A Novel Index for Agricultural Drought Measurement: Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI)

Abstract

Droughts are common across various climates, typically caused by prolonged decreases in rainfall. Several factors contribute to drought, including the temperature, wind speed, and relative humidity and the timing, amount, and intensity of rainfall during the growing season. This study introduces the Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI), a new index that combines soil moisture and evapotranspiration (calculated using the Penman–Monteith method) to enhance drought early warning systems. To validate the SERDI, we compared it with other established indices such as the Land Surface Temperature (LST), Vegetation Health Index (VHI), Normalized Difference Vegetation Index (NDVI), and Normalized Difference Water Index (NDWI), using metrics like the R-squared (R²), root mean square error (RMSE), mean absolute percentage error (MAPE), and p-value to assess the accuracy, data variability, and forecast conditions. The results showed a low RMSE and high R² between the SERDI and the LST and VHI, indicating a strong correlation. However, weaker correlations were observed between the SERDI and NDVI/NDWI, as shown by the lower R² and higher RMSE values in semi-arid areas. Regions across Iran, Iraq, Syria, Jordan, and Israel experienced mostly moderate to severe drought conditions, with a few areas in Iran and Syria showing normal conditions. The SERDI’s strong correlation with the LST and moderate correlation with the VHI can be attributed to the direct influence of the soil moisture and evapotranspiration on the surface temperature and vegetation health. On the other hand, the weaker correlation with the NDVI and NDWI is due to variability in the vegetation response, irrigation practices, and regional differences. This study concludes that the SERDI is an effective tool for the detection of drought based on soil moisture and evapotranspiration.

1. Introduction

The unpredictability of the drought severity is due to a variety of factors, including the frequency and spread of precipitation, evaporation requirements, and the ability of soils to store moisture. There are two common types of drought stress: intermittent drought, which is caused by periods without rain, and terminal drought, which happens at the end of the growing season when the soil moisture is depleted [1]. Droughts occur across various climates, from wet to dry regions, and are often linked to a significant decrease in rainfall over an extended period. The temperature, wind speed, and timing of rainfall during the growing season all play critical roles in determining the drought severity. Unlike aridity, which is permanent, droughts are usually temporary and can vary in intensity and duration [2].

To quantify the drought severity, several drought indices have been developed. These indices provide numerical values based on factors like precipitation, soil moisture loss, and changes in reservoir levels. Over time, advances in remote sensing have led to the development of more than 150 different drought indices [3,4,5]. Each index is tailored to specific climates or applications, combining data such as precipitation, temperature, and soil moisture data [3]. Different types of droughts require different input variables for forecasting. Meteorological drought indices like the Standardized Precipitation Index (SPI) rely on precipitation data, while hydrological drought indices use streamflow and reservoir levels to assess water shortages, and agricultural drought is measured using soil moisture and crop yield data. These indices can also be used to assess the impact of a drought and determine its intensity, duration, severity, and spatial extent [6]. Despite these advances, existing drought indices have limitations [4]. For instance, meteorological indices like the SPI do not account for soil moisture, which is a critical factor in agricultural drought. Hydrological indices often provide late warnings since they rely on the downstream effects of precipitation deficits. Vegetation-based indices like the Vegetation Health Index (VHI) and Land Surface Temperature (LST) can detect the impacts of drought on crops, but they miss early warning signs that could help to mitigate drought’s effects on agriculture.

Early drought indices, such as the Palmer Drought Severity Index (PDSI) [7], were designed to assess long-term meteorological drought by combining temperature and precipitation data. However, the PDSI struggled with short-term drought detection and regional variability. The Standardized Precipitation–Evapotranspiration Index (SPEI) was introduced to improve on the SPI by incorporating evapotranspiration, which captures the effect of the temperature on the water demand. While this was an important step forward, the SPEI still does not fully address the soil moisture availability, a crucial factor in agricultural drought [3,4].

Recent advancements in remote sensing have made it possible to assess both vegetation health and soil moisture through satellite imagery. Indices like the VHI and LST are commonly used for this purpose. However, they are limited in their ability to detect drought in its early stages because they focus on observable impacts rather than the underlying climatic conditions that lead to a drought [8,9]. As a result, there is a need for a more comprehensive index that integrates key variables like soil moisture and evapotranspiration to provide early warnings and better monitor drought conditions.

There are many indices for drought, but we still require a more accurate index for the detection of the drought onset, intensity, and frequency. Several indices use evapotranspiration and soil moisture for drought, but the Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI) combines both to detect droughts early due to high temperatures and low precipitation.

This study introduces the Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI), which combines soil moisture and evapotranspiration (calculated using the Penman–Monteith method) to offer a more accurate and timely assessment of drought, particularly in semi-arid regions. By leveraging both variables, the SERDI addresses the limitations of traditional drought indices, providing a more comprehensive tool for early drought detection. The SERDI’s integration of soil moisture makes it particularly effective in detecting agricultural drought, where water availability directly impacts crop growth. Additionally, the SERDI’s ability to measure both evapotranspiration and soil moisture gives it a clear advantage in regions where water scarcity is a critical issue.

The newly developed SERDI differs from traditional agricultural drought indices in several key ways. Unlike indices such as the SPI and the SPEI, which primarily rely on precipitation and temperature data, the SERDI uniquely integrates soil moisture and evapotranspiration, two critical variables that directly influence the agricultural drought severity. This combination allows the SERDI to provide a more accurate and timely assessment of the drought conditions, particularly in semi-arid regions, where water availability is limited. Furthermore, while indices like the VHI and LST capture drought impacts only after significant vegetation stress has occurred, the SERDI focuses on early detection, identifying moisture deficits before visible signs of drought emerge. This makes the SERDI a valuable tool for proactive drought management, offering a more nuanced reflection of agricultural drought compared to traditional indices, which often overlook the critical role of soil moisture in crop growth and yields.

