Assessing the Vulnerability of Water Resources to Drought by Applying the Standardized Drought Vulnerability Index to the Lake Karla Basin ^†

Abstract

Drought as an extreme weather phenomenon has recently become more frequent with significant impacts on water resources, such as reduced infiltration and surface runoff. To assess the vulnerability of water resources to drought, the Standardized Drought Vulnerability Index (SDVI) was applied to the hydrological basin of Lake Karla in Thessaly, Central Greece. The Lake Karla basin has a semi-arid climate and is an agricultural basin in which water-demanding crops are cultivated. The SDVI is a composite index that integrates all types of droughts and, with its holistic approach, can be used as a monitoring tool to provide knowledge for the delineation of vulnerable areas.

1. Introduction

Drought, a natural phenomenon associated with reduced precipitation and increased temperatures, is recognized as one of the costliest natural disasters [1]. As water availability is the main constraint to crop production in areas affected by drought and subsequent water scarcity, addressing this issue is essential [2].

The identification of drought is a complex process, typically based on the effects or impacts on various systems, including agriculture, water resources, and the economy, among others [3]. However, it should be noted that there is no physical variable which can be measured in order to quantify drought [4].

The increasing frequency of droughts in recent years, attributable mainly to climate change, has had significant impacts on water resources, as evidenced by a reduction in both infiltration and surface runoff [5,6]. The Mediterranean region is particularly vulnerable to the effects of drought. It has been asserted that this vulnerability is further substantiated by the region’s designation as a hot spot of significant concern [7].

The IPCC [8] defines vulnerability as “the propensity or predisposition to be adversely affected”, and states that it is the result of “diverse historical, social, economic, political, cultural, institutional, natural resource, and environmental conditions and processes”. The term ‘vulnerability’ can also be expressed as a function of two variables, namely ‘hazard’ and ‘impacts’ [9].

The vulnerability of water resources to drought is a complex process caused by several factors such as environmental, anthropogenic, and economic, especially for the Mediterranean semi-arid conditions [10]. A comprehensive understanding and assessment of drought vulnerability is key to developing appropriate drought management strategies and mitigating its impacts [11].

The present research aims to determine the impact of droughts on the vulnerability of water resources in the hydrological basin of Lake Karla in Thessaly, Central Greece. For this reason, the Standardized Drought Vulnerability Index (SDVI), a composite index that integrates all types of droughts (meteorological, hydrological, agricultural, and socio-economic), was used to assess the vulnerability of water resources on a single scale.

2. Study Area

The study area was the watershed of Lake Karla, in Thessaly. The wider region of Thessaly is an agricultural area, one of the most important in Greece. The role of water resources (surface and groundwater) is crucial for meeting the irrigation needs of this region [12].

This basin, covering a total area of 1170 km², is characterized by a semi-arid climate and the dominant cultivation by water-demanding crops (such as cotton, maize and alfa-alfa), and it is considered as one of the most productive agricultural regions of Greece. The elevation of the basin varies between 40 and 1980 m with an average altitude of about 230 m (Figure 1).

In the past, Karla Lake was one of the largest lakes in the country (4–6 m deep), as well as an important wetland. Following the draining of the natural lake in the 1960s, there has been severe depletion of the groundwater resource due to excessive exploitation for irrigation purposes [13]. In 2018, a reservoir was built up at the basin’s lowest part (with an average area of 38 km²) to prevent further degradation. After Storm Daniel that affected Thessaly in September 2023 and caused record-breaking rainfall in Greece, the lake tolerated its old extent.

The mean annual precipitation in the Karla basin ranges from 400.1 mm to 742.3 mm, with a mean of 547 mm and a standard deviation of 82.4. The distribution of rainfall is seasonally uneven (79% fall in the rainy season from October to May). The mean annual temperature is 14.8 °C, with a standard deviation of 1.5 °C. The range of temperatures is from 12.0 °C to 17.8 °C, with an upward trend of approximately 0.04 °C per year, which is not statistically significant.

From a geological point of view, the Karla basin belongs to the Pelagonian zone, whose hard rocks (marbles, gneisses, slates, ophiolites, etc.) form the bedrock of the basin. Quaternary and Neogene sediments have been deposited on the bedrock. In these sediments, which consist of an alternation of fine-grained (clays, silts) and coarse-grained (sands) materials, a multiple aquifer system is developed.

3. Methodology

To assess the vulnerability of water resources to the drought phenomenon, the Standardized Drought Vulnerability Index (SDVI) was applied, which is a composite index that integrates all types of droughts (meteorological, hydrological, agricultural, social, and economic). The SDVI was developed by the research team of the Agricultural University of Athens and was applied in Greece as well as in several countries of South-Eastern Europe [14,15].

The SDVI incorporates Water Supply, Water Demand, Infrastructure, Impact, and composite drought indices, such as cSPI-6 and cSPI-12, as sub-indices (Figure 2). It should be noted that this study presents a modified version of the SDVI estimating method, which is as follows. To apply the SDVI, the Supply, Demand, Infrastructure, Impact, and cSPI-6 and cSPI-12 sub-indices were initially assessed. For a homogeneous way of extracting conclusions, the results were recorded on a scale of values of 0–3. The estimation was made for the period from the 1987/88 hydrological year to 2017/18. This date was chosen as a starting point because 1987 is a landmark period of increased agricultural production with extensive cotton cultivation. In addition, population and crop records were not available for earlier years.

