Experiment for mapping land cover and it’s change in southeastern Sri Lanka utilizing 250 m resolution MODIS imageries

Forest-cover change has become an important topic in global biodiversity conservation in recent decades because of the high rates of forest loss in different parts of the world, especially in the tropical region. While human interventions are the major cause, natural disasters also contribute to forest cover changes. During the past decades, several studies have been conducted to address different aspects of forest cover changes (e.g., drivers of deforestation, degradation, interventions) in different parts of the world. In Sri Lanka, increasing rates of forest loss have been recorded during the last 100 years on a regional basis, especially in the dry zone. However, Sri Lanka needs detailed studies that employ contemporary data and robust analytical tools to understand the patterns of forest cover changes and their drivers. The dry zone of Sri Lanka encompasses 59% of the total land area of the country, ergo, the most extensive forest cover. Our study analyzed forest cover dynamics and its drivers between 1992 and 2019. Our specific objectives included (i) producing a forest cover map for 2019, (ii) analyzing the spatiotemporal patterns of forest cover changes from 1992 to 2019, and (iii) determining the main driving forces. Landsat 8 images were used to develop forest-cover maps for 2019, and the rest of the forest cover maps (1992, 1999, and 2010) were obtained from the Forest Department of Sri Lanka. In this study, we found that the dry zone had undergone rapid forest loss (246,958.4 ha) during the past 27 years, which accounts for 8.0% of the net forest cover changes. From 2010 to 2019, the rates of forest loss were high, and this can be associated with the rapid infrastructure development of the country. The findings of this study can be used as a proxy to reform current forest policies and enhance the forest sustainability of the study area.

https://www.mdpi.com/1999-4907/11/8/836/pdf

Multi-Decadal Forest-Cover Dynamics in the Tropical Realm: Past Trends and Policy Insights for Forest Conservation in Dry Zone of Sri Lanka

By the end of this century, the global population will increase by another billion people, making a growing threat to the natural resources. The situation in developing countries is critical, and only less than 10 per cent of low income countries are having an increasing annual growth rate of their forest and woodland areas. Population control is beyond the curriculum of environmental scientists although resources management is a timely important topic to them. Newly developed satellite remote sensing and GIS have crucial roles in this field, but the application of these sciences is showing large differences in developed and developing countries. A number of barriers for remote sensing and GIS in developing countries have been identified. Clear recognition of these limitations gives valuable solutions, which we have forwarded here as a cycle of solutions. A simple case study is included in this paper to explain the research steps, alternative low cost routings and the usefulness of a satellite da...

Do remote sensing and GIS have a practical applicability in developing countries? (including some Sri Lankan experiences)

GIS supplies comprehensive techniques for land suitability classifications. GIS based research has been carried out to extract new land for paddy cultivation in south Sri Lanka with special concern on environmental conservation. The locations of the worth considering lands and impractical lands for paddy cultivation were analyzed and merged with the GIS data base by a specially arranged point system. About 71% of the study area was directed to the suitability classification and rest was reserved as forest and water with buffer zones, present paddy, river and roads. More than 72% of the selected land was classified as highly suitable or suitable for paddy cultivation. The methodology can be implemented on any area with slight adjustments in the points system according to the physical and climatological conditions of the particular area. >

Land suitability assessment by GIS for paddy cultivation in Sri Lanka. With a concern on environmental protection

Digitally processed Landsat MSS data and GIS technology were applied to work out a Land Suitability classification for reforestation activities in southern Sri Lanka. The methodology was designed to solve three main problems, i, e., 1) Where dose forest cover exist? 2) Where are the regions suitable for reforestation? and 3) Where are the regions impossible for reforestation?. About 17% of the total land area was classified as practically considerable for reforestation directed to the land suitability classification. More than 80% of this area was categorized as highly suitable or suitable lands for reforestation. If this reforestation program is implemented, the total forest area will increase from 12% to 25% in the study area.

Land Suitability Assessment for Reforestation in Southern Sri Lanka

Based on Landsat MSS data of 1977 and 1987 together with in situ surveys, landcover and landuse maps are made for the central southern part of Sri Lanka then the change of landcover and deforestation in 10 years period is estimated. The change amounts to 16 % decrease in evergreen forests, 39 % increase in scrublands and 63 % decrease in open water areas.Analyses of 30 yeras monthly and annual mean precipitation and air temperature data collected from the surrounding observing stations indicate a complicated distribution pattern with a fairly large variability in case of precipitation while in case of air temperature 5 years moving average of the annual air temperature indicates a steady increase. These facts well coincide with the villagers perception on the natural environments.

Identification of Forest Cover Changes by Landsat MSS Data and Environmental Effects of Such Changes in Central South Sri Lanka

Classification of Categories on Remote Sensing Multispectral Image Data by the Angular Distribution in the Feature Space

On Category Classification of Remotely Sensed Multispectral Image by Linear Algbraic Method

https://www.jstage.jst.go.jp/article/rssj1981/15/3/15_3_232/_pdf

Hiromichi Toyota

Papers

Identification of Forest Cover Changes by Landsat MSS Data and Environmental Effects of Such Changes in Central South Sri Lanka

Classification of Categories on Remote Sensing Multispectral Image Data by the Angular Distribution in the Feature Space

On Category Classification of Remotely Sensed Multispectral Image by Linear Algbraic Method

Three Dimensional Histogram for Analysis of Remote Sensing Multi-spectral Image.

Measurement of Diffuse Property Based on the Change of Spatial Distribution of Sensitivity in the Field of View of an IR Radiometer