Urban Heat Island (UHI) refers to the phenomenon in which metropolitan regions experience significantly higher temperatures than their surrounding rural areas. This effect is primarily driven by human activities that alter the natural properties of the land surface and atmosphere. As cities grow and expand, they replace natural land cover with impermeable surfaces such as asphalt, concrete, and buildings. These materials absorb and retain heat, leading to elevated temperatures. The significant modifications in land use and land cover contribute to the formation of the UHI effect [1]. Assessing land cover changes provides valuable insights into the impact of human activities on the ecological conditions of urban environments [2].

Urbanization and rapid population growth are among the leading causes of the degradation and loss of natural vegetation and arable land [3]. Developing countries, in particular, exhibit pronounced land use and land cover transformations resulting from extensive human activities [4]. Researchers worldwide have observed considerable urban expansion associated with thermal variations and ecological disturbances in rapidly growing cities. Urbanization influences the microclimate by altering local land surface temperatures, increasing energy consumption, and contributing to greenhouse gas emissions [5, 6].

Several studies have highlighted the growing importance of understanding urban heat islands and their impacts on cities. Researchers have found that urban expansion has significantly affected major cities such as New Delhi, Mumbai, Chennai, Tokyo, Los Angeles, and Cairo. As urbanization continues to expand, more cities are likely to experience similar challenges, emphasizing the urgent need for effective mitigation strategies to address UHI and promote sustainable urban development. Many Mediterranean cities, for instance, share common characteristics of densely built settlements with limited green spaces and extensive use of impervious materials, exacerbating heat stress [7, 8]. Climate change further amplifies these challenges, making it essential to implement adaptation strategies.

In densely populated metropolitan areas with minimal vegetation, the increasing frequency of heat waves has drawn attention to the correlation between temperature rise and land use patterns. The loss of natural spaces, combined with urban layouts that disrupt wind circulation, contributes to the intensification of the UHI effect [9]. Poor urban planning is a key factor exacerbating this issue [10].

Latakia, Syria, serves as a prime example of a city struggling with unplanned urban expansion, especially following waves of population displacement caused by the Syrian crisis. As one of the country's safest cities, Latakia has experienced significant population pressure, resulting in urban sprawl at the expense of green spaces. This study assesses the impact of increasing population density and subsequent urban expansion in Latakia, particularly in relation to climate change, to propose effective mitigation and adaptation strategies. 

The study area is located in the northwestern part of Syria [Fig. 1], within the administrative boundaries of Latakia governorate. This study covered 16 sectors. The highest population density is in the western sectors of the city, which are associated with both commercial and residential activities. The majority of the city’s marketplaces and logistical infrastructure are located in these highly populated sectors. Eastern and northern sectors have business, residential, and agricultural activities [11]. Low-density informal housing was created when the rich agricultural fields were progressively urbanized.

In these images, the study area remained cloud-free, as the cloud cover was concentrated in a different part of the scene.

This study was conducted in August 2013 and August 2023, representing the hottest summer months when the UHI effect is most intense in Syria. 2013 was chosen as it marks the onset of large-scale displacement to Latakia due to the Syrian crisis. A ten-year comparison assesses how this population shift impacted UHI. Landsat 8 satellite images were used to derive Land Surface Temperature (LST) and Normalized Difference Vegetation Index (NDVI), key indicators for analyzing UHI effects and human impacts on the environment [12].

