Seasonal variability of NDVI-LST relationship in Hyderabad, India

doi:10.21203/rs.3.rs-3202750/v1

Download PDF

Research Article

Seasonal variability of NDVI-LST relationship in Hyderabad, India

https://doi.org/10.21203/rs.3.rs-3202750/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

This study investigates the spatiotemporal and seasonal dynamics of NDVI, LST and their relationship in Hyderabad, India from 1993 to 2018, using Landsat data. Focused on the unique urban context, the research unveils distinct patterns in both NDVI and LST. Specifically, the highest NDVI values are observed during the monsoon and post-monsoon seasons, while elevated LST values are found in the pre-monsoon and post-monsoon seasons. This temporal analysis underscores the evolving nature of urban vegetation cover, indicating its stability and resilience over the study duration. Furthermore, the study highlights an observed increase in LST during the pre-monsoon and post-monsoon seasons, signifying potential risks such as heatwaves and extended mosquito breeding seasons. These observations align with the intensification of the urban heat island effect and a reduction in urban vegetation cover. Additionally, a negative correlation between LST and NDVI is established across all seasons and years, with the strongest negative correlation observed during the monsoon season and weakest during the winter season. The findings enrich existing knowledge on NDVI-LST dynamics, informing urban planning and management strategies to mitigate the adverse impacts of rapid urbanization on human health and the environment. The interdisciplinary approach, combining remote sensing techniques, geographic analysis, and urban environmental research, offers a comprehensive understanding of the NDVI-LST dynamics.

NDVI

Land Surface Temperature

Urban vegetation

Urban Heat Island

Urbanization is one of the most significant global environmental and social challenges of the twenty-first century (Seto et al., 2012), with more than half of the world’s population residing in cities (United Nations, 2018). The rapid expansion of urban areas is associated with several environmental and health problems, such as air and water pollution, heatwaves, and the loss of green spaces (Akbari et al., 2001; Hirano & Fujita, 2012; Schlink et al., 2006; Zhang et al., 2009). Urbanization affects the urban landscape by transforming natural habitats into artificial environments, leading to changes in land cover, land use, and vegetation cover. As a result, urban areas tend to have higher surface temperatures, known as the urban heat island (UHI) effect, which can have negative impacts on human health, energy consumption, and air quality (Oke, 1995; Voogt & Oke, 2003; Zhou et al., 2011). To monitor and understand these changes, remote sensing technology has emerged as a crucial tool for mapping and assessing urban environmental dynamics, including urban vegetation cover, land surface temperature (LST), and the impact of UHI effect.

In recent years, there has been a growing interest in using remote sensing techniques to monitor urban environmental changes (Chen et al., 2019; Hegazy & Kaloop, 2015; Weng et al., 2004a). Among the most commonly used remote sensing techniques are the Normalized Difference Vegetation Index (NDVI) and LST. NDVI is an indicator of the vegetation cover, while LST is an indicator of thermal conditions on the land surface (Zhang et al., 2009). Studies have shown that vegetation cover plays a crucial role in regulating urban temperatures, reducing the UHI effect, and mitigating heat stress (Liu & Zhang, 2011; Weng et al., 2004b, 2008). In contrast, high LST values are linked to increased heat stress, heat-related illness, and air pollution (Guhathakurta & Gober, 2007; Nandi & Swain, 2022; Tan et al., 2010). These two parameters are closely related, and their relationship can provide insights into the state of urban ecosystems, such as impact of climate change, urbanization, and other human activities on vegetation cover and surface temperature. Though several studies have investigated the relationship between NDVI and LST in urban areas (Govil et al., 2020; Guha et al., 2020; Sara Afrasiabi et al., 2013; Subzar Malik et al., 2019; Sun & Kafatos, 2007), limited research has been conducted on the seasonal variability of the NDVI-LST relationship in urban areas.

Hyderabad, the capital of the Indian state of Telangana, has experienced rapid urbanization in recent years, leading to significant changes in land use and land cover. Previous studies have reported a decline in vegetation cover and an increase in urban heat island effect in Hyderabad (Indira & Srinagesh, 2021; V. R. K. Sharma & Kaur, 2019). However, there is a lack of comprehensive understanding of the spatiotemporal and seasonal variability of NDVI and LST in Hyderabad.

The objective of this study is to analyse the spatiotemporal and seasonal variation of NDVI, LST and their relationship in Hyderabad city, India, from 1993 to 2018. This study aimed to identify the patterns across seasons and years of NDVI, LST distribution and their relationship in Hyderabad city. The results of this study can provide valuable insights for developing effective urban environmental management strategies to mitigate the impacts of rapid urbanization on human health and environment.

2.1 Study area

The study region for this research is Hyderabad, the capital and largest city of Telangana, India, which is also India’s sixth-largest urban agglomeration (Sannigrahi et al., 2017). The Greater Hyderabad Municipal Corporation (GHMC), which oversees the metropolitan region, is one of India’s largest municipal corporations, with a population of 7.9 million people and an area of 650 square kilometres. The GHMC boundary extends from 17°17΄30΄΄N to 17°33΄40΄΄N latitude and 78°14΄20΄΄E to 78°37΄20΄΄E longitude with an elevation of 462–635 meters. Figure-1 shows the map of the study area including the map of India (a), the map of the Telangana state (b) and the map of the GHMC boundary with its zones (c). GHMC has a total of 6 zones namely the Charminar zone, Khairatabad zone, Kukatpally zone, L.B.Nagar zone, Secunderabad zone and the Serilingampally zone.

The study region has a tropical wet and dry climate with four distinct seasons, namely pre-monsoon, monsoon, post-monsoon and winter (Parthasarathy et al., 1994). 20 years of climatic data have been provided in Table-1 (Source: Climate-Data.org). The pre-monsoon season extends from March through May, with average temperatures ranging from 28.2°C to 32.0°C and hot, dry weather. The monsoon season lasts from June to September, with August having the highest average monthly rainfall at 162mm. In the post-monsoon months of October and November, the weather is pleasant. Low pollution, moderate temperature, moisture content in plants and air, and a high percentage of green vegetation characterise the post-monsoon months (Guha & Govil, 2020a). Winter lasts from December to February, with December being the coldest month of the year.