2. Methodology

2.1. Study Area

The semi-arid regions in the Middle Eastern countries include Iran, Iraq, Syria, Jordan, Israel, and Turkey. It is located between 30 and 60 E and 28 and 40 N. Broad semi-arid areas lie in Iran; these dominate most of Central Iran and stretch from the east to the west of the country. Although the provinces’ names are used to show the drought’s occurrence, the entire province does not always have a semi-arid climate in many Middle Eastern countries. For example, only small parts of the Sulaimaniyah and Erbil provinces are located in semi-arid areas. The temperature and precipitation vary because the semi-arid region covers a huge part of the Middle East. It also has two different semi-arid zones, including cold and hot semi-arid areas. Most importantly, the Köppen classification is used to extract the semi-arid regions in Middle Eastern countries [10].

The Köppen climate classification system is commonly used to categorize the world’s climates based on the annual and monthly averages of the temperature and precipitation. Below are the general Köppen climate classifications for the semi-arid regions of the Middle East [10] (Figure 1).

• Iran: BSh (hot semi-arid climate), with some regions classified as BWk (cold desert climate).
• Iraq: BSk (cold semi-arid climate) and BWh (hot desert climate) in the western desert areas in other parts of the country.
• Syria: BSh (hot semi-arid climate) in most parts of the country, with some areas classified as BSk (cold semi-arid climate).
• Jordan: BSh (hot semi-arid climate) in most parts of the country, with some areas classified as BSk (cold semi-arid climate).
• Israel: BSh (hot semi-arid climate) in most parts of the country, with some areas classified as Csa (hot-summer Mediterranean climate).
• Turkey: BSh (hot semi-arid climate) in the southeastern part of the country, with some areas classified as Csa (hot-summer Mediterranean climate).

It is important to note that these classifications are general and that there can be significant variations in the climate within each country due to differences in topography, altitude, and other factors. In addition, climate patterns may be affected by global climate change, which causes variations in the average temperatures and precipitation patterns over time.

2.2. Data

Soil moisture and evapotranspiration data were used in this study. Both types of data were obtained from TerraClimate, which is a dataset on the monthly climate and climatic water balance for global terrestrial surfaces, developed by (“IDAHO_EPSCOR/TERRACLIMATE”). The TerraClimate data have a spatial resolution of ~4 km (1/24°). We used Google Earth Engine to collect data from TerraClimate. The soil moisture data were derived using a one-dimensional soil water balance model. Moreover, the reference evapotranspiration data were based on the ASCE Penman–Montieth method.

The data used for correlations were collected on a monthly basis from 2000 to 2020, while the maps generated using Google Earth Engine (GEE) were based on an annual time resolution. This approach ensured accurate long-term drought pattern analysis.

2.3. Methods

The Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI) is a new drought index that was developed to improve upon existing indices such as the Palmer Drought Severity Index (PDSI) and the Standardized Precipitation Evapotranspiration Index (SPEI). The SERDI takes into account both soil moisture and evapotranspiration (the combined water loss from plants and the soil surface) to provide a more comprehensive picture of drought conditions. This index will be useful when using satellite data to estimate both soil moisture and evapotranspiration, and it will allow for the more accurate and timely monitoring of drought conditions in future studies. The SERDI is calculated by comparing the current soil moisture and evapotranspiration values to long-term averages for the same time of year. If the current values are significantly lower than the long-term averages, a drought is indicated. One of the advantages of the SERDI is that it can be used to monitor drought conditions in areas where ground-based data are limited or nonexistent. The SERDI is a promising new drought index that has the potential to enable the more accurate and timely monitoring of drought conditions. The SERDI was used for the first time in the current study. It includes series equations, which are similar to the steps used for the Vegetation Health Index (VHI).

The Evapotranspiration Condition Index (PETCI) is a drought index that measures the availability of water for vegetation in a particular area. It is often used in conjunction with other drought indices, such as the Soil Moisture Condition Index (SMCI). The PETCI is based on the potential evapotranspiration (PET), which is the amount of water that would be lost from an area through evapotranspiration if there were unlimited water available. The PET is affected by factors such as the temperature, humidity, wind, and solar radiation. One of the advantages of the PETCI is that it provides information specifically about the availability of water for vegetation. This can help to improve drought monitoring and management, especially in areas where vegetation is a key resource. The following equation (Equation (1)) requires a minimum and maximum PET.

$$PETCI={\displaystyle \frac{PE{T}_{max}-PET}{PE{T}_{max}-PE{T}_{min}}}\times 100$$

(1)

where PETCI is the evapotranspiration condition index, and PET_max and PET_min are the minimum and maximum evapotranspiration.

The Soil Moisture Condition Index (SMCI) is a drought index that measures the soil moisture conditions in a particular area. One advantage of the SMCI is that it relies on actual soil moisture measurements rather than just precipitation data, which can be influenced by factors such as the temperature and wind. This makes it a more accurate indicator of the soil moisture conditions and it can help to improve drought monitoring and management. The SMCI is a useful tool for the monitoring of soil moisture conditions and drought conditions in the United States. It provides a more accurate picture of the soil moisture conditions than traditional precipitation-based indices and is widely used in agriculture and water management. The soil moisture required for this index is based on the following equation (Equation (2)):

$$SMCI={\displaystyle \frac{SM-S{M}_{min}}{S{M}_{max}-S{M}_{min}}}\times 100$$

(2)

where SMCI is the soil moisture condition index; SM_min and SM_max are the minimum and maximum soil moisture.

