Evaluate the Effect of Relative Humidity in the Atmosphere of Baghdad City urban expansion Using Remote Sensing Data

The effects of land use/cover change are recognized as one of the challenges facing humans in the twenty-first century. In contrast to less developed regions, its characteristics are characterized by a variety of climatic conditions. The changes result in “Urban Heat Island”, in which the temperature in cities is higher than the temperature in the rest of the country. By monitoring the city of Baghdad for a short period, Baghdad's urbanization progressed quickly, which negatively affected the region’s climate through the decreasing of agricultural lands surrounding the area. Understanding the extent of effects on the environment is critical for long-term development. Climate change and environmental cleanup include making the right decisions and taking actions to reduce the detrimental impact of urbanization on both the urban population and ecosystems. As a result, we investigated the relationship between urbanization and relative humidity in Baghdad city, using remote sensing images and data downloaded from the European Center for Forecasting (ECMWF) for Baghdad city. By calculating several factors, such as relative humidity (RH), temperature (Ta), and evaporation, and confirming the changes observed in urban areas Landsat 5 and Landsat 8 images were processed and analyzed in the (ArcGIS 10.8) program for the years 2010 to 2020. The study proves that there is a clear association between growing urbanization and the relative humidity rate and how increase in building areas and the rate of increase in temperature rise in them. The purpose of this study is to assess the impact of relative humidity level on the microclimate of Baghdad city from 2010 to 2020, the built-up increased from 19.60% to 27.44%. While, NDVI calculation healthy vegetation has almost disappeared with its percentage going down from 0.05% to 0.00. This study compares the microclimates of two study areas: ancient and modern, both of which differ in terms of vegetation cover and urban distribution and is based on remote sensing data.


Introduction
Green spaces in cities can play an important role in coping with climate change and can be used for both adaptation and mitigation. Their impact on improving thermal conditions and ventilation, as well as adjusting relative humidity within the region, promises to aid in regulating the microclimate conditions that are deteriorating due to climate change [1]. However, According to studies on the impact of urban space configuration on urban climate, vegetation alone is not representative of micro climate conditions [2]. Urban microclimates are known to be influenced by factors such as street widths, building heights, street orientations, and built-up densities [3]. Urban Heat Island (UHI) effect is one of the effects of rapid urbanization and urban growth. Because of its implications for environment and its associated development, the urban heat island is getting more and more widespread. It happens when the temperature in an urban city is higher than its surrounding area. The anthropocentric characteristics and behaviors of urban areas cause urban heat island effect. During the process of urbanization and spatial transformation, natural materials are converted into impervious surfaces, houses, and a road network variations in building surfaces, land use, and morphology in urban areas affect local temperatures, keeping urban areas warmer than non-urbanized areas [4,5]. The microclimatic impact of urbanization is analyzed in terms of the following variables: relative humidity (RH), temperature of the air (Ta), evaporation, and the wind speed (Va) [3]. The most important components of natural environment that have a significant impact climate is vegetation, water, and soil. Therefore, the study of climatic conditions is important because it represents the full picture of the nature prevailing in the area to be studied [6]. Iraq's climate has continental characteristics, as it is warm and dry in the summer and cold in the winter. In this research, the ECMWF (European Center for Medium-Range Weather Forecasts) used climate data that included temperature, evaporation, total precipitation, and relative humidity [7]. In determining and evaluating land use and land cover shifts, remote sensing and geographic information systems have proven to be extremely useful. Satellite-based remote sensing has revolutionized the study of land use and land cover change due to its ability to provide synoptic details of land use and land cover at a specific time and place [8]. Land use and land cover data over time can be used to recognize areas of change in a country. Changes in land use and land cover can be designated spatial connotations using geo informatics, such as population pressure, environment, landscape, and other factors that drive these changes [9]. The various phases of remote sensing data can be studied for quantitative analysis and defining the characteristics and processes of surface changes. It includes ground surface forms, boundary changes, and patterns before and after the changes, as well as the form, distribution, and quantity of changes. Because of its high temporal frequency, digital format suitable for computation, synoptic vision, and wider range of spatial and spectral resolutions, remote sensing data has become a major source for change detection studies [10].