Table 1

Climate data for Hyderabad City (1999–2019)

(Source: CLIMATE-DATA.ORG)
Weather elements	January	February	March	April	May	June	July	August	September	October	November	December
Maximum temp.(°C)	28.6	31.4	34.6	36.9	37.8	32.3	28.9	28.1	28.7	29.3	28.6	27.9
Minimum temp.(°C)	16.0	18.4	21.4	24.4	26.6	24.3	22.9	22.4	22.1	20.6	18.0	15.9
Mean temp.(°C)	22.2	24.9	28.2	30.7	32.0	27.9	25.6	24.9	25.0	24.7	23.2	21.8
Mean monthly rainfall (mm)	7.0	3.0	10.0	19.0	27.0	101.0	159.0	162.0	129.0	98.0	26.0	4.0
Average relative humidity(%)	50.0	43.0	37.0	38.0	37.0	62.0	74.0	77.0	78.0	69.0	59.0	53.0

The city witnessed growth rates of more than 50% during 1981–1991 and 27% during 1991–2001, and it is anticipated to grow to 11–12 million people by 2025, making it one of the megacities of the future (DESA, 2019; Kit et al., 2012). Hyderabad has reaped the benefits of India’s high-tech boom, with the information technology, pharmaceutical, and chemical industries accounting for the majority of the city’s GDP (Kit et al., 2012). Between 1975 and 1985, several industrial estates were developed, with the primary industrial sectors being chemical and bulk drugs, metals, timber, plastic, rubber, textiles and electronics. Hyderabad is also home to a number of multinational organizations and public sector enterprises, as well as universities and professional colleges (Taubenbock et al., 2007). The constant influx of people from all across the country exerted tremendous strain on the urban landscape, resulting in uncontrolled spatial development, reduced greenery, and excessive pollution levels. The city is also highly vulnerable to heatwaves, which have been intensified by unprecedented urbanisation and climate change (Government of Telangana (Revenue Department), 2021). The ‘Telanganaku Harithaharam (Greenery for Telangana)’ project, which was initiated in 2015 with the goal of increasing the state’s forest cover from 22.6–33.0%, is one of the several climate change mitigation measures the state has adopted. Around 2.5 million saplings were planted in Hyderabad through a huge plantation effort conducted by GHMC and Hyderabad Municipal Development Authority (HMDA) in 2016 as part of this project, which aims to increase the green belt within the city (Lonavath & Virugu, 2021). One of the programme’s main objectives is to conduct mass plantation activities in the city at various locations such as roadside avenues, rivers and canal banks, barren hills, institutional premises, religious places, housing colonies, community lands, graveyards, and industrial parks to make the city greener and less polluted.

2.2. Data descriptions and satellite data processing

A total of 12 radiometrically calibrated and atmospherically corrected Landsat Collection 2 Level 2 satellite data of different sensors from 1993 to 2018 are freely collected from the United States Geological Survey (USGS) data centre. Four Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensors (TIRS) data of 2018; four Landsat 7 Enhanced Thematic Mapper Plus (ETM+) data of 2008; and four Landsat 5 Thematic Mapper (TM) data of 1993 are selected for this study (Table-2). Based on the data available, 1993 was chosen as the base year for determining the LST and NDVI pattern in the city’s developing stage, 2008 was chosen to represent the city’s intermediate stage after development, and 2018 was chosen to represent the city’s current stage. During the morning time between 04:32 and 05:10 Greenwich Mean Time (10:02 to 10:40 Indian Standard Time (IST)), various Landsat satellite sensors pass over the Indian city of Hyderabad. The Collection 2 Level 2 Landsat datasets provide surface reflectance values for all the multispectral bands and surface temperature in kelvin (K) for the TIR bands. With band 11 reporting calibration uncertainty (Barsi et al., 2014), only the surface temperature of TIR band 10 is provided by the USGS for Landsat 8 and the surface temperature of TIR band 6 is provided for both Landsat 5 & 7.

Table 2

Description of Landsat datasets

(Source: USGS)
Season	Date of acquisition	Satellite sensor	Time (IST)	Cloud cover (%)
Pre-monsoon	21-March-1993	Landsat 5 TM	10:02:04	0.00
	23-April-2008	Landsat 7 ETM+	10:29:46	0.00
	27-April-2018	Landsat 8	10:38:52	0.01
Monsoon	29-Sep-1993	Landsat 5 TM	10:01:57	6.00
	30-Sep-2008	Landsat 7 ETM+	10:29:06	0.00
	30-June-2018	Landsat 8	10:38:43	4.85
Post-monsoon	16-Nov-1993	Landsat 5 TM	10:01:36	0.00
	01-Nov-2008	Landsat 7 ETM+	10:28:37	0.00
	21-Nov-2018	Landsat 8	10:39:36	0.70
Winter	17-Feb-1993	Landsat 5 TM	10:01:37	0.00
	18-Jan-2008	Landsat 7 ETM+	10:30:01	0.00
	05-Jan-2018	Landsat 8	10:39:48	1.79

Surface reflectance values for Landsat 8 were derived using the Land Surface Reflectance Code (LaSRC) algorithm version 1.5.0, while Landsat 5 and 7 employed the Land Ecosystem Disturbance Adaptive Processing System (LEDAPS) algorithm version 3.4.0 (Masek et al., 2006; Vermote et al., 2016). To generate the surface temperature product, all three Landsat sensors utilized the single-channel algorithm with various input data, including Top of Atmosphere (TOA) reflectance, TOA brightness temperature, ASTER Normalized Difference Vegetation Index (NDVI) data, and atmospheric profiles from the Goddard Earth Observing System (GEOS) model version 5 Forward Processing Instrument Teams (FP-IT) or Modern Era Retrospective analysis for Research and Applications version 2 (MERRA-2) (Cook et al., 2014; Laraby & Schott, 2018; Malakar et al., 2018; USGS, 2020). The Landsat Collection 2 Level 2 products adhere to the Committee on Earth Observation Satellite (CEOS) Analysis Ready Data for Land (CARD4L) standards and have undergone enhanced processing, geometric accuracy, and radiometric calibration compared to Collection 1 products. This internationally recognised certification ensures that the Collection 2 Level 2 products meet standardised requirements for immediate analysis (Committe on Earth Observation Satellite (CEOS), 2017; Digital Earth Africa, 2021). The spatial resolution for both multispectral surface reflectance bands and thermal infrared surface temperature bands for all three Landsat sensors is 30m x 30m. The surface temperature values in kelvin were converted to degrees Celsius (°C) using the Eq. (1).