Lastly, the Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI) uses both the Soil Moisture Condition Index (SMCI) and the Evapotranspiration Condition Index (PETCI) in its calculations. By incorporating both indices, the SERDI provides a more comprehensive and accurate picture of drought conditions, taking into account both soil moisture and the availability of water for vegetation. The SERDI was developed to address some of the limitations of traditional drought indices. The SERDI has shown promising results in drought monitoring and prediction, particularly in areas where ground-based data are limited or unavailable. The Soil Moisture and Evapotranspiration Revealed Drought Index (SERDI) is calculated using the following equation (Equation (3)).

$$SERDI=\left(SMCI\times 0.5)+(PETCI\times 0.5\right)$$

(3)

where SERDI is the Soil Moisture and Evapotranspiration Revealed Drought Index; SMCI and PETCI are the Soil Moisture Condition Index and Evapotranspiration Condition Index; 0.5 indicates a coefficient to quantify the relative contributions of the SMCI and PETCI to the SERDI. In this study, the following classification scheme for drought monitoring is proposed (

Table 1. Classification of normal, wet, and dry conditions based on SERDI.

SERDI Class	Minimum Value	Maximum Value
Extreme Dry	≤0	≤10
Severe Dry	>10	≤20
Moderate Dry	>20	≤27
Normal	>27	≤38
Wet	>38	≤50
Extreme Wet	>50	100

).

The 0.5 coefficient in the equation represents an equal weighting between the SMCI and PETCI in the final SERDI value. This means that both soil moisture and evapotranspiration are considered equally important in determining drought conditions. Both soil moisture and evapotranspiration are critical indicators of drought. Soil moisture indicates the availability of water in the soil for plants, while evapotranspiration represents the demand for water by vegetation and the atmosphere. By giving them equal weight, the index balances both the supply (SM) and demand (ET) sides of the water equation.

In most climates, especially semi-arid regions, both factors typically exert an equal influence on drought conditions. Therefore, applying equal weighting is a simple yet effective way to represent these factors’ combined effect on drought.

In future work or in more specific case studies, one could explore whether changing these weights improves the predictive power of the SERDI in different climatic zones. For example, in very arid regions, evapotranspiration may have a greater influence, whereas, in humid areas, soil moisture could be the dominant factor.

The categorization of the drought severity in Table 1 is based on the SERDI, which integrates soil moisture and evapotranspiration as key indicators for the determination of drought conditions. These thresholds are designed to reflect varying levels of water availability and moisture stress, which are critical in both hydrological and agricultural drought contexts.

Extreme Dry (≤10): This category represents the most severe drought conditions, where the soil moisture and evapotranspiration levels are critically low. This threshold aligns with previous indices that associate extreme dryness with significant crop failures and ecosystem stress. Such conditions are often observed in long-term drought periods with minimal rainfall and high temperatures [11,12].

Severe Dry (>10 to ≤20): Severe drought is characterized by prolonged water deficits that begin to heavily impact agricultural productivity and water resources. This range was selected based on similar classifications found in drought indices like the SPEI, where substantial deviations from normal moisture levels denote severe drought stress [13,14].

Moderate Dry (>20 to ≤27): Moderate drought represents a level where drought conditions begin to have observable impacts on water resources and agriculture, but not to the same extent as severe or extreme drought. Indices like the SPI and SPEI have adopted comparable ranges to classify moderate drought, focusing on periods of sustained but manageable water deficits [15].

Normal (>27 to ≤38): Normal conditions reflect average levels of soil moisture and evapotranspiration, where ecosystems and agriculture function without significant moisture stress. This classification is supported by empirical observations of standard climatological conditions during non-drought periods [15,16].

Wet (>38 to ≤50): The wet category is associated with above-normal moisture levels, indicating surplus water availability. Such conditions often correspond to periods of excessive rainfall, which can benefit agriculture but may lead to flooding risks depending on the intensity of the water surplus [17].

Extreme Wet (>50 to 100): This class represents extreme wet conditions, typically corresponding to periods of excessive rainfall or sustained high soil moisture levels, which can saturate ecosystems and exceed the water-holding capacity of soils. Such thresholds are observed in flood-prone regions following prolonged wet periods [18].

3. Results

3.1. Spatial Distributions of Dry and Wet Conditions Based on SERDI

The SERDI was applied to semi-arid regions in the Middle East, focusing on the period from 2000 to 2020. Figure 2a–u illustrate the distribution of dry and wet conditions across different countries during this time.

From 2000 to 2002, severe drought affected large areas of Iraq and Iran, while some regions in Iran exhibited normal conditions, particularly in the mountainous areas. The severe drought expanded in Central Iran, while the moderate drought remained stable in Iraq. Turkey’s semi-arid regions experienced normal to extremely wet conditions, especially in 2001, with some reductions in wetness by 2002. Syria saw a mix of moderate dry and normal conditions, while Jordan and Israel were predominantly covered by moderate drought.

Between 2003 and 2005, the prevalence of normal conditions increased in the northwest and northeast of Iran, while that of wet and extremely dry conditions decreased. The severe droughts intensified in Iraq and Syria, particularly in the southern semi-arid regions. Turkey’s wet conditions remained mostly stable, with minor fluctuations in normal and wet conditions.