The heat risk index was calculated using 3 criteria in the ArcGIS Pro software, where the weighted sum analysis techniques are used to determine the heat risk over the study area. The population density, LST values, and non-vegetation coverage are used for heat risk identification. These criteria were used together to show the impact of population distribution dynamics on the change in urban areas and its impact on the shrinkage of green spaces in the city and the resulting increase in the effect of urban heat islands. 

a) LST was calculated using the following steps:

Land Surface Temperature (LST) data plays a vital role in analyzing the impacts of climate change on both urban and rural regions. It offers a direct measure of surface temperature, enabling researchers to identify areas with notable thermal fluctuations. These variations are often influenced by factors such as population density and human activities, which significantly contribute to the UHI effect and broader surface temperature changes across different landscapes [13, 14]. To calculate the LST, the algorithm applied by [15] using ERDAS IMAGINE was adopted as follows:

1. The transformation of the digital number (DN) to spectral radiance (L) (Conversion to Top of Atmosphere (TOA) Radiance):

L(λ) = ML x Band 10 + AL – Oi

Where:

2. Conversion to Top-of-Atmosphere (TOA) Brightness Temperature, Kelvin (k) to Celsius degree Co:

BT = K2 / ln (k1/L(λ)+1) – 273.15

Where:

3. Normalized Difference Vegetation Index (NDVI):

NDVI = (NIR-RED) / (NIR+RED)

4. Minimum and maximum NDVI values are used for the proportion of vegetation calculation:

PV = ((NDVI – NDVI min) / (NDVImax – NDVImin))2

5. The land surface emissivity (LSE) is calculated based on the proportion of vegetation:

E = 0.004 * PV + 0.986

6. Land Surface Temperature (LST) Calculation:

LST = BT / (1+ (λ * BT / C2) * ln (E))

Where:

We choose the highest temperature in each sector within the study area after driving the LST index. Consequently, the HRI first input is done.

b) Derivation of the NDVI:

The Normalized Difference Vegetation Index (NDVI) is a widely used indicator for monitoring vegetation conditions and distinguishing green vegetation from bare soil [16]. It helps assess vegetation quantity and detect changes over time [17]. Spectral vegetation indices are globally applied to evaluate vegetation health in both natural and cultivated landscapes [18-22].

After generating the NDVI map, the region was classified into vegetation and non-vegetation areas. The number of pixels in each class was calculated, and their percentage was determined. The non-vegetation percentage serves as a second input for computing the HRI.

c) Calculation of population density:

Each of the 16 sectors population densities was determined by dividing the total number of inhabitants by the neighborhood's area. Thus, the third input for RHI is done.

Because a higher population density will result in a greater demand for structures, which is linked to a reduction in the amount of green space in cities, it will be easier to show how population density exacerbates the urban heat island phenomenon, so Heat Risk Index.

d) Calculation of Heat Risk Index (HRI):

Once the input variables have been processed, they can be integrated to generate HRI [Fig. 2]. The three derived input variables for HRI should be combined into a single shapefile layer. This layer will encompass all the studied sectors, ensuring that each sector is associated with its corresponding data from the three input variables within the attribute table.

The input variables use different units of measurement and must be standardized before being combined. The Standardize Field geoprocessing tool was applied to standardize the units for the three input variables using ArcGIS Pro. The minimum-maximum standardization method is used to standardize the input variables and create a 5-point scale where 1 represents areas with the lowest heat risk index and 5 represents areas with the highest heat risk index, as follows: No Risk, Mild, Moderate, Severe, and Extreme Risk.

The three-input criteria data were converted into raster format classified from 1 to 5, as the weighted sum tool requires raster inputs. The number 1 indicates a low value in the impact on the HRI, and the higher the number, the higher the impact, until we reach the number 5, which is the most influential class on the HRI.

The first input generated is LST. As observed in the maps [Fig. 3], a comparison between 2013 and 2023 reveals a slight temperature increase over the decade. This rise is attributed to recurrent drought events and a general decline in rainfall during this period [23-25]. 

In 2023, there is a noticeable reduction in low-temperature areas compared to 2013, as these regions have shifted to higher temperatures. This trend is particularly evident in the Al-Datour and Demsarkho sectors, where urban expansion and green space reduction, driven by population pressure and internal displacement, have led to rising temperatures. Similar patterns were observed in Tartous, where built-up areas nearly doubled during the crisis [26].

