Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan
Dry land ecosystems extend over 40 % of the Earth, supporting an estimated 3 billion human population. Thus, quantifying LCLU changes in such ecosystems is essential for achieving sustainable development goals. In this context, this research aimed to examine the LCLU changes in the past three decade...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | Ecological Indicators |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470160X24001274 |
| _version_ | 1797295794623086592 |
|---|---|
| author | Zulqadar Faheem Jamil Hasan Kazmi Saima Shaikh Sana Arshad Noreena Safwan Mohammed |
| author_facet | Zulqadar Faheem Jamil Hasan Kazmi Saima Shaikh Sana Arshad Noreena Safwan Mohammed |
| author_sort | Zulqadar Faheem |
| collection | DOAJ |
| description | Dry land ecosystems extend over 40 % of the Earth, supporting an estimated 3 billion human population. Thus, quantifying LCLU changes in such ecosystems is essential for achieving sustainable development goals. In this context, this research aimed to examine the LCLU changes in the past three decades (1990 – 2020) in an arid ecosystem of Pakistan, i.e., the Cholisatn desert. Three remote sensing indices, the normalized difference vegetation index (NDVI), normalized difference barren index (NDBaI), and top grain soil index (TGSI) are taken as LCLU representatives to examine their temporal relationship associated with meteorological drought, e.g. the standardized precipitation index (SPI). Moreover, machine learning-based random forest (RF) classification followed by change detection techniques was implemented. Results from RF classifier revealed the applicability of RF in accurately predicting LULC with validation overall accuracy of 0.99. Output of the research revealed an interesting finding where the desert experienced significant LCLU change over the last three decades. The highest vegetation expansion (4.4 %) took place from 2014 to 2020 at the expense of the highest reduction of barren land (-6.3 %). Mann-Kendall trend (MK) and Sen’s slope (SS) analysis showed a significant (P < 0.001) increasing trend of NDVI (SS = 0.004), SPI (SS = 0.01 and 0.04) and decreasing trend of NDBaI and TGSI (SS = -0.001, −0.005). Interestingly, the significant positive Pearson correlation range (r = 0.6–0.8) of NDVI with SPI-1 to 6, and negative correlation range (r = 0.5–0.7) of NDBaI with SPI indices reveals a strong linear relationship between LCLU and meteorological drought. The research provides substantial implications for policy makers and stakeholders emphasizing the need for proactive strategies such as drought resistant vegetation to improve and maintain the ecological health of desert and combating the negative impacts of climatic change. |
| first_indexed | 2024-03-07T21:53:06Z |
| format | Article |
| id | doaj.art-38ff1018826f41aba6dabc320f664ccb |
| institution | Directory Open Access Journal |
| issn | 1470-160X |
| language | English |
| last_indexed | 2024-03-07T21:53:06Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ecological Indicators |
| spelling | doaj.art-38ff1018826f41aba6dabc320f664ccb2024-02-25T04:35:06ZengElsevierEcological Indicators1470-160X2024-02-01159111670Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of PakistanZulqadar Faheem0Jamil Hasan Kazmi1Saima Shaikh2Sana Arshad3 Noreena4Safwan Mohammed5Department of Geography, University of Karachi, Karachi 75270, PakistanDepartment of Geography, University of Karachi, Karachi 75270, PakistanDepartment of Geography, University of Karachi, Karachi 75270, PakistanDepartment of Geography, the Islamia University of Bahawalpur, Bahawalpur 63100, PakistanDepartment of Geography, the Islamia University of Bahawalpur, Bahawalpur 63100, PakistanInstitute of Land Use, Technical and Precision Technology, Faculty of Agricultural and Food Sciences and Environmental Management, University of Debrecen, Debrecen, 4032 Hungary; Institutes for Agricultural Research and Educational Farm, University of Debrecen, Böszörményi 138, 4032 Debrecen, Hungary; Corresponding author.Dry land ecosystems extend over 40 % of the Earth, supporting an estimated 3 billion human population. Thus, quantifying LCLU changes in such ecosystems is essential for achieving sustainable development goals. In this context, this research aimed to examine the LCLU changes in the past three decades (1990 – 2020) in an arid ecosystem of Pakistan, i.e., the Cholisatn desert. Three remote sensing indices, the normalized difference vegetation index (NDVI), normalized difference barren index (NDBaI), and top grain soil index (TGSI) are taken as LCLU representatives to examine their temporal relationship associated with meteorological drought, e.g. the standardized precipitation index (SPI). Moreover, machine learning-based random forest (RF) classification followed by change detection techniques was implemented. Results from RF classifier revealed the applicability of RF in accurately predicting LULC with validation overall accuracy of 0.99. Output of the research revealed an interesting finding where the desert experienced significant LCLU change over the last three decades. The highest vegetation expansion (4.4 %) took place from 2014 to 2020 at the expense of the highest reduction of barren land (-6.3 %). Mann-Kendall trend (MK) and Sen’s slope (SS) analysis showed a significant (P < 0.001) increasing trend of NDVI (SS = 0.004), SPI (SS = 0.01 and 0.04) and decreasing trend of NDBaI and TGSI (SS = -0.001, −0.005). Interestingly, the significant positive Pearson correlation range (r = 0.6–0.8) of NDVI with SPI-1 to 6, and negative correlation range (r = 0.5–0.7) of NDBaI with SPI indices reveals a strong linear relationship between LCLU and meteorological drought. The research provides substantial implications for policy makers and stakeholders emphasizing the need for proactive strategies such as drought resistant vegetation to improve and maintain the ecological health of desert and combating the negative impacts of climatic change.http://www.sciencedirect.com/science/article/pii/S1470160X24001274Change DetectionClimate ChangeSPIDevelopmentArid ecosystem |
| spellingShingle | Zulqadar Faheem Jamil Hasan Kazmi Saima Shaikh Sana Arshad Noreena Safwan Mohammed Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan Ecological Indicators Change Detection Climate Change SPI Development Arid ecosystem |
| title | Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan |
| title_full | Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan |
| title_fullStr | Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan |
| title_full_unstemmed | Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan |
| title_short | Random forest-based analysis of land cover/land use LCLU dynamics associated with meteorological droughts in the desert ecosystem of Pakistan |
| title_sort | random forest based analysis of land cover land use lclu dynamics associated with meteorological droughts in the desert ecosystem of pakistan |
| topic | Change Detection Climate Change SPI Development Arid ecosystem |
| url | http://www.sciencedirect.com/science/article/pii/S1470160X24001274 |
| work_keys_str_mv | AT zulqadarfaheem randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan AT jamilhasankazmi randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan AT saimashaikh randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan AT sanaarshad randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan AT noreena randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan AT safwanmohammed randomforestbasedanalysisoflandcoverlanduselcludynamicsassociatedwithmeteorologicaldroughtsinthedesertecosystemofpakistan |