$$LST\left(^\circ \text{C} \right)=LST-273.15$$

where LST(°C) is the land surface temperature in degrees Celsius and LST is the land surface temperature in kelvin. The NDVI is calculated using the near-infrared (NIR) and red spectral bands surface reflectance values as per Eq. (2).

$$NDVI= \frac{{\rho }_{NIR}- {\rho }_{Red}}{{\rho }_{NIR}+ {\rho }_{Red}}$$

where NDVI is the normalized difference vegetation index; ${\rho }_{NIR}$ is the surface reflectance values of band 5 for Landsat 8 and band 4 for Landsat 5 & 7; ${\rho }_{Red}$ is the surface reflectance values of band 4 for Landsat 8 and band 3 for Landsat 5 & 7.

2.3. Data analysis

The processing and extraction of NDVI and LST values for overall Hyderabad city as well as the individual GHMC zones were done using the software ArcMap 10.4. The spatiotemporal and seasonal variations of NDVI and LST distribution were analysed using statistical techniques. Descriptive statistics such as mean, minimum, maximum, and standard deviation were calculated for NDVI and LST in each zone and for each year. The relationship between NDVI and LST were examined through correlation analysis to investigate the seasonal variation of their relationship.

3.1. Spatiotemporal and seasonal variation of NDVI distribution

The study analysed the spatiotemporal and seasonal variation of NDVI distribution over the time period 1993 to 2018 in Hyderabad, India. Table-3 displays the temporal and seasonal variation of NDVI values. It was observed that the maximum NDVI values gradually increased in all seasons between 1993 and 2018, suggesting a possible improvement of vegetation cover in the urban areas of Hyderabad. One potential reason for this increase could be the implementation of Telanganaku Haritaharam program, which aims to increase green cover in urban areas of Telangana by planting millions of trees across the state. While we cannot definitely attribute the increase in maximum NDVI values to this program, it is possible that it has contributed to the observed trend.

The monsoon season had the highest maximum NDVI values (0.85 in 1993, 0.87 in 2008, and 0.98 in 2018), followed by post-monsoon (0.83 in 1993, 0.85 in 2008, and 0.87 in 2018), winter (0.79 in 1993, 0.84 in 2008, 0.86 in 2018) and pre-monsoon (0.71 in 1993, 0.78 in 2008, and 0.84 in 2018). A similar trend was observed for mean NDVI values as well with the highest being observed during monsoon followed by post-monsoon. This could indicate that the urban vegetation in Hyderabad is largely dependent on rainfall.

The mean NDVI values increased in the pre-monsoon and winter seasons from 1993 to 2018, but declined in the monsoon and post-monsoon seasons. It suggests that the vegetation cover in the urban areas of Hyderabad city may be influenced by seasonal changes and climatic factors such as temperature, precipitation, and sunlight availability. This information could be useful for urban planners to implement appropriate measures for sustainable management of urban green spcaes. For instance, during the monsoon and post-monsoon seasons, there could be a need to increase the maintenance activities such as watering, fertilizing, and pruning to maintain the vegetation cover (Chandel & Chauhan, 2023; Palmate et al., 2017). Similarly, during the pre-monsoon and winter seasons, efforts could be made to promote the growth of vegetation cover through planting new trees and shrubs, providing sufficient sunlight, and protecting existing vegetation cover (Brahma et al., 2023).

Table 3

Temporal and seasonal variation of NDVI (1993–2018)
Season	Date	NDVI(min.)	NDVI(max.)	NDVI(mean)	NDVI(std.)
Pre-monsoon	21-March-1993	-0.17	0.71	0.20	0.10
	23-April-2008	-0.15	0.78	0.30	0.14
	27-April-2018	-0.18	0.84	0.27	0.12
Monsoon	29-Sep-1993	-0.37	0.85	0.46	0.17
	30-Sep-2008	-0.30	0.87	0.40	0.20
	30-June-2018	-0.19	0.98	0.36	0.18
Post-monsoon	16-Nov-1993	-0.29	0.83	0.34	0.13
	01-Nov-2008	-0.30	0.85	0.33	0.16
	21-Nov-2018	-0.24	0.87	0.31	0.14
Winter	17-Feb-1993	-0.27	0.79	0.22	0.10
	18-Jan-2008	-0.33	0.84	0.20	0.11
	05-Jan-2018	-0.23	0.86	0.27	0.14

The seasonal difference in mean NDVI values decreased significantly from 0.26 in 1993 to 0.09 in 2018, indicating minimum seasonal fluctuation in urban vegetation in recent years. This suggests that the urban vegetation cover in Hyderabad has become more stable and resilient to seasonal variations, which can have significant ecological and environmental benefits, such as improved air and water quality, reduced urban heat island effects, and increased biodiversity (Adosi, 2007; Al-Kindi et al., 2023; Baniya et al., 2022; Wang et al., 2016).

In terms of spatial distribution, there was a decline in NDVI values in major regions of the L.B.Nagar zone (east), Serilingampally zone (west), and Kukatpally zone (north) between 1993 and 2018, as shown in Figure-2(a) to Figure-2(c). The annual average NDVI statistics for the six GHMC zones are shown in Table-4. They are calculated by averaging the four images from each season in a given year. Although the maximum NDVI value increased in all six zones, the mean NDVI values remained almost constant during the study period. The increase in maximum NDVI value could indicate an overall improvement in the quality of vegetation in the city. These results suggest that the spatiotemporal variation of NDVI distribution could be influenced by a complex interplay of factors such as urbanization, land use changes, and climate variability (Gao et al., 2023). The observed patterns of NDVI distribution can provide useful insights for urban planners and policymakers in identifying areas that require special attention for green cover enhancement and sustainable urban development.