After classifying temperatures into five classes, Basnada, Al-Pharous, and Al-Raml Aljanobi followed class 5 (high temperatures) in 2013. By 2023, Demsarkho, Al-Datour, and Tishrine had also reached class 5, while Al-Pharous shifted to class 4 due to declining temperatures [Fig. 4].

Only a few sectors, like Mar Takla, Al Owaynah and Tabiyat, kept the same temperature class over these years and remained associated with temperature class 1, whereas the majority of sectors had temperature increases during these ten years. Land cover type, especially vegetation cover, plays a crucial role in shaping the microclimate by influencing the exchange of energy between the land and atmosphere [27] which in turn alters temperature and precipitation patterns. Urbanization and shrinking green spaces in cities are key processes that can significantly increase exacerbating global warming [5].

4.2 Non-Vegetation Cover Effects

Vegetation-covered areas play a crucial role in moderating the climate in cities as they reduce warmth and increase shade. Therefore, locating vegetation-free zones within the city is a key step toward limiting the sites that contribute to increasing the severity of the HRI.

The NDVI maps for the study area were generated for the years 2013 and 2023. As observed in [Fig. 5], the majority of sectors maintained a relatively stable percentage of vegetation cover, with minimal changes in either increase or decrease. However, a noticeable decline in green areas is evident in certain northern sectors, particularly in Demsarkho and Al-Datour. 

This reduction can be attributed to urban expansion, driven by the increased demand for housing due to internal displacement during the Syrian crisis. Notably, these sectors previously consisted of more than 50% agricultural land, making the loss of vegetation particularly significant. The impact extends beyond the intensification of the urban heat island effect, leading to adverse environmental consequences for agricultural systems and economic repercussions. Furthermore, as indicated in [Table. 2], the proportion of non-vegetated land increased by approximately 4% in 2023 compared to 2013.

By classifying the NDVI map into two main categories vegetation cover and non-vegetation cover a new map was derived based on the percentage of non-vegetated areas [Fig. 6]. This map was further classified into five categories according to the proportion of vegetation cover. Class 1 represents the areas with the highest percentage of vegetation cover, while Class 5 corresponds to those with the lowest vegetation cover. Consequently, Class 5 has the most significant influence on increasing the severity of the heat risk index due to the limited presence of green spaces.

As illustrated in [Fig. 6], most sectors have experienced a decline in vegetation cover. Notably, the city center, particularly the sectors of Mar Takla, Alpharos, and Qannis, has shifted to Class 5, indicating the lowest percentage of vegetation cover. The primary factor contributing to the reduction in green spaces in these central sectors is high population density. Additionally, these sectors have a smaller spatial extent compared to others. In the case of AL-Raml Aljanobi sector, vegetation cover has also declined, placing it in Class 4. This sector has undergone significant unplanned and irregular urban expansion, largely due to the high influx of displaced persons.

4.3 Population Density Effects

In our case study of Latakia city, demographic changes due to internal displacement during the Syrian crisis have been the most significant factor influencing the heat risk index. A comparison of population density between 2013 and 2023 [Fig. 7] indicates that the overall population increase across all sectors was substantial; its impact on urban heat island intensity is evident. The increasing population has been observed to drive substantial urban expansion, encroaching upon forested and agricultural lands [28-31].

This is particularly due to the city's reception of a disproportionately high number of displaced persons relative to its area, whereas as of February 2022, the number of IDPs in Latakia governorate was stated to be 449,317. According to UNOCHA, as the administrative center of the governorate, Latakia absorbed the largest share of displaced individuals, resulting in significant population pressure across the city. 

Notably, [Fig. 7] shows that the Al Owaynah and Tabiyat sectors, both located in the city center, experienced the most significant population growth in 2023, nearly doubling compared to 2013. In analyzing the population factor, population density was calculated and classified into five categories, similar to the previous factors. The maps [Fig. 8] illustrate the spatial distribution of these classes across the studied sectors. The highest population density was recorded in Al Owaynah, the smallest sector, which transitioned from Class 3 in 2013 to Class 5, the highest density category, in 2023. Additionally, five other sectors Mar Takla, Al Sleibeh, Tabiyat, 7 Nisan, and Al-Datour experienced an increase in population density classification over this period.