From 2006 to 2008, the prevalence of moderate dryness increased across the Middle East. The severe drought intensified in Central Iran, while the wet conditions declined in Northern Iran. Iraq and Syria were largely affected by moderate and severe droughts, with Turkey experiencing a decrease in wet conditions.

In the period from 2009 to 2011, severe dry conditions persisted in Iraq and Iran, with a moderate drought spreading across Syria, Jordan, and Israel. The wet conditions in Turkey fluctuated, with some increases in the west but a decline in the east.

From 2012 to 2014, the severe and moderate droughts increased across the Middle East, with extreme dryness peaking in Southeastern Iran in 2015. Iraq and Syria experienced rising drought severity, while wet conditions declined across the region.

By 2017, the severe drought had decreased significantly in Iran, Iraq, and Syria, with normal conditions increasing. In Turkey, the wet conditions improved in the west but remained stable in the east.

Finally, from 2018 to 2020, the prevalence of normal conditions increased across Iran, Iraq, and Syria, while the severe drought declined. The wet conditions persisted in parts of Iran, while the moderate and severe droughts continued in Israel and Jordan. Turkey saw some fluctuations in its wet conditions, particularly in its eastern semi-arid areas (Figure 2).

The application of the SERDI to monitor drought conditions in semi-arid regions across the Middle East from 2000 to 2020 reveals substantial variations in drought severity across multiple countries, including Iraq, Iran, Turkey, Syria, Jordan, and Israel.

Iraq: Between 2000 and 2002, large parts of Iraq were affected by severe drought, particularly in the south. However, normal conditions were observed in the west and some northern parts. From 2006 to 2008, the drought conditions worsened, with moderate and severe dryness dominating most semi-arid areas. Although the conditions slightly improved in 2017, the severe drought persisted in the southern regions. By 2020, normal conditions had returned to some parts of the north and west, but the severe drought remained prevalent in the southeast.

Iran: Central and Eastern Iran saw some of the most prolonged and severe drought conditions throughout the study period. From 2000 to 2002, the severe drought covered much of Central Iran, while an extreme drought affected the southeast. These patterns intensified from 2006 to 2008, when the extreme dryness expanded significantly. By 2020, the severity of the drought had eased in many areas, although severe dryness still persisted in the southeast, while some normal conditions emerged in the northeast.

Turkey: In contrast to Iraq and Iran, Turkey’s semi-arid regions experienced fluctuating conditions between normal and wet from 2000 to 2008. Extreme wet conditions were noted in the west during 2001, followed by a slight decrease in wet conditions by 2002. By 2017, most of Turkey’s semi-arid areas experienced normal conditions, but severe drought was observed in some southwestern regions by 2020.

Syria: From 2000 to 2002, moderate dry conditions dominated most of Syria’s semi-arid regions, particularly in the west and southeast. These patterns intensified from 2006 to 2008, as the severe drought became more widespread. By 2020, Syria continued to experience moderate dryness, particularly in the west and southeast, while normal conditions were limited to the north.

Jordan and Israel: Moderate dry conditions covered most of Jordan and Israel throughout the early 2000s. While Jordan’s semi-arid regions remained mostly moderate from 2006 to 2008, a few areas experienced normal conditions. However, by 2020, severe drought became more pronounced in Israel, while Jordan saw a mix of normal and moderately dry conditions.

Comparisons between these two periods, 2000–2002 and 2006–2008, have shown that severe drought conditions intensified significantly between these two periods, particularly in Iraq and Iran. While Iraq experienced moderate drought in the earlier period, by 2006–2008, severe dryness dominated large areas. Similarly, in Iran, the central and southeastern regions shifted from severe to extreme drought conditions, highlighting the worsening drought severity over time.

Conversely, Turkey maintained relatively wet and normal conditions during 2000–2002, but the prevalence of wet conditions slightly decreased by 2006–2008, particularly in the western and central areas.

Another two periods, 2012–2014 and 2018–2020, displayed a marked improvement in drought conditions in Iraq, Iran, and Syria, particularly during 2018–2020, compared to the peak dryness recorded in 2012–2014. The severe drought decreased significantly, and the normal conditions expanded in the northern and western regions of these countries.

In contrast, Jordan and Israel displayed persistent dryness throughout both periods, with only limited areas showing improvements in 2018–2020. Severe drought in Israel remained prevalent, whereas Jordan experienced a mix of normal and moderately dry conditions.

Turkey consistently showed better resilience to severe drought compared to Iraq and Iran throughout the study period. The semi-arid regions of Turkey oscillated between normal and wet conditions, with only localized severe droughts in certain years, such as 2020. This contrasts with Iraq and Iran, where prolonged and extensive severe drought was common.

Iran exhibited the most prolonged and severe drought across all periods, particularly in the southeast, where extreme dryness persisted even as other regions showed improvements.

Iraq’s southern regions also faced recurring severe drought, with only minor improvements by the end of the study period.

Turkey’s semi-arid regions consistently displayed better drought resilience, with frequent wet conditions noted in the west during the early 2000s and gradual normalization by 2017. However, isolated severe drought events were still recorded in the southwest by 2020.

Moderate drought dominated Jordan and Israel throughout the study period, with severe drought becoming more prevalent in Israel by 2020. This suggests a greater challenge in achieving recovery compared to other countries.

The drought intensity generally peaked in the Middle East between 2006 and 2014, followed by gradual easing in some regions, as evidenced by the expansion of normal conditions in Iraq, Iran, and Syria by 2020.