Table 4

Annual variation of NDVI across the zones of GHMC (1993–2018)
Zone	Year	NDVI(min.)	NDVI(max.)	NDVI(mean)	NDVI(std.)
Charminar	1993	-0.08	0.73	0.29	0.11
	2008	-0.07	0.74	0.30	0.15
	2018	-0.10	0.78	0.29	0.15
Khairatabad	1993	-0.01	0.70	0.26	0.11
	2008	-0.05	0.74	0.26	0.14
	2018	-0.14	0.76	0.27	0.15
Kukatpally	1993	-0.13	0.70	0.31	0.09
	2008	-0.06	0.77	0.31	0.14
	2018	-0.09	0.79	0.31	0.13
L.B.Nagar	1993	-0.10	0.76	0.34	0.11
	2008	0.01	0.79	0.34	0.15
	2018	-0.05	0.80	0.33	0.13
Secunderabad	1993	0.02	0.73	0.26	0.11
	2008	-0.002	0.74	0.25	0.12
	2018	-0.03	0.79	0.28	0.13
Serilingampally	1993	-0.19	0.71	0.32	0.09
	2008	-0.12	0.75	0.34	0.13
	2018	-0.07	0.79	0.32	0.13

3.2. Spatiotemporal and seasonal variation of LST distribution

The spatiotemporal and seasonal variation of LST was analysed for Hyderabad city from 1993 to 2018 (Figure-4). Table-5 shows the temporal and seasonal variation of LST values over the time period studied. The annual average mean LST values increased from 36.22°C in 1993 to 38.14°C in 2018, indicating an overall rise in LST over the study period. The highest mean LST value (41.36°C) was recorded during the pre-monsoon season, whereas the lowest mean LST value (31.69°C) was recorded during the winter season. The mean LST values during the monsoon and post-monsoon seasons were 40.86°C and 34.94°C, respectively. The highest and lowest mean LST were observed on 23-April-2008 (45.10°C) and 05-Janurary-2018 (29.20°C) respectively.

The study found that the mean LST increased by 1.92°C from 1993 to 2018, although this increase was not seen all four seasons. During the years 1993–2018, the city’s mean LST rose by 7.48°C in the pre-monsoon and 6.33°C in the post-monsoon seasons, but fell by 1.24°C in the monsoon and 4.90°C in the winter seasons (Figure-3). This implies that the warming trend in the city is more pronounced during the dry seasons, particularly in the pre-monsoon season, which has important implications for urban heat management strategies.

Moreover, the study found an inverse relationship between LST and NDVI, as the seasonal difference in mean LST values increased from 11.87°C in 1993 to 14.08°C in 2018, whereas the NDVI seasonal difference decreased from 0.29 in 1993 to 0.19 in 2018. This suggests that urbanization and land-use changes have led to increased urban heat island effects and reduced vegetation cover over time.

Table 5

Temporal and seasonal variation of LST in Degree Celsius (1993–2018)
Season	Date	LST(min.)	LST(max.)	LST(mean)	LST(std.)
Pre-monsoon	21-March-1993	22.38	45.84	35.80	2.78
	23-April-2008	30.09	59.56	45.10	3.09
	27-April-2018	29.26	53.78	43.28	2.84
Monsoon	29-Sep-1993	24.01	58.86	43.51	2.87
	30-Sep-2008	26.75	54.23	36.94	3.02
	30-June-2018	20.86	54.72	42.27	3.07
Post-monsoon	16-Nov-1993	22.46	41.80	31.64	2.18
	01-Nov-2008	21.97	48.91	35.29	2.60
	21-Nov-2018	26.29	48.53	37.97	2.33
Winter	17-Feb-1993	22.30	43.41	34.10	2.60
	18-Jan-2008	21.47	46.61	31.86	2.53
	05-Jan-2018	21.65	37.48	29.20	1.70

The study also found that the mean LST increased in all six GHMC zones between 1993 and 2018, with the biggest jump in Charminar zone (2.56°C), followed by L.B.Nagar zone (1.91°C), Serilingampally zone (1.89°C), Kukatpally zone (1.83°C), Khairatabad zone (1.74°C), and Secunderabad zone (1.35°C). The seasonal pattern of LST is similar across all six zones of GHMC, with winter recording the lowest LST values and pre-monsoon recording the highest LST values, with the exception of Charminar, Khairtabad, and Secunderabad zones, where the highest mean LST values are reported during the monsoon seasons. The seasonal variance of mean LST in the L.B.Nagar zone is the smallest, with 8.77°C, and the Secunderabad has the largest variance, with 10.73°C. The seasonal variance of mean LST in Charminar zone is 10.13°C, Khairatabad zone is 10.51°C, Kukatpally zone is 9.48°C, and Serilingampally zone is 10.46°C.

The observed rise in LST values has several implications. Higher LST values during pre-monsoon season can result in an increased risk of heatwaves, which can cause various health issues (Guha & Govil, 2021; M. Sharma et al., 2022). In Hyderabad, there have been several instances of severe heatwaves in the past decade, causing heatstroke, dehydration, and even death among vulnerable population. The heatwave in May 2015 was one of the worst in recent history, claiming over 2,000 lives in Telangana alone (Nandi & Swain, 2022; Rathi & Sodani, 2021; Wehner et al., 2016). The increase in LST values during the post-monsoon season can lead to a longer mosquito breeding season, which in turn can increase the risk of mosquito-borne diseases such as dengue fever and malaria (Deichstetter, 2017). According to a report by the National Vector Borne Disease Control Programme (NVBDCP), Hyderabad witnessed a sharp rise in the number of dengue fever cases from 2014 to 2019, with a peak in 2017, where over 4,000 cases were reported (Gupta et al., 2012; Moulika, 2021; News18, 2019; Pilot et al., 2020). Similarly, malaria cases were also increased from 2015 to 2018, with a peak in 2016, where over 1,600 cases were reported (Arvind, 2022; HansIndia, 2022; Shrivastava & Saurabh, 2020; TNN, 2016). The rise in LST values during the post-monsoon season could be a contributing factor to the increase in these diseases, as it creates conducive environment for mosquitoes to breed and thrive. Furthermore, the variations in LST values across different zones of GHMC can be useful in identifying areas that are more susceptible to temperature-related health issues and designing appropriate mitigation strategies.

