10:45 AM - 11:00 AM
[HTT15-06] Identification of Abandoned Tea Lands in Kandy District, Sri Lanka Using Satellite Remote Sensing
Keywords:Tea abandonment, LULC, Trajectory, Landsat, Remote Sensing, GIS
Tea (Camellia sinensis) stands as a prominent cash crop in global agriculture, and it is also the top agricultural export in Sri Lanka, known as 'Ceylon Tea.’ It provides employment to nearly 1 million people, while the land area covered in tea lands amounted to 202,540 ha. The establishment of the commercial tea plantation in Sri Lanka dates back to 1867 at the Loolkandura Estate in Kandy District. However, over the past decade, many tea lands in Sri Lanka have been abandoned due to several reasons. Consequently, Sri Lankan tea production has experienced a gradual decline, amounting to a 2.6% decrease, and directly affected the economic revenue loss. Given the context, proper spatial identification of the abandoned tea lands in Sri Lanka is needed, as no significant studies have been conducted to examine abandoned locations concerning land use changes. Therefore, this research aims to identify the spatial distribution of abandoned tea lands in response to land use/cover (LULC) to provide a substantial knowledge base on the abandonment of tea lands. This study is significant because previous research has only focused on quantifying the impact of climate factors and assessing climate suitability for tea cultivation. Additionally, studies have analyzed tea land dynamics by considering factors such as climate, topography, and population in specific areas, as well as simulating future occurrences. Some research has also discussed the environmental covariates of tea lands and the role of policy in the Sri Lankan tea sector from non-spatial perspectives.
The study area consists of five local administrative divisions called Divisional Secretariat Divisions in the Kandy District, an economic hub in the central part of Sri Lanka, namely Delthota, Doluwa, Udapalatha, Ganga Ihala Korale, and Pasbage Korale, where approximately 70% of the tea lands are located. This study employed Random Forest (RF) classification and LULC trajectory methods with the Landsat satellite images which were represented every two years, from 2015 to 2023, and Google Earth images to assess the tea land abandonment. Six LULC classes were used for the classification: Tea, Home Garden and Other Crops, Forest, Grass and Bare Land, Built-up Area, and Water Body. The abandonment of tea lands was considered if they were converted to another land use between 2015 and 2023.
The results of RF classification show an overall accuracy of 85%, indicating that tea is the second-largest land use type. The study area is mostly covered with Home Garden and Other Crops while the Built-up Area expanded year by year. The temporal analysis reveals a notable decline in tea lands, with a shift from 9,818 ha (21.6%) in 2015 to 8,446.4 ha (18.6%) in 2023. A total of 22 LULC trajectories were identified in trajectory analysis, which is based on the time series of each pixel, rather than searching for single change events between two datasets since it involves tracking the changes in specific features over time. The trajectory results emphasize that tea abandonment mainly occurred between 2017 and 2023, with only 3.7% (1676.7 ha) of pixels continuously classified as tea lands over the 8 years, while most of the tea lands were converted to Home Garden and Other Crops. This instability may result from various factors, such as shifts in land use, agricultural practices, socioeconomic problems, or environmental changes affecting tea lands. About 35% of the abandoned occurred within the length of 7 years and significant changes were observed between the two urban centers of Gampola and Nawalapitiya towns suggesting a dynamic interaction between urban development and tea land abandonment. These findings provide vital geographical information for the stakeholders involved in tea cover enforcement. Spatially evaluating the factors influencing tea land abandonment in the study area is one of the main directions for future studies.
The study area consists of five local administrative divisions called Divisional Secretariat Divisions in the Kandy District, an economic hub in the central part of Sri Lanka, namely Delthota, Doluwa, Udapalatha, Ganga Ihala Korale, and Pasbage Korale, where approximately 70% of the tea lands are located. This study employed Random Forest (RF) classification and LULC trajectory methods with the Landsat satellite images which were represented every two years, from 2015 to 2023, and Google Earth images to assess the tea land abandonment. Six LULC classes were used for the classification: Tea, Home Garden and Other Crops, Forest, Grass and Bare Land, Built-up Area, and Water Body. The abandonment of tea lands was considered if they were converted to another land use between 2015 and 2023.
The results of RF classification show an overall accuracy of 85%, indicating that tea is the second-largest land use type. The study area is mostly covered with Home Garden and Other Crops while the Built-up Area expanded year by year. The temporal analysis reveals a notable decline in tea lands, with a shift from 9,818 ha (21.6%) in 2015 to 8,446.4 ha (18.6%) in 2023. A total of 22 LULC trajectories were identified in trajectory analysis, which is based on the time series of each pixel, rather than searching for single change events between two datasets since it involves tracking the changes in specific features over time. The trajectory results emphasize that tea abandonment mainly occurred between 2017 and 2023, with only 3.7% (1676.7 ha) of pixels continuously classified as tea lands over the 8 years, while most of the tea lands were converted to Home Garden and Other Crops. This instability may result from various factors, such as shifts in land use, agricultural practices, socioeconomic problems, or environmental changes affecting tea lands. About 35% of the abandoned occurred within the length of 7 years and significant changes were observed between the two urban centers of Gampola and Nawalapitiya towns suggesting a dynamic interaction between urban development and tea land abandonment. These findings provide vital geographical information for the stakeholders involved in tea cover enforcement. Spatially evaluating the factors influencing tea land abandonment in the study area is one of the main directions for future studies.