This analysis underscores the significant regional differences in the drought dynamics, with the northern and western regions of Iraq, Northeastern Iran, and Western Turkey showing recovery trends, while Southeastern Iran and parts of Syria and Israel remained persistently affected by severe drought.

3.2. Validation of SERDI Based on Common Drought Indices in Semi-Arid Areas in Middle Eastern Countries

To validate the SERDI, we compared it with other common drought indices in the semi-arid areas of the Middle East, including the LST, VHI, NDWI, and NDVI. The SERDI showed strong correlations with the LST (R² = 0.85) and VHI (R² = 0.72), indicating a close relationship, while moderate correlations were found with the NDWI (R² = 0.39) and NDVI (R² = 0.38). These results suggest that the LST and VHI better capture drought conditions as reflected by the SERDI compared to the NDWI and NDVI.

The RMSE between the SERDI and LST was the lowest (8.3), making the LST the most accurate predictor of the SERDI. Similarly, the VHI had a reasonably low RMSE (11.3), while the NDWI and NDVI showed higher error rates (16.7 and 16.9, respectively). The mean absolute percentage error (MAPE) analysis indicated that the LST and VHI provided reasonable forecasting accuracy, with values of 24.4% and 32.8%, respectively, while the NDWI and NDVI showed less reliable forecasts, with 61.8% and 60.2%.

Further statistical tests confirmed the significant relationships between the SERDI and the other indices, with p-values below the significance threshold (p ≤ 0.05), supporting the model’s robustness (Appendix A,

Table A1. Correlation matrix (Pearson correlation coefficients) for Bushehr Province.

	SERDI	VHI	NDVI	LST	NDWI
SERDI	1	0.850	0.615	−0.921	0.625
VHI	0.850	1	0.869	−0.926	0.770
NDVI	0.615	0.869	1	−0.665	0.678
LST	−0.921	−0.926	−0.665	1	−0.722
NDWI	0.625	0.770	0.678	−0.722	1

and Figure A1 and Figure A2).

3.2.1. Comparing SERDI with LST and VHI

The relationship between the SERDI and LST in various semi-arid provinces of Iran shows strong correlations, with R² values ranging from 0.819 to 0.905, indicating that the LST explains 82% to 90% of the variability in the SERDI. The strongest correlation is observed in Tehran (R² = 0.905), while the weakest is in West Azerbaijan (R² = 0.819). The correlation between the SERDI and VHI varies more, with R² values ranging from 0.404 in Golestan to 0.722 in Bushehr.

The RMSE values for the SERDI and LST are relatively low, ranging from 5.1 in Tehran to 9.95 in West Azerbaijan, suggesting good model accuracy. For the SERDI and VHI, the RMSE ranges from 9.7 to 18, indicating more variation in the prediction accuracy.

Regarding the MAPE, values below 10% indicate excellent forecasting reliability, while values between 20% and 50% indicate reasonable forecasts. Tehran shows useful forecasting, with an MAPE of 18.27% for the SERDI and LST, while other provinces show reasonable forecasts, with MAPE values slightly exceeding 20%. The forecasts for the SERDI and VHI are considered reasonable in most provinces, with values between 30.48% and 47.35%.

The p-values for all indices and provinces are statistically significant (p ≤ 0.05), suggesting that the observed relationships are not due to chance (

Table 2. The R², RMSE, MAPE, and p-values between the SERDI and the LST and VHI in semi-arid areas in Iran.

Province	LST					VHI
	R2	Adjusted R2	RMSE	MAPE	p-Value	R2	Adjusted R2	RMSE	MAPE	p-Value
Alborz	0.849	0.848	7.596	21.806	<0.0001	0.556	0.554	13.008	38.974	<0.0001
Bushehr	0.849	0.848	8.316	24.386	<0.0001	0.722	0.721	11.278	32.749	<0.0001
Fars	0.879	0.879	6.736	21.815	<0.0001	0.656	0.655	11.352	34.993	<0.0001
Golestan	0.857	0.856	7.375	22.290	<0.0001	0.404	0.401	15.043	47.351	<0.0001
Kermanshah	0.889	0.888	9.470	20.114	<0.0001	0.670	0.669	16.314	33.464	<0.0001
Tehran	0.905	0.905	5.100	18.268	<0.0001	0.656	0.655	9.721	30.479	<0.0001
West Azerbaijan	0.819	0.818	9.943	21.521	<0.0001	0.401	0.399	18.019	48.709	<0.0001

).

In Iraq, the SERDI demonstrates strong correlations with the LST, with R² values ranging from 0.87 to 0.89. The RMSE values range from 7.09 to 9.71, confirming the good prediction accuracy. The correlations with the VHI are moderate, with R² values between 0.68 and 0.77. The MAPE values for the SERDI and LST range between 20% and 50%, indicating reasonable forecasting, similar to the SERDI and VHI (

Table 3. The R², RMSE, MAPE, and p-values between the SERDI and the LST and VHI in the semi-arid areas of Iraq.

Province	LST					VHI
	R2	Adjusted R2	RMSE	MAPE	p-Value	R2	Adjusted R2	RMSE	MAPE	p-Value
Erbil	0.885	0.884	9.707	20.604	<0.0001	0.679	0.678	16.179	35.267	<0.0001
Kirkuk	0.893	0.892	7.088	21.633	<0.0001	0.770	0.769	10.389	32.712	<0.0001
Ninawa	0.870	0.870	9.175	22.761	<0.0001	0.725	0.724	13.341	32.400	<0.0001
Sulaimaniyah	0.878	0.878	9.640	20.452	<0.0001	0.750	0.749	13.807	26.926	<0.0001
Diyala	0.869	0.868	7.019	22.159	<0.0001	0.752	0.751	9.638	30.756	<0.0001

).