3.3. Seasonal variation of LST-NDVI relationship

The study investigated the seasonal variation of the LST-NDVI relationship for Hyderabad city from 1993 to 2018 across four seasons. Figure-5 depicts the LST-NDVI relationship for three years (1993, 2008, 2018) across four seasons. The LST-NDVI correlation coefficients were found to be negative in all seasons and years, indicating an inverse relationship between LST and NDVI. Moreover, the correlation coefficients drifted towards zero, suggesting a weakening of the negative relationship between LST and NDVI. This finding implies that urban vegetation in Hyderabad city is becoming less effective at moderating local temperatures, which could have negative consequences for urban heat island effects and overall urban climate resilience.

Seasonally, the strongest negative correlation coefficients were observed during the monsoon season in all three years, with − 0.53 in 1994, -0.59 in 2008 and − 0.27 in 2018. In contrast, the weakest negative correlation coefficients were observed during the winter season in all three years, with − 0.35 in 1993, -0.08 in 2008 and − 0.05 in 2018. In pre-monsoon season, the LST-NDVI correlation coefficient values were − 0.42 in 1993, -0.43 in 2008 and − 0.16 in 2018. In post-monsoon season, the correlation coefficient values of LST-NDVI relationship were − 0.22 in 1993, -0.28 in 2008 and − 0.13 in 2018. The results indicate that the strength of the relationship between LST and NDVI is highest during the monsoon season, followed by pre-monsoon and post-monsoon seasons, and is weakest during the winter season. During the growing season (monsoon), an increase in vegetation cover due to rainfall can lead to a reduction in LST. In contrast, during the non-growing season (winter), decline in vegetation leads to less vegetation cover, leading to an increase in LST. Also, the soil moisture in non-growing season has a dominant effect on LST, leading to a weak relationship between LST and NDVI (Sun & Kafatos, 2007).

The study’s results are consistent with past literature, which suggests that seasonal factors such as precipitation, temperature, and vegetation health, and growth patterns influence the strength of the LST-NDVI relationship(Govil et al., 2020; Guha & Govil, 2020b; Nikkala et al., 2022; Subzar Malik et al., 2019). For instance, Meng et al., 2022 observed a strong correlation between LST and NDVI during the monsoon and pre-monsoons seasons than the winter season in Beijing, China. Sun & Kafatos, 2007) reported a positive correlation between LST and NDVI in winter season but a negative correlation in pre-monsoon season in North America. Similary, Sara Afrasiabi et al., 2013 found that the correlation between LST and NDVI was negative in Mashhad, Iran. In Raipur city, India, Guha & Govil, 2020b found that the correlation varied across seasons and was weakest for the winter season compared to others, similar to the findings of this study.

In conclusion, this study examined the spatiotemporal and seasonal variation of NDVI, LST and their relationship in Hyderabad city, India from 1993 to 2018. The findings reveal an increase in maximum NDVI values in all seasons over the years, indicating a potential improvement in vegetation cover in the urban areas of Hyderabad. The monsoon season showed the highest maximum and mean NDVI values, suggesting the dependence of urban vegetation in Hyderabad on rainfall. Mean NDVI values increased in the pre-monsoon and winter seasons but declined in the monsoon and post-monsoon seasons, indicating the influence of seasonal changes and climatic factors on vegetation cover. However, the seasonal difference in mean NDVI values decreased significantly over time, indicating a more stable and resilient urban vegetation cover. Spatially, a decline in NDVI values was observed in major regions of the city, potentially influenced by urbanization, land use changes, and climate variability.

The increase in LST values from 1993 to 2018, particularly during dry seasons such as pre-monsoon, poses a significant risk of heatwaves in Hyderabad city. Furthermore, higher LST values during the post-monsoon season may prolong the mosquito breeding season, leading to increased risk of mosquito-borne diseases. The observed rise in LST values suggests that urbanization and land-use changes have contributed to increased urban heat island effects and reduced vegetation cover over time. The negative correlation between LST and NDVI in all seasons and years, with the strongest negative correlation during the monsoon season and weakest during the winter season, indicates that urban vegetation in Hyderabad city is becoming less effective at mitigating local temperatures. Understanding the seasonal variation of the LST-NDVI relationship is crucial in developing effective urban heat mitigation strategies that can improve urban climate resilience in Hyderabad and similar urban areas.

First, the study relied on remote sensing data, which has inherent limitations such as its resolution and the influence of atmospheric factors. Secondly, the study focused only on NDVI and LST as indicators of vegetation cover and heat stress, without considering other important environmental factors such as air pollution, humidity, and wind patterns. Third, the study was limited to Hyderabad city area and did not include data from surrounding rural areas or other cities in India. Fourth, the study period only covered 25 years, which may not be sufficient to capture long-term trends in vegetation and heat stress. Finally, the study did not investigate the underlying drivers of the observed trends, such as urbanization, land use change, and climate change, which could provide valuable insights for policymakers and urban planners. Despite these limitations, the findings of this study contribute a better understanding of the spatiotemporal and seasonal dynamics of urban vegetation and heat stress, which can inform the development of effective management and mitigation strategies to enhance urban resilience and sustainability.

Ethics approval

Not applicable

Competing interests

The author(s) declare they have no conflict of interests.

Author contributions

As the sole author of this manuscript, I declare that the research was conducted in accordance with ethical principles, and all necessary ethical considerations were taken into account during the study. The author(s) declare they have no conflict of interest.