Notably, the AL-Raml Aljanobi sector remained in Class 3 in both 2013 and 2023. However, despite this classification, it had the highest population among all sectors and is one of the largest in the study area. As a result, its population density did not appear high relative to its size. Nevertheless, the sector experienced significant population pressure due to the limited number of residential buildings and the presence of barren lands.

4.4 Heat Risk Index (HRI) Distribution 

Using the three previously analyzed factors, a map of HRI classes was generated, [Fig. 9]. In 2013, none of the sectors fell into the extreme risk class. However, four sectors Damsarkho, Al-Datour, Mar Takla, and Mashrou Al Ziraa were classified under the Mild class. These sectors were characterized by extensive green spaces, except for Mar Takla, where vegetation cover played a crucial role in mitigating the local climate conditions.

Furthermore, in 2013, none of the sectors fell into the No Risk category, indicating that all sectors were already at a critical stage. Without intervention measures to enhance green spaces and implement climate mitigation strategies, the risk of urban heat island impacts could escalate.

By 2023, most sectors saw an increase in HRI levels, with Al Sleibeh and Al-Raml Aljanobi reaching the extreme risk category due to significant population pressure. Al Sleibeh experienced 24% population growth, while Al-Raml Aljanobi’s population surged by 64%, alongside a rise in buildings and a decline in green spaces [Fig. 10]. This contradicts modern urban planning research, which emphasizes large-scale tree planting as the most effective solution for mitigating urban heat stress, improving air quality, and enhancing overall environmental sustainability [10].

The Al-Datour sector was the only area where the HRI increased by two levels, reaching a severely dangerous category in 2023. This escalation is mainly due to an 8% increase in urban expansion, particularly over agricultural lands. As shown in [Fig. 11], extensive agricultural fields and fruit orchards have been replaced with buildings, exacerbating heat retention. 

The loss of vegetation, essential for temperature regulation, has intensified the urban heat island effect.

This sector highlights the urgent need for sustainable urban planning that balances development with green space preservation. Proposed mitigation strategies include tree-planting programs, high-albedo materials, and climate-responsive land-use planning [32-34].

The impact of effective urban planning is particularly evident in the Southern Corniche sector, which has maintained a moderate HRI classification over the past decade [Fig. 12]. This stability can be attributed to its development in accordance with urban planning standards that prioritize adequate green spaces, as well as the absence of significant building expansion during this period. Additionally, population growth in this sector was minimal, nearly negligible. It is also a sector overlooking the sea, as the proximity to the sea helps mitigate heat accumulation from concrete structures, reducing the overall urban heat effect.

The findings revealed that in 2013, none of the sectors experienced extreme heat risk; instead, risk levels ranged from mild to severe. However, by 2023, heat risk levels increased in most sectors, with Al Sleibeh and Al-Raml Aljanobi reaching extreme heat risk due to significant population growth. Meanwhile, Southern Corniche and Mashrou Alqalaa maintained a moderate heat risk level. Notably, Al-Datour was the only sector where the HRI increased by two levels, reaching a severely dangerous category. This surge was driven by rapid urban expansion, particularly over agricultural lands.

The analysis shows that most HRI changes in urban areas are linked to rising population density or urban expansion, often at the expense of green spaces. Temperature played an indirect role, whereas direct impacts were due to internal displacement, increasing population pressure. This highlights the need for re-planning to accommodate population growth, encouraging suburban development and expanding green spaces to counterbalance the city's current deficiency.

To combat urban heat island effects, the city must adopt innovative adaptation strategies. The HRI can serve as a foundational tool to identify high-risk areas, guiding localized adaptation plans that prioritize climate resilience, sustainable urban planning, and increased green coverage.

The first author expresses gratitude to the Indian Council for Cultural Relations (ICCR) for providing all the available support.