In Turkey, the SERDI shows strong correlations with the LST, particularly in Aksaray and Konya, where the R² values exceed 0.90. However, the correlation with the VHI is weaker, particularly in Van, where the R² is 0.086. The MAPE and RMSE values confirm the model’s accuracy with the LST, while larger errors are observed with the VHI (

Table 4. The R², RMSE, MAPE, and p-values between the SERDI and the LST and VHI in the semi-arid areas of Turkey.

Province	LST					VHI
	R2	Adjusted R2	RMSE	MAPE	p-Value	R2	Adjusted R2	RMSE	MAPE	p-Value
Aksaray	0.901	0.900	9.856	20.465	<0.0001	0.549	0.547	21.032	44.835	<0.0001
Ankara	0.871	0.871	10.439	19.683	<0.0001	0.107	0.103	27.521	94.300	<0.0001
Konya	0.903	0.902	9.785	20.661	<0.0001	0.549	0.547	21.069	46.413	<0.0001
Nigda	0.850	0.849	10.562	19.223	<0.0001	0.332	0.330	22.347	60.017	<0.0001
Van	0.760	0.759	14.755	25.052	<0.0001	0.086	0.083	28.724	90.012	<0.0001

).

In the semi-arid regions of Syria, the SERDI shows strong correlations with the LST, with R² values between 0.85 and 0.87, and moderate correlations with the VHI, ranging from 0.61 to 0.78. The RMSE values for the SERDI and LST are relatively low, ranging from 7.95 to 9.94, confirming the model’s accuracy in these regions. For the SERDI and VHI, the RMSE values are slightly higher, ranging from 9.89 to 14.34. The p-values for both indices indicate statistical significance (p ≤ 0.05), confirming the validity of the relationships (

Table 5. The R², RMSE, MAPE, and p-values between the SERDI and the LST and VHI in the semi-arid areas of Syria, Jordan, and Israel.

Province	Country	LST (R2)	LST (RMSE)	LST (MAPE)	p-Value (LST)	VHI (R2)	VHI (RMSE)	VHI (MAPE)	p-Value (VHI)
Aleppo	Syria	0.868	8.96	24.54	<0.0001	0.77	11.82	36.44	<0.0001
Al-Hasaka	Syria	0.852	9.94	22.56	<0.0001	0.73	13.41	29.93	<0.0001
Raqqa	Syria	0.86	7.95	23.91	<0.0001	0.784	9.89	31.85	<0.0001
Swieda	Syria	0.849	9	20.24	<0.0001	0.617	14.34	40.06	<0.0001
Karak	Jordan	0.879	6.78	23.43	<0.0001	0.742	9.91	38.06	<0.0001
Al-Balqa	Jordan	0.909	6.72	16.66	<0.0001	0.783	10.37	29.84	<0.0001
Hadarom	Israel	0.848	6.98	25.59	<0.0001	0.683	10.11	38.34	<0.0001

).

In Jordan, the SERDI also shows strong correlations with the LST, with R² values between 0.88 and 0.91. The VHI correlations are slightly lower, ranging between 0.74 and 0.78. The RMSE values between the SERDI and LST are low, ranging from 6.72 to 6.78, while the RMSE for the VHI ranges from 9.91 to 10.37. The p-values in all cases are statistically significant (p ≤ 0.05) (Table 5).

In Israel, the SERDI shows a strong correlation with the LST, with an R² of 0.85, and a moderate correlation with the VHI (R² = 0.68). The RMSE values are 6.98 for the LST and 10.11 for the VHI, with the p-values indicating statistical significance for both indices (Table 5).

3.2.2. Comparing SERDI with NDVI and NDWI

Table 6. The R², RMSE, MAPE, and p-values between the SERDI and the NDVI and NDWI in the semi-arid areas of Iran, Iraq, and Turkey.

Province	Country	NDVI (R2)	NDVI (RMSE)	NDVI (MAPE)	p-Value (NDVI)	NDWI (R2)	NDWI (RMSE)	NDWI (MAPE)	p-Value (NDWI)
Alborz	Iran	0.613	12.148	48.246	<0.0001	0.786	9.041	36.348	<0.0001
Bushehr	Iran	0.378	16.869	60.188	<0.0001	0.39	16.696	61.749	<0.0001
Fars	Iran	0.003	19.339	98.544	0.408	0.461	14.22	70.247	<0.0001
Golestan	Iran	0.549	13.08	56.741	<0.0001	0.349	15.723	67.645	<0.0001
Kermanshah	Iran	0.005	28.327	112.8	<0.0001	0.524	19.588	63.652	<0.0001
Tehran	Iran	0.476	12.001	53.953	<0.0001	0.69	9.231	39.778	<0.0001
West Azerbaijan	Iran	0.456	17.211	64.437	<0.0001	0.567	15.393	53.539	<0.0001
Aksaray	Turkey	0.107	29.584	119.815	<0.0001	0.42	23.829	81.562	<0.0001
Ankara	Turkey	0.021	28.806	106.617	<0.0001	0.38	22.939	73.006	<0.0001
Konya	Turkey	0.096	29.834	118.232	<0.0001	0.406	24.186	81.395	<0.0001
Nigda	Turkey	0.426	20.716	79.01	<0.0001	0.614	16.995	58.076	<0.0001
Van	Turkey	0.466	22.025	74.372	<0.0001	0.673	17.187	45.389	<0.0001
Erbil	Iraq	0.178	25.904	94.107	<0.0001	0.282	24.217	79.749	<0.0001
Kirkuk	Iraq	0.304	18.069	82.602	<0.0001	0.322	17.831	75.443	<0.0001
Ninawa	Iraq	0.186	22.96	88.806	<0.0001	0.27	21.745	80.048	<0.0001
Sulaimaniyah	Iraq	0.32	22.719	74.265	<0.0001	0.407	21.299	63.878	<0.0001
Diyala	Iraq	0.297	16.213	67.265	<0.0001	0.307	16.112	61.551	<0.0001