2.The Study area
The study area is situated between longitudes (33.452°N and 33.184°N) and longitudes (44.189°E to 44.576°E) in Baghdad district, Iraq's capital. Baghdad is 25 miles (40 kilometers) west of the Euphrates River, where the Tigris meets the Euphrates. The Tigris River, which runs through the city, divides it into two parts: Karkh (Western part) and Rusafa (Eastern part) Figure 1 [11].

Data Presentation
This study based on Landsat 5 TM Thematic Mapper and Landsat 8 OLI Operational Land Imager satellite image, obtained satellite imagery from the official website of the United States Geological Survey (USGS). These images were used to achieve the study's objectives for three period (2010, 2015, and 2020). The projection of the images is WGS84-UTM_38N, with a spatial resolution of 30m. Three different time periods were used to capture the images: ArcGIS10.8, processed satellite images, and generated maps of land cover and land use (LCLU) for the study area. The maps are generally consistent in terms of size, units, and methods.
The relative humidity, temperature of the air, evaporation, and the wind speed factor was then calculated using meteorological data obtained from the European Center for Medium-Range Weather Forecasts (ECMWF) (https://www.ecmwf.int/).

Image pre-processing
Satellite images needs to be pre-processed so that a variety of studies on land use change identification could be carried out using various analysis methods. The surface temperature and its changing patterns areas are monitored and detected using GIS and remote sensing. The main goal of this study is to determine the temporal changes in LULC and to observe the urban land surface temperature and perception period to see how much it varies, Using ArcGIS version 10.8, the image data was corrected, which resulted in the development of information to extract land use/land cover information from remote sensor data. The study area's final image is produced and used in the image classification process [12].

Unsupervised Classification
Unsupervised classification is a process that checks a large number of unknown pixels and divides them several categories dependent on natural groupings existing in Landsat image value [13]. The unsupervised classification method does not require analyst-identified training samples. The key rule is that data from different groups should be kept separate in the detection zone within a defined cover type. The unsupervised classification method yielded spectral groups, which are based on natural grouping of Landsat image values. The existence of a spectral group would be unknown beforehand should compare classified data from real data (such as maps, or location visit) to know the identity and information values of the specified spectral categories [14].

Land use/land cover change, 2010-2020
Changes in land use and land cover have been among the most visible changes in our environment. Although perceptible, the intensity, variety, and spatial variability of the changes occurring have made quantifying and assessing land use and land cover changes a difficult task for scientists. Remote Sensing and GIS has proved to be very important in assessing and analyzing (LCLU) changes. Because of its ability to provide synoptic details of land use and land cover for a specific time and place, space-based remote sensing has revolutionized the study of land use and land cover [15]. The temporal data on (LCLU) aids in the identification of areas of change in a field. According to the use of geoinformatics, land use and land cover changes are given spatial connotations, such as population growth, environment, landscape, and other factors that influence these changes [16].

Normalized Difference Vegetation Index (NDVI) calculation of Satellite Images of Baghdad
The NDVI (Normalized Differential Vegetation Index) is a standardized vegetation index that can be used to create a picture of relative biomass. The NDVI is calculated using chlorophyll absorption in the red band and relatively high vegetation reluctance in the near infrared band (NIR) [17]. The NDVI is widely used around the world to monitor droughts, as well as to control and forecast agricultural production. Since it compensates for changing illumination conditions, surface slope, aspect, and other extraneous variables, the NDVI is favored for global vegetation monitoring [18]. In the near-infrared band, green vegetation reflects more energy than in the visible spectrum. As a result of its development, the NDVI method produces a single-band dataset that only shows greenery. Rock and bare soil are represented by values close to zero, while water, snow, and clouds are represented by negative values.The vegetation growth signal is distinguished from the background signal by taking the ratio or difference of two bands. The NDVI method was created by NASA scientists (equation 1), by taking a ratio of two bands, and droping the values between -1 to +1 [19]. Water has an NDVI value less than zero, bare soils between zero and 0.1, and and vegetation over 0.1. Increase in the positive NDVI value means greener vegetation, and the below equation is used to calculate (NDVI), where NIR-reflection in the near-infrared spectrum; RED-reflected in the red range spectrum [20] :

