarkar A (2010) Application of meteorological and vegetation indices for evaluation of drought impact: a case study for Rajasthan, India. Nat Hazards 54:643–656. doi:10.1007/s11069-009-9493-x, Springer Science + Business Media B.V Karavitis CA, Alexandris S, Tsesmelis DE, Athanasopoulos G (2011) Application of the Standardized Precipitation Index (SPI) in Greece. Water J 3:787–805. doi:10.3390/w3030787 Kenny P (2008) Assessment of the social impacts of drought and government and non-government social support services. Issues Paper. http://www.daff.gov.au/__data/assets/pdf_file/0004/ 710644/social_panels_isues_paper.pdf Linsely RK Jr, Kohler MA, Paulhus JLH (1959) Applied hydrology. McGraw Hill, New York Livada I, Assimakopoulos VD (2007) Spatial and temporal analysis of drought in Greece using the Standardized Precipitation Index (SPI). Theor Appl Climatol 89:143–153. doi:10.1007/s00704005-0227-z, Printed in The Netherlands Mastrangelo AM, Mazzucotelli E, Guerra D, De Vita P, Cattivelli L (2012) Improvement of drought resistance in crops: from conventional breeding to genomic selection. In: Venkateswarlu et al (eds) Crop stress and its management: perspectives and strategies. Springer Science + Business Media B.V. doi:10.1007/978-94007-2220-0-7 McKee TB, Doesken NJ, Kleist J (1993) The relationship of drought frequency and duration to time scales. Preprints, 8th Conference on Applied Climatology, 17–22 January, Anaheim, CA, pp. 179–184 Mishra AK, Singh VP (2010) A review of drought concepts. J Hydrol 391:202–216 Mishra AK, Singh VP (2011) Drought modeling—a review. J Hydrol 403:157–175 MWD (Mediterranean Water Scarcity and Drought Working Group) (2007) Mediterranean Water Scarcity and Drought Report. Technical report-009. http://www.emwis.net/topics/WaterScarcity/ PDF/MedWSD_FINAL_Edition Nagarajan R (2009) Drought assessment. Capital Publishing Company, co-published by Springer Science + Business Nagarajan R (2010) Drought assessment. Earth Environ Sci 2010:160– 204. doi:10.1007/978-90-481-2500-5_5, Springer Science+Business Ni S, Gu Y (2005) Study on frangibility zoning of agricultural drought in China. Adv Water Sci 16(5):705–709 (in Chinese)

Owrangi MA, Adamowski J, Rahnemaei M, Mohammadzadeh A, Sharifan RA (2011) Drought monitoring methodology based on AVHRR images and SPOT vegetation maps. J Water Resour Prot 3:325–334 Palmer WC (1965) Meteorological drought. US Department of Commerce, Weather Bureau, Research Paper No. 45, p. 58 Palmer WC (1968) Keeping track of crop moisture conditions, nationwide: the new crop moisture index. Weatherwise 21:156–161 Paltineanu C, Mihailescu IF, Prefac Z, Dragota C, Vasenciuc F, Claudia N (2009) Combining the standardized precipitation index and climatic water deficit in characterizing droughts: a case study in Romania. Theor Appl Climatol 97:219–233 Quiring SM, Ganesh S (2010) Evaluating the utility of the Vegetation Condition Index (VCI) for monitoring meteorological drought in Texas. Agric For Meteorol 150:330–339 Quiring S, Nielsen-Gammon JW, Srinivasan R, Miller T, Narasimhan B (2007) Drought Monitoring Index for Texas. Final Technical Report submitted by the Texas A&M Research Foundation (RF468511) to the Texas Water Development Board on behalf of the project investigators (June 30, 2007) Schneider SH (1996) Encyclopedia of climate and weather. Oxford University Press, New York Schubert S, Koster R, Hoerling M (2007) Predicting drought on seasonal-to-decadal time scales. Bull Am Meteorol Soc 88:1625–1630 Shahid S, Behrawan H (2008) Drought risk assessment in the western part of Bangladesh. Nat Hazards J 46:391–413. doi:10.1007/ s11069-007-9191-5, Springer Science + Business Media B.V Singh RP, Roy S, Kogan F (2003) Vegetation and temperature condition indices from NOAAAVHRR data for drought monitoring over India. Int J Remote Sens 24(22):4393–4402 Tao J, Zhongfa Z, Shui C (2011) Drought monitoring and analyzing on typical karst ecological fragile area based on GIS. Procedia Environ Sci 10:2091–2096. Published by Elsevier Ltd. Selection. doi:10.1016/j.proenv.2011.09.326. Available online at www.sciencedirect.com UN Secretariat General (1994) United Nations convention to combat drought and desertification in countries experiencing serious droughts and/or desertification, particularly in Africa. Paris Wang PX, Wan ZM, Gong JY, Li XW, Wang JD (2003) Advances in drought monitoring by using remotely sensed normalized difference vegetation index and land surface temperature products. Adv Earth Sci 18(4):527–533 (in Chinese and cited from Han et al. 2010) Wilhite DA (2000) Drought as a natural hazard: concepts and definitions. In: Wilhite DA (ed) Droughts: a global assessment. Routledge, London, pp 3–18 Wilhite DA, Buchanan-Smith M (2005) Drought as hazard: understanding the natural and social context. In: Wilhite DA (ed) Drought and water crises: science, technology, and management issues. CRC Press, Taylor & Francis Group, Florida, pp 3–29 Wilhite DA, Glantz MH (1985) Understanding the drought phenomenon: the role of definitions. Water Int 10:111–120 Wilhite DA, Glantz MH (1987) Understanding the drought phenomena: the role of definitions. In: Wilhite DA, Easterling WE, Deobarah A (eds) Planning of drought: towards a reduction of societal vulnerability. Westview Press, Wood, pp 11–27 Wipulanusat W, Nakrod S, Prabnarong P (2009) Multi-hazard risk assessment using GIS and RS applications: a case study of Pak Phanang Basin. Walailak J Sci Tech 6(1):109–125 World Meteorological Organization (WMO) (1986) Report on drought and countries affected by drought during 1974–1985. WMO, Geneva, p. 118 Wu J, He B, Lu A, Zhou L, Liu M, Zhao L (2011) Quantitative assessment and spatial characteristics analysis of agricultural drought vulnerability in China. Nat Hazards 56:785–801

Arab J Geosci Yevjevich VM (1967) An objective approach to definitions and investigations of continental hydrologic droughts. Hydrologic Paper 23. Colorado State University, Fort Collins, CO Zhang D, Wang G, Zhou H (2011a) Assessment on agricultural drought risk based on variable fuzzy sets model. Chin Geogr Sci 21(2):167–175

Zhang X-P, Rong-Fang L, Sheng L, Quna F, Xiao-Xiao W (2011b) The study of dynamic monitor of rice drought in Jiangxi Province with remote sensing. Procedia Environ Sci 10:1847–1853 Zhou F (2005) Evaluation index system of regional drought and drought strategy. Shandong University, Jinan