Funding

Not applicable

Availability of data and material

Data is openly available from USGS website.

Adosi JJ (2007) Seasonal variation of carbon dioxide, rainfall, NDVI and it’s association to land degradation in Tanzania. Environ Sci Eng (Subseries: Environ Sci 9783540724377:373–389. https://doi.org/10.1007/978-3-540-72438-4_20/COVER
Akbari H, Pomerantz M, Taha H (2001) Cool surfaces and shade trees to reduce energy use and improve air quality in urban areas. Sol Energy 70(3):295–310. https://doi.org/10.1016/S0038-092X(00)00089-X
Al-Kindi KM, Nadhairi R, Al, Akhzami S, Al (2023) Dynamic Change in Normalised Vegetation Index (NDVI) from 2015 to 2021 in Dhofar, Southern Oman in Response to the Climate Change. Agric 2023 13(3):592. https://doi.org/10.3390/AGRICULTURE13030592. 13
Arvind N (2022) A Study of Malaria in Telangana. Telangana J IMA 2(2). https://doi.org/10.52314/tjima.2022.v2i2.88
Baniya B, Tang Q, Chhipi-Shrestha G, Baniya HB, Haile GG (2022) Vegetation Dynamics, Land Use and Ecological Risk in Response to NDVI and Climate Change in Nepal. Clim Risk Sustainable Water Manage 160–178. https://doi.org/10.1017/9781108787291.011
Barsi JA, Schott JR, Hook SJ, Raqueno NG, Markham BL, Radocinski RG (2014) Landsat-8 Thermal Infrared Sensor (TIRS) Vicarious Radiometric Calibration. Remote Sens 2014 6(11):11607–11626. https://doi.org/10.3390/RS61111607. 6
Brahma H, Pant S, Micheli L, Smestad GP, Sarmah N (2023) Effect of Environmental Factors on Photovoltaic Soiling: Experimental and Statistical Analysis. Energies 16(1):45. https://doi.org/10.3390/EN16010045/S1
Chandel V, Chauhan T (2023) Attributing Vegetation Recovery During the Indian Summer Monsoon to Climate Drivers in Central India. Ecol Econ Soc 6(1):109–122. https://doi.org/10.37773/EES.V6I1.927
Chen T, Lang W, Li X (2019) Exploring the Impact of Urban Green Space on Residents’ Health in Guangzhou, China. J Urban Plan Dev 146(1):05019022. https://doi.org/10.1061/(ASCE)UP.1943-5444.0000541
Committe on Earth Observation Satellite (CEOS) (2017) Analysis Ready Data For Land
Cook M, Schott JR, Mandel J, Raqueno N (2014) Development of an operational calibration methodology for the Landsat thermal data archive and initial testing of the atmospheric compensation component of a land surface temperature (LST) product from the archive. Remote Sens 6(11):11244–11266. https://doi.org/10.3390/RS61111244
Deichstetter P (2017) The Effect of Climate Change on Mosquito-Borne Diseases. Am Biology Teacher 79(3):169–173. https://doi.org/10.1525/ABT.2017.79.3.169
DESA U (2019) World Population Prospects 2019. United Nations. Department of Economic and Social Affairs. World Population Prospects 2019
Digital Earth Africa (2021) Landsat Collection 2 Level-2 Surface Reflectance. https://docs.digitalearthafrica.org/en/latest/data_specs/Landsat_C2_SR_specs.html
Gao S, Dong G, Jiang X, Nie T, Guo X (2023) Analysis of factors influencing spatiotemporal differentiation of the NDVI in the upper and middle reaches of the Yellow River from 2000 to 2020. Front Environ Sci 10:2584. https://doi.org/10.3389/FENVS.2022.1072430
Government of Telangana (Revenue Department) (2021) Telangana State HeatwaveAction Plan – 2021
Govil H, Guha S, Diwan P, Gill N, Dey A (2020) Analyzing Linear Relationships of LST with NDVI and MNDISI Using Various Resolution Levels of Landsat 8 OLI and TIRS Data. Adv Intell Syst Comput 1042:171–184. https://doi.org/10.1007/978-981-32-9949-8_13/COVER
Guha S, Govil H (2020a) Seasonal impact on the relationship between land surface temperature and normalized difference vegetation index in an urban landscape. Geocarto Int 1–21. https://doi.org/10.1080/10106049.2020.1815867
Guha S, Govil H (2020b) Land surface temperature and normalized difference vegetation index relationship: a seasonal study on a tropical city. SN Appl Sci 2(10):1–14. https://doi.org/10.1007/S42452-020-03458-8/FIGURES/7
Guha S, Govil H (2021) An assessment on the relationship between land surface temperature and normalized difference vegetation index. Environ Dev Sustain 23(2):1944–1963. https://doi.org/10.1007/S10668-020-00657-6/TABLES/7
Guha S, Govil H, Diwan P (2020) Monitoring LST-NDVI Relationship Using Premonsoon Landsat Datasets. Advances in Meteorology, 2020, 1–15. https://doi.org/10.1155/2020/4539684
Guhathakurta S, Gober P (2007) The impact of the Phoenix urban heat Island on residential water use. J Am Plann Association 73(3):317–329. https://doi.org/10.1080/01944360708977980
Gupta N, Srivastava S, Jain A, Chaturvedi UC (2012) Dengue in India. Indian J Med Res 136(3):373. /pmc/articles/PMC3510884/
HansIndia. GHMC ropes in entomology wing to clear mosquito menace in Hyderabad. The Hans India. https://www.thehansindia.com/news/cities/hyderabad/ghmc-ropes-in-entomology-wing-to-clear-mosquito-menace-in-hyderabad-766241
Hegazy IR, Kaloop MR (2015) Monitoring urban growth and land use change detection with GIS and remote sensing techniques in Daqahlia governorate Egypt. Int J Sustainable Built Environ 4(1):117–124. https://doi.org/10.1016/J.IJSBE.2015.02.005
