*3.1. Datasets*

To study the land cover dynamics and spatial analysis of ICT, various datasets have been collected from primary and secondary sources. The data collected from the primary sources include Landsat medium resolution satellite imagery and field observations using Geographical Positioning System (GPS) receiver, while the secondary or ancillary data consist of population census and ground truth data from Very High Resolution Satellite (VHRS) imagery.

#### 3.1.1. Landsat Satellite and Digital Elevation Model (DEM) Data

For land cover mapping, we used 30 m spatial resolution orthorectified and cloud free Landsat Multispectral Scanner System (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper Plus (ETM+), and Operational Land Image (OLI) sensor images. The TM, ETM+, and OLI sensors have the same 30 m spatial resolution, while the MSS has a spatial resolution of 57 m. The overall ICT region is covered within one Landsat scene (185 × 185 km). A total of five images for years 1976, 1990, 2000, 2010, and 2016 were downloaded from the United States Geological Survey (USGS) Earth Resources Observation and Science (EROS) archive. All satellite data was collected for the months of October and November, to avoid cloud cover (Table 1).

**Table 1.** Landsat images used for land cover mapping.


A 30 m spatial resolution Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) from USGS was used to understand the topography of the area and to prepare the study area maps.

#### 3.1.2. Population Census Data

For the calculation of SDG Goal 11 indicator 11.3.1 LCRPGR, we used the data collected from secondary sources including demographic data (primary census abstracts for the year 1972, 1998, and 2017) from the Pakistan Bureau of Statistics. We used actual census data of 1972 and 1998 to determine the shape of the population curve, which was exponential, as expected. Using the fitted exponential regression function, we derived the growth rate through the exponential function derivative—Equation (1), and then Equation (2) was utilized to determine the approximate population for the year 1976. The population data used for the calculation of LCRPGR is given in Table 2.

$$\text{Groupth rate} = (\text{ab})\text{e}^{\text{ac}} \tag{1}$$

$$\text{Pop}\_{1976} = \text{Pop}\_{1972} + 4 \times \text{Groupth rate} \tag{2}$$

where a and b are regression coe fficients, c represents the population variable, POP1976 = Estimated Population in 1976, and POP1972 = Past Population in 1972.

**Table 2.** Population data of ICT for 1972, 1976, 1998/2000, and 2016/2017.


#### 3.1.3. Ground Truth Data

In this study, we collected ground truth data from two sources: the primary ground data was collected using a Global Positioning System (GPS) receiver in October 2016 during field survey at various locations of ICT, while for the secondary ground data we relied on sub-meter VHRS imagery from Google Earth. A total of 205 samples were gathered, which were utilized as reference data for the satellite image classification and accuracy assessment of results. Using the Google Earth temporal VHRS data, we visually identified and geo-tagged the sites witnessing land degradation, urban development, and tree loss. The geo-tagged dataset helped to validate Landsat-based land cover changes.