Drought indices are very useful because they recognize the episodes of drought in time and take negative values (drought) and positive values (wet years). In this work, based on precipitation data for the years of 1960–2018 from nearby stations, the Standardized Precipitation Index was applied to identify drought. The SPI is probably the best-known drought index with wide application in the international literature [16]. A drought begins when the index takes negative values and ends when it takes positive values. For the calculation of SPI, DrinC 1.2 software (Drought Indices Calculator), developed at National Technical University of Athens, was used [17]. The reclassified cSPI drought sub-indices (6 and 12) were derived from the SPI.

The Water Supply sub-index was calculated considering the surface water that is available through runoff and the amount of rainfall that infiltrates the soil and recharges the aquifer of the Karla basin. The conceptual hydrological model for water balance estimation UTHBAL [18] was applied to calculate the Supply sub-index. The UTHBAL model requires monthly values of mean temperature, precipitation, and potential evapotranspiration and produces values for actual evapotranspiration, soil moisture, natural groundwater recharge, and watershed surface runoff. The model separates the total precipitation into rainfall and snowfall, as the correct apportionment of precipitation is essential for modeling mass balance, seasonal snow cover, and accurate runoff simulation.

The Demand sub-index was calculated by considering urban, industrial, and agricultural water demands. The estimation of the crops’ needs was performed with the Near Irrigation Requirement (NIR) method. However, as the Karla basin has mainly an agricultural characteristic, the calculation is based on the difference between potential and actual evapotranspiration.

For the calculation of the Economic Impact sub-index, the reduction in mainly agricultural production was used. More specifically, the percentage difference (%) in agricultural production (cereals, wheat, maize, and cotton are the main crops cultivated in the Karla basin) between each hydrological year and the average agricultural production for the study period (1988–2018) were calculated. The corresponding data for each hydrological year were collected from the Hellenic Statistical Authority (ELSTAT).

The calculation of the Infrastructure sub-index was performed according to the actual capacity of the reservoirs (dams) and other pertinent infrastructure of the Karla basin in relation to the total water needs.

4. Results

From the application of the SPI, it is found that important drought episodes occurred between the years of 1988 and 1990, 1999 and 2002, and 2013 and 2014, when the rainfall was lower than the average in Greece and the Mediterranean region. Generally, it is observed that for the short-time thresholds (SPI-3 and SPI-6), more drought episodes are recorded; however, for the long-time thresholds (SPI-9 and SPI-12), the droughts last longer (Figure 3).

The SDVI results identify 15 hydrological years with high drought vulnerability values (values greater than 1.5), i.e., about half of the study period: 1987/88, 1988/89, 1989/90, 1990/91, 1991/92, 1992/93, 1993/94, 1996/97, 1999/00, 2000/01, 2003/04, 2004/05, 2010/11, 2012/13, and 2013/14 (Figure 4).

5. Conclusions

The aim of this work was to assess the vulnerability of water resources to drought and its impacts. The hydrological basin of Lake Karla, in Thessaly, North Greece, was chosen as the study area. The Standardized Drought Vulnerability Index was used to determine the vulnerability of water resources to drought. A modified method of applying the index was applied.

The SDVI results for the Lake Karla basin for 30 hydrological years (from 1978/88 to 2017/18) show that the index has a value greater than 1.5 for about 15 years, indicating a high vulnerability of water resources to drought in the Karla basin, whose main use is irrigation. It is pointed out that the Karla basin belongs to the water district of Thessaly, which is probably the most water-deficient district in the country. The availability of water resources (groundwater and surface water) does not meet the actual needs of crops for irrigation.

The established phenomenon of decreasing precipitation and increasing temperature will inevitably lead to more severe (extreme) and longer drought episodes in the coming decades. A drought forecasting and monitoring system should be developed to warn and detect the onset, intensity, and end of the phenomenon. SDVI, with its holistic approach, can be used as a monitoring tool for detecting drought and identifying vulnerable areas.

Continuous monitoring of the SDVI, which can correlate precipitation deficits with subsequent demand deficits, will contribute to the early diagnosis and vulnerability mitigation of drought in all its dimensions, provided the appropriate measures are taken, and the necessary infrastructure are built, including reservoirs, the artificial recharge of aquifers, and the reuse of treated wastewater for irrigation.

It is imperative that good agricultural practices are adopted and implemented in conjunction with the application of appropriate crop patterns. Furthermore, it is essential to implement measures for water conservation, including the use of appropriate irrigation systems to reduce irrigation water usage through the integration of water-saving techniques and the improvement of water transfer networks. Finally, the reuse of treated wastewater, as part of a circular economy, will provide water for irrigation.

The replacement of the SPI drought index (as a sub-index for the SDVI estimation) by other drought indices that take into account temperature (e.g., RDI) or even effective precipitation (e.g., aSPI, eRDI) [19] should be investigated, especially in rural areas such as the Karla basin. This may provide a more representative picture of the water availability situation in a region.