Hirano Y, Fujita T (2012) Evaluation of the impact of the urban heat island on residential and commercial energy consumption in Tokyo. Energy 37(1):371–383. https://doi.org/10.1016/j.energy.2011.11.018
Indira D, Srinagesh B (2021) Review on Mitigation Technologies for Controlling Urban Heat Island Effect in Housing and Settlement Areas in Housing and Settlement Areas in Hyderabad. 8.
Kit O, Lüdeke M, Reckien D (2012) Texture-based identification of urban slums in Hyderabad, India using remote sensing data. Appl Geogr 32(2):660–667. https://doi.org/10.1016/j.apgeog.2011.07.016
Laraby KG, Schott JR (2018) Uncertainty estimation method and Landsat 7 global validation for the Landsat surface temperature product. Remote Sens Environ 216:472–481. https://doi.org/10.1016/J.RSE.2018.06.026
Liu L, Zhang Y (2011) Urban heat island analysis using the landsat TM data and ASTER Data: A case study in Hong Kong. Remote Sens 3(7):1535–1552. https://doi.org/10.3390/rs3071535
Lonavath AK, Virugu K (2021) Climate Change Mitigation Measures in Telangana State, India. In: Mishra RK, Singh RB, Dubey A (eds) Sustainable Climate Action and Water Management. Springer, Singapore, pp 191–201. https://doi.org/10.1007/978-981-15-8237-0_16
Malakar NK, Hulley GC, Hook SJ, Laraby K, Cook M, Schott JR (2018) An Operational Land Surface Temperature Product for Landsat Thermal Data: Methodology and Validation. IEEE Trans Geosci Remote Sens 56(10):5717–5735. https://doi.org/10.1109/TGRS.2018.2824828
Masek JG, Vermote EF, Saleous NE, Wolfe R, Hall FG, Huemmrich KF, Gao F, Kutler J, Lim TK (2006) A landsat surface reflectance dataset for North America, 1990–2000. IEEE Geosci Remote Sens Lett 3(1):68–72. https://doi.org/10.1109/LGRS.2005.857030
Meng Q, Liu W, Zhang L, Allam M, Bi Y, Hu X, Gao J, Hu D, Jancsó T (2022) Relationships between Land Surface Temperatures and Neighboring Environment in Highly Urbanized Areas: Seasonal and Scale Effects Analyses of Beijing, China. Remote Sens 2022 14(17):4340. https://doi.org/10.3390/RS14174340. 14
Moulika K (2021), July 10 Dengue mosquitoes breeding in Hyderabad, fear of disease looms | Hyderabad News - Times of India. The Times of India. https://timesofindia.indiatimes.com/city/hyderabad/dengue-mosquitoes-breeding-in-hyderabad-fear-of-disease-looms/articleshow/84285222.cms
Nandi S, Swain S (2022) Analysis of heatwave characteristics under climate change over three highly populated cities of South India: a CMIP6-based assessment. Environ Sci Pollut Res 1:1–13. https://doi.org/10.1007/S11356-022-22398-X/FIGURES/7
News18 (2019), October 2 Children More Vulnerable to Dengue, Say Hyderabad Doctors as Mosquito-borne Diseases Plague Telangana. News18. https://www.news18.com/news/india/children-more-vulnerable-to-dengue-say-hyderabad-doctors-as-mosquito-borne-diseases-plague-telangana-2330927.html
Nikkala S, Peddada JR, Neredimelli R (2022) Correlation analysis of land surface temperature on landsat-8 data of Visakhapatnam Urban Area, Andhra Pradesh, India. Earth Sci Inf 15(3):1963–1975. https://doi.org/10.1007/S12145-022-00850-3/FIGURES/7
Oke TR (1995) The Heat Island of the Urban Boundary Layer: Characteristics, Causes and Effects. In Wind Climate in Cities (pp. 81–107). Springer Netherlands. https://doi.org/10.1007/978-94-017-3686-2_5
Palmate SS, Pandey A, Kumar D, Pandey RP, Mishra SK (2017) Climate change impact on forest cover and vegetation in Betwa Basin, India. Appl Water Sci 7(1):103–114. https://doi.org/10.1007/S13201-014-0222-6/FIGURES/8
Parthasarathy B, Munot AA, Kothawale DR (1994) All-India monthly and seasonal rainfall series: 1871–1993. Theoretical and Applied Climatology 1994 49:4(4):217–224. https://doi.org/10.1007/BF00867461. 49
Pilot E, Murthy GVS, Nittas V (2020) Understanding India’s urban dengue surveillance: A qualitative policy analysis of Hyderabad district. Glob Public Health 15(11):1702–1717. https://doi.org/10.1080/17441692.2020.1767674/SUPPL_FILE/RGPH_A_1767674_SM3899.PDF
Rathi SK, Sodani PR (2021) Summer temperature and all-cause mortality from 2006 to 2015 for Hyderabad, India. Afr Health Sci 21(3):1474–1481. https://doi.org/10.4314/ahs.v21i3.59
Sannigrahi S, Rahmat S, Chakraborti S, Bhatt S, Jha S (2017) Changing dynamics of urban biophysical composition and its impact on urban heat island intensity and thermal characteristics: the case of Hyderabad City, India. Model Earth Syst Environ 3(2):647–667. https://doi.org/10.1007/s40808-017-0324-x
Sara Afrasiabi G, Mahdi P, Fatemeh R (2013) The Relationship between NDVI and LST in the urban area of Mashhad, Iran. International Conference on Civil Engineering Architecture & Urban Sustainable Development. https://www.researchgate.net/publication/265601825_The_Relationship_between_NDVI_and_LST_in_the_urban_area_of_Mashhad_Iran
Schlink U, Herbarth O, Richter M, Dorling S, Nunnari G, Cawley G, Pelikan E (2006) Statistical models to assess the health effects and to forecast ground-level ozone. Environ Model Softw 21(4):547–558. https://doi.org/10.1016/j.envsoft.2004.12.002
Seto KC, Güneralp B, Hutyra LR (2012) Global forecasts of urban expansion to 2030 and direct impacts on biodiversity and carbon pools. Proc Natl Acad Sci USA 109(40):16083–16088. https://doi.org/10.1073/PNAS.1211658109/SUPPL_FILE/PNAS.201211658SI.PDF