Normalized difference built-up index (NDBI) (NDBI) is effective for mapping urban Land and uses Landsat -5 TM information and also
Landsat-8 OLI information [11]. The NDBI index for Landsat 5,8 uses the difference and ratio of the blue band and the near infrared band (NIR) to indicate built-up areas, and it is given by the equation (2): … ….
(2) Where BLUE = blue band; NIR = near infrared band. This index takes advantage of its simplicity as well as the speed at which it can be computed [21]. The unsupervised classification of the built-up maps was achieved using the program ArcGIS Version 10.8 for satellite images, for the period (2010,2015,2020).
Iraqi Journal of Science, 2022, Vol. 63, No. 4, pp: 1848-18591854 The total built-up area in 2010 was 1020.1113 km2, which includes all residential, commercial, and industrial properties. It rose to 1241.559 km2 in 2015, and it raised to 1427.9967 km2 in 2020, indicating an increase in urban sprawl as a result of population growth over time. Figures (8) and (9) clearly show the rate of change in built-up area from 2010 to 2020, indicating the urban sprawl pattern caused by urbanization [12]. The analyzing of satellite images shown in Figure (6 ,7,8,9) and Tables (3,4) illustrate Baghdad's rapid growth in built-up areas from 2010 to 2020 years, by NDBI index we managed to isolate the built-up category from a different category.

The Relative Humidity
One of the most important parameters in describing the physical state of the atmosphere is relative humidity. It regulates cloud formation as well as the optical properties of aerosols, affecting visibility [22]. Relative humidity is a measurement of how much moisture is present in the air compared to how much moisture the air can retain at that temperature. In weather reports, the relative humidity is expressed as a percentage. Figures (10,11) illustrate the average relative humidity for April and August for the years of the study [23].  . This expansion in the construction areas during the last ten years happened by exploiting the agricultural lands. Therefore, we notice, through images analysis from satellites by (NDVI), how the dense vegetation in the city of Baghdad (i.e. the large orchards) decreased and the lack of healthy trees in recent years, where the value of (NDVI) went from -0.982609 in 2010 to 0.992188. While in 2015, the value has dropped from -0.322738 to 0.626365, and in 2020 the value of (NDVI) it changed from -0.303526 to 0.587578. This means the healthy vegetation decreased dramatically. The maps show clearly the increase in unhealthy vegetation, which values ranges from (0.2-0.4) from 2015 to 2020, and the reason is attributed to the increase in the percentage of rain in recent years that led to the emergence of weeds, which were classified within this value. As for the average relative humidity, it increased in the year 2020 during April and August despite the increase in building areas due to the abundance of rain in this year and the factor that is considered as a basic climatic factor in increasing the relative humidity in the atmosphere layer.

Conclusions
The aim of this research was to analyze and classify satellite images in different seasons of the year to show the difference between the climatic factors of two different seasons of the year, which are the months of April and August, and how the scarcity of plants affected by the expansion of construction. It concludes with the following results: 1-The change in vegetation cover from 2010 to 2020 years by using agricultural land in 2020for construction. there is almost no vegetation cover from the healthy plant category and the increase in vegetation cover from the unhealthy plant type, which represents the weeds that have increased as a result of the increase in the rate of rain in recent years. 2-This study showed the spatial change occurring from 2010 to 2020 years for the following items: the built-up from 19.60% to 27.44%, unhealthy vegetation from 11.10% to 31.76%, moderate vegetation 0.74% to 3.89%, healthy vegetation from 0.05% to 0.00%. 3-The data indicate an increase in the average relative humidity in the last year of study, the temperature also decreased. The reason for the first is the large amount of rain during 2020 in addition to the large decrease in global warming that occurred because of the general ban for all inhabitants of the earth during the year 2020 due to the global pandemic of COVID-19 Disease. 4-It was found that the use of the spectral indexes method is better to determine the changes in the land cover/land use (LCLU) than the classifications (supervised, unsupervised).