presents the R² values for the relationship between the SERDI and NDVI across various semi-arid regions. The Alborz province leads, with an R² of 0.613, indicating that the NDVI explains 61.3% of the variability in the SERDI. In contrast, the Kermanshah and Fars provinces have very low R² values, indicating no explanation of the variability by the NDVI. For the NDWI, Alborz (0.786), West Azerbaijan (0.567), and Tehran (0.690) show the strongest correlations, while Bushehr and Golestan have the weakest.

The RMSE values for the NDVI are the lowest in Alborz (12.148), Tehran (12.001), and Golestan (13.080). For the NDWI, the lowest RMSE values are also found in Alborz (9.041) and Tehran (9.231). The MAPE analysis indicates that the forecasts are generally inaccurate, except for those in Alborz and Tehran for the NDWI, which have reasonable forecasts. The p-values are statistically significant across most regions, except in Fars for the NDVI, where the relationship is not significant.

In Turkey, Nigda and Van have higher R² values, reaching 0.426 and 0.466 for the NDVI and 0.614 and 0.673 for the NDWI, respectively. However, high RMSE and MAPE values persist, indicating challenges in forecasting accuracy, with reasonable forecasts only in Van for the NDWI.

In Iraq, the R² values for the NDVI range from 0.178 in Erbil to 0.304 in Kirkuk, while the NDWI values are generally low, reflecting limited explanatory power. The high RMSE values suggest that the forecasts in these regions are also inaccurate (Table 6).

Table 7. The R², RMSE, MAPE, and p-values between the SERDI and the NDVI and NDWI in the semi-arid areas of Syria, Jordan, and Israel.

Province	Country	NDVI (R2)	NDVI (RMSE)	NDVI (MAPE)	NDWI (R2)	NDWI (RMSE)	NDWI (MAPE)	p-Value (NDVI)	p-Value (NDWI)
Aleppo	Syria	0.295	20.677	84.104	0.352	19.822	78.413	<0.0001	<0.0001
Al-Hasaka	Syria	0.213	22.923	85.992	0.239	22.482	81.525	<0.0001	<0.0001
Raqqa	Syria	0.189	19.182	83.193	0.293	17.891	73.512	<0.0001	<0.0001
Swieda	Syria	0.152	21.329	83.238	0.333	18.861	65.345	<0.0001	<0.0001
Karak	Jordan	0.158	17.921	87.293	0.321	16.077	71.991	<0.0001	<0.0001
Al-Balqa	Jordan	0.449	16.519	58.896	0.504	15.628	53.597	<0.0001	<0.0001
Hadarom	Israel	0.292	15.119	65.181	0.439	13.458	52.091	<0.0001	<0.0001

presents the R², RMSE, MAPE, and p-values for the semi-arid areas in Syria, Jordan, and Israel. Generally, the R² values indicate weak correlations between the SERDI and both the NDVI and NDWI, except for Al-Balqa province in Jordan, which shows higher R² values than the other provinces. The high RMSE and MAPE values across all regions suggest a poor fit and inaccurate forecasts for the relationships examined. Additionally, the p-values indicate statistical significance for the observed relationships in these countries.

4. Discussion

The results of the SERDI analysis showed that a significant portion of the semi-arid regions in Iraq and Iran experienced severe to moderate drought conditions, while some areas at higher elevations, such as Taurus, Alborz, and the Zagros Mountains, remained relatively normal. In Jordan, Syria, and Israel, moderate drought conditions were observed, with some parts of Israel facing severe drought. Turkey exhibited diverse conditions, ranging from normal to wet, depending on the region, with most of its semi-arid areas experiencing normal conditions.

Furthermore, the SERDI demonstrated a strong correlation with the Land Surface Temperature (LST) and a moderate correlation with the Vegetation Health Index (VHI) across all study areas, as indicated by the high R² values and low RMSE. However, the relationships between other indices, such as the NDWI and NDVI, varied. For example, moderate correlations between the NDWI and NDVI were observed in regions with above-normal precipitation, while weak correlations were found in low-precipitation regions like Alborz and Fars. These variations suggest that the correlation strength between the indices is influenced by the specific environmental conditions in each region. Additionally, strong correlations were observed in regions where the variables used in each index aligned more closely, such as semi-arid zones.

The prevalence of moderate and severe drought in semi-arid regions is largely due to the low average annual rainfall (166 mm) and high evapotranspiration rates that characterize these areas [19]. This climate naturally limits water availability, making drought more frequent and severe. The inclusion of soil moisture and evapotranspiration in the SERDI provides a more accurate representation of drought conditions, as these two variables directly influence water availability and ecosystem health.