Sharma M, Bangotra P, Gautam AS, Gautam S (2022) Sensitivity of normalized difference vegetation index (NDVI) to land surface temperature, soil moisture and precipitation over district Gautam Buddh Nagar, UP, India. Stoch Env Res Risk Assess 36(6):1779–1789. https://doi.org/10.1007/S00477-021-02066-1/FIGURES/5
Sharma VRK, Kaur A (2019) Green Infrastructure Measures to Mitigate Urban Heat Island Effect: a Case of Hyderabad City. Int J Eng Res Technol 8(4). https://doi.org/10.17577/IJERTV8IS040485
Shrivastava S, Saurabh P (2020) Dengue and malaria: A spatial–temporal study across the Greater Hyderabad Municipal Corporation limits. MRIMS J Health Sci 8(4):88. https://doi.org/10.4103/MJHS.MJHS_3_20
Subzar Malik M, Shukla P, J., Mishra S (2019) Relationship of LST, NDBI and NDVI using Landsat-8 data in Kandaihimmat Watershed, Hoshangabad, India. Indian J Geo Mar Sci 48(01):25–31
Sun D, Kafatos M (2007) Note on the NDVI-LST relationship and the use of temperature-related drought indices over North America. Geophys Res Lett 34(24):L24406. https://doi.org/10.1029/2007GL031485
Tan J, Zheng Y, Tang X, Guo C, Li L, Song G, Zhen X, Yuan D, Kalkstein AJ, Li F, Chen H (2010) The urban heat island and its impact on heat waves and human health in Shanghai. Int J Biometeorol 54(1):75–84. https://doi.org/10.1007/s00484-009-0256-x
Taubenbock H, Pengler I, Schwaiger B, Cypra S, Hiete M, Roth A (2007) A multi-scale urban analysis of the Hyderabad metropolitan area using remote sensing and GIS. 2007 Urban Remote Sensing Joint Event, 1–6. https://doi.org/10.1109/URS.2007.371823
TNN. Dengue, malaria spreading in Hyderabad | Hyderabad News - Times of India. The Times of India. https://timesofindia.indiatimes.com/city/hyderabad/dengue-malaria-spreading-in-hyderabad/articleshow/53609267.cms
United Nations (2018) The Sustainable Development Goals Report 2018
USGS (2020) Landsat Collection 2 Surface Temperature | U.S. Geological Survey. https://www.usgs.gov/landsat-missions/landsat-collection-2-surface-temperature
Vermote E, Justice C, Claverie M, Franch B (2016) Preliminary analysis of the performance of the Landsat 8/OLI land surface reflectance product. Remote Sens Environ 185:46–56. https://doi.org/10.1016/J.RSE.2016.04.008
Voogt JA, Oke TR (2003) Thermal remote sensing of urban climates. Remote Sens Environ 86(3):370–384. https://doi.org/10.1016/S0034-4257(03)00079-8
Wang R, Gamon JA, Montgomery RA, Townsend PA, Zygielbaum AI, Bitan K, Tilman D, Cavender-Bares J (2016) Seasonal Variation in the NDVI–Species Richness Relationship in a Prairie Grassland Experiment (Cedar Creek). Remote Sensing 2016, Vol. 8, Page 128, 8(2), 128. https://doi.org/10.3390/RS8020128
Wehner M, Stone D, Krishnan H, AchutaRao K, Castillo F, 16. THE DEADLY COMBINATION OF HEAT AND HUMIDITY IN INDIA AND PAKISTAN IN SUMMER (2016) 2015. Bulletin of the American Meteorological Society, 97(12), S81–S86. https://www.jstor.org/stable/26265443
Weng Q, Liu H, Liang B, Lu D (2008) The spatial variations of urban land surface temperatures: Pertinent factors, zoning effect, and seasonal variability. IEEE J Sel Top Appl Earth Observations Remote Sens 1(2):154–166. https://doi.org/10.1109/JSTARS.2008.917869
Weng Q, Lu D, Schubring J (2004a) Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies. Remote Sens Environ 89(4):467–483. https://doi.org/10.1016/j.rse.2003.11.005
Weng Q, Lu D, Schubring J (2004b) Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies. Remote Sens Environ 89(4):467–483. https://doi.org/10.1016/j.rse.2003.11.005
Zhang X, Zhong T, Feng X, Wang K (2009) Estimation of the relationship between vegetation patches and urban land surface temperature with remote sensing. Int J Remote Sens 30(8):2105–2118. https://doi.org/10.1080/01431160802549252
Zhou W, Huang G, Cadenasso ML (2011) Does spatial configuration matter? Understanding the effects of land cover pattern on land surface temperature in urban landscapes. Landsc Urban Plann 102(1):54–63. https://doi.org/10.1016/j.landurbplan.2011.03.009

No competing interests reported.

Download PDF

Reviews received at journal
10 Nov, 2023
Reviewers agreed at journal
10 Oct, 2023
Reviewers agreed at journal
10 Oct, 2023
Reviewers invited by journal
07 Oct, 2023
Editor assigned by journal
27 Jul, 2023
Submission checks completed at journal
27 Jul, 2023
First submitted to journal
25 Jul, 2023

You are reading this latest preprint version

Seasonal variability of NDVI-LST relationship in Hyderabad, India

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and Methods

2.1 Study area

2.2. Data descriptions and satellite data processing

2.3. Data analysis

3. Results and discussion

3.1. Spatiotemporal and seasonal variation of NDVI distribution

3.2. Spatiotemporal and seasonal variation of LST distribution

3.3. Seasonal variation of LST-NDVI relationship

4. Conclusion

5. Limitations of the study

Declarations

Ethics approval

Competing interests

Author contributions

Funding

Availability of data and material

References

Additional Declarations

Status:

Version 1