Please use this identifier to cite or link to this item:
Title: Spatial Resource Assessment of Forests in Azad Jammu and Kashmir, Pakistan using Optical Remote Sensing Techniques
Authors: Ahmad, Adeel
Keywords: Physical Sciences
Issue Date: 2022
Publisher: University of the Punjab, Lahore
Abstract: Forest resources, including forest cover, forest aboveground biomass (FAGB), and forest species diversity, are an integral part of our environment that provides substantial organic infrastructure for the habitats of many life forms and livelihood for human beings. The Forest Resource Assessment (FRA) report by the Food and Agriculture Organization (FAO) of United Nations includes forest cover, forest characteristics, ecosystem services, biodiversity, and forest disturbances in its mainly focused categories related to the world's forests. Spatial assessment of these forest resources provides vital information for forest conservation and management practices. The review of the spatial forest assessment studies in Pakistan shows that out of seventy-three studies, three were conducted in Azad Jammu and Kashmir (AJK), one in Balochistan, three in Gilgit-Baltistan (GB), twelve in Islamabad Capital Territory (ICT), thirty-one in Khyber Pakhtunkhwa (KP), six in Punjab, ten in Sindh, and remaining seven studies were conducted on national/ regional scales. While a global literature review of FAGB resources using optical high-resolution satellite data shows that twenty-one studies were conducted across six continents in Asia, eight in North America and Africa, five in South America, and four in Europe. This study aims to report these forest resources in the AJK region, which is the administrative area of Pakistan with forests lying in the Western Himalayas using optical remote sensing data and techniques. Among these forest resources, firstly, this study reports the forest cover change and evolution of spatial configuration of the forest using Landsat satellite imagery over the past three decades (1990-2020) in the region of AJK, Pakistan. The results revealed that in 2000-2010 the forest cover of AJK declined at the highest rate of change of -3.00 km2 (-0.071 %) per year, and in 1990-2020 the overall rate of ii change of -2.50 km2 (-0.059 %) per year was calculated. The forest fragmentation analysis shows that the core areas (>500 acres) of forests are being fragmented into small (<250 acres) or medium (250-500 acres) core areas. Landscape metrics obtained at the class level show that except few areas, the overall forests of AJK present a connected and aggregated pattern temporally. The pattern-based spatial and dissimilarity analysis, based on forest cover and ecoregions in AJK revealed forest hotspots (gains or losses) and forest clusters. Secondly, this study aims to spatially predict the potential distribution of four native tree species (Abies pindrow, Olea ferruginea, Pinus roxburghii, and Pinus wallichiana) using 21 bio-climatic, two biophysical, and three topographical variables in MaxEnt modelling. 739 species occurrence points from 45 circular plots of selected species were used with selected variables, filtered through multi-collinearity tests. The jackknife test showed different essential variables influencing the prediction of species distribution, including elevation, vegetation indices, temperature, and precipitation. Satisfactory results were achieved with Area under ROC (receiver operating characteristic) Curve (AUC) testing and training values greater than 0.74 and 0.88, respectively. Based on the 10-percentile training presence threshold-dependent values, the True Skill Statistics (TSS) test attained at least 76% overall accuracy for species distribution. Abies pindrow covered 429.58 km2, Pinus wallichiana 346.28 km2, Pinus roxburghii 341.93 km2, and Olea ferruginea 27.53 km2 within the very highly suitable category of predicted potential distribution. The results also showed that the Himalayan subtropical pine forests ecoregion has the highest species diversity. Thirdly, the AGB in the study area was estimated using the Random Forest (RF) modelling approach. A mean AGB of 74.9 tons per hectare (t ha-1) for the entire region of AJK with a coefficient of determination (R2) of 0.79. The northern and northeastern districts of AJK, including Neelum, Muzaffarabad, and Hattian Bala showed the highest mean AGB estimations compared to the southern districts of the region. The overall analysis for the assessment of forest resources of AJK shows that the region has high biodiversity and has an overall connected forest pattern temporally. But the declining forest trend, even though not an abrupt decline, between 1990 and 2020 should be a matter of concern for the authorities and forest conservation groups. This study can be taken as a prototype for other administrative areas of Pakistan, especially forest-rich KP province, GB region, and mangrove forest resources of Pakistan.
Gov't Doc #: 27045
Appears in Collections:PhD Thesis of All Public / Private Sector Universities / DAIs.

Files in This Item:
File Description SizeFormat 
Adeel Ahmad geomatics 2022 uop lhr.pdf 29.9.22.pdfPh.D thesis7.16 MBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.