Evapotranspiration is a critical factor in drought monitoring. During drought, increased evaporative demands and reduced moisture supplies exacerbate the effects of drought by rapidly depleting water resources. This interplay between evapotranspiration and drought occurs globally, affecting 44.4% of the months identified as drought periods [20]. Evapotranspiration is particularly significant during drought as it continues to diminish the already limited water reserves in lakes, streams, and soil [21]. Additionally, the drought severity can be determined by the imbalance between the water lost through evapotranspiration and the water uptake by plant roots, which serves as an indicator of plant drought stress [22].

Soil moisture plays a key role in influencing water availability and crop yields. Variations in soil moisture directly impact plant productivity, making deficiencies in soil moisture particularly significant for agriculture and water supplies [23]. Drought evaluation methods now utilize a combination of rainfall, soil moisture, and vegetation indices to assess drought more comprehensively, taking into account both temporal and spatial factors [24].

Several drought indices incorporate evapotranspiration and soil moisture as inputs, including the Aridity Anomaly Index [25], the Standardized Precipitation Evapotranspiration Index (SPEI) [26], the Evapotranspiration Deficit Index (ETDI) [27], the Aggregate Dryness Index (ADI) [28], the Evaporative Stress Index (ESI) [29], and the Vegetation Drought Response Index (VegDRI) [30]. For soil moisture, indices such as the Soil Moisture Anomaly (SMA) [31] and the Soil Moisture Deficit Index (SMDI) [27] are commonly used.

The strong correlation observed between the SERDI and both the LST and VHI can be explained by several factors that vary across semi-arid regions. The LST serves as a critical indicator of surface moisture, where higher temperatures typically correlate with lower soil moisture availability. In regions where the SERDI integrates soil moisture data effectively, the correlation with the LST is particularly robust, especially in areas with limited vegetation cover, where temperature variations are more directly influenced by the soil moisture levels. Conversely, regions with abundant vegetation may exhibit a weaker correlation due to factors such as shading, evapotranspiration rates, and diverse land-use patterns that can mask the underlying relationship.

Similarly, the correlation between the SERDI and VHI is closely tied to the vegetation’s response to water availability. In areas dominated by agricultural practices, the SERDI effectively captures changes in vegetation health linked to drought conditions, resulting in stronger correlations with the VHI. However, in regions with diverse land-use patterns, including urbanization or varying agricultural practices, this relationship may diminish. Regional climate characteristics, such as precipitation patterns, temperature extremes, and humidity levels, further influence these correlations. Areas that experience more frequent droughts may show a stronger relationship between the SERDI and both the LST and VHI, as drought conditions directly impact vegetation health and surface temperatures.

While the SERDI demonstrates strong correlations with the LST and VHI in many regions, the strength of these relationships is influenced by the land-use patterns, regional climate characteristics, and vegetation responses, which are crucial in accurately interpreting the index’s performance in different contexts.

The SERDI offers a promising tool for drought detection, especially in semi-arid regions, where evapotranspiration and soil moisture are critical indicators of the drought severity. Its strong performance in these areas suggests that it could be highly useful for agricultural drought monitoring, providing early warning signals for water scarcity and helping to inform water management strategies. However, the applicability of the SERDI to other climates, such as tropical or temperate zones, remains uncertain due to the unique environmental factors at play in these regions.

One limitation of the SERDI is the global availability of soil moisture and evapotranspiration data. These data are often derived from remote sensing platforms like Google Earth Engine (GEE), which may not cover all areas comprehensively. As a result, the index’s utility could be constrained in regions with limited data coverage. Additionally, while the SERDI has shown strong correlations with the LST and VHI, it is important to be cautious with high or statistically significant values of R², as they may not necessarily correspond to the magnitude of differences between indices. Small differences between indices can still occur even with low or negative R values [32].

Therefore, further research is required to validate the SERDI in diverse climatic zones to assess its broader applicability. Future studies should focus on refining the index for use in areas with different environmental characteristics, such as varying soil properties, vegetation types, and rainfall patterns, to ensure that the SERDI can become a more universally applicable tool for drought monitoring.

5. Conclusions

The SERDI was developed to detect drought by using soil moisture and evapotranspiration, which are obtained from TerraClimate. We used GEE for the SERDI in the semi-arid areas of Middle Eastern countries. The SERDI analysis provided insights into the drought conditions in the semi-arid regions of Iraq, Iran, Jordan, Syria, Israel, and Turkey. The majority of the semi-arid areas in Iraq and Iran experienced severe and moderate dryness, with some exceptions in elevated regions. The prevalence of dryness can be attributed to the low average annual rainfall in these areas. Jordan, Syria, and Israel experienced moderate drought, with sporadic areas of normal conditions and severe drought in parts of Israel. Turkey, on the other hand, had a range of normal, wet, and extremely wet conditions, with most of its semi-arid regions remaining in a normal state.

It is important to acknowledge that weak, moderate, and strong correlations between indices are normal due to their distinct methodologies and inputs. Therefore, the utilization of the LST and VHI can be valuable in detecting agricultural drought, as they exhibit high R² values and low RMSE values in various semi-arid regions. Furthermore, similar drought conditions were observed between the NDWI and NDVI in the northern part of Iran and most of Turkey’s semi-arid areas. It is important to examine this further and validate it with ground data in future studies.

While the SERDI has proven effective for drought monitoring in semi-arid regions, further validation in other climates is necessary to confirm its global applicability. Future studies should focus on testing the SERDI in different environments to ensure its reliability across diverse climatic conditions.