Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Thirty-year Spatiotemporal Change Record of Sundarban Mangrove Forest in Bangladesh

K. M. M. Uzzaman
GIS and Remote Sensing Lab., Department of Agroforestry and Environment, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

M. G. Miah
GIS and Remote Sensing Lab., Department of Agroforestry and Environment, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

H. M. Abdullah
GIS and Remote Sensing Lab., Department of Agroforestry and Environment, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

M. R. Islam
Department of Agronomy, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

M. S. I. Afrad
Department of Agricultural Extension and Rural Development, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

M. J. Hossain
GIS and Remote Sensing Lab., Department of Agroforestry and Environment, Bangabandhu Sheikh Mujibur Rahman Agricultural University, Bangladesh.

Abstract/ Executive Summary:

Accurate and realistic forest cover change assessment is essential for the conservation and management of the Sundarban mangrove forest of Bangladesh. With these views, an integrated way of the vegetation cover assessment was conducted using time-series Landsat satellite imageries of 1991, 2001, 2011, and 2021. During the last 30-year (1991-2021), variations in four land cover classes viz. healthy vegetation, unhealthy vegetation, water body, and sandbar were recorded. It showed a decreasing trend of forest vegetation and a subsequent increase of water bodies during the study period. The healthy vegetation and unhealthy vegetation decreased at 1.33 and 1.66%, respectively, whereas water bodies increased 2.55% at the same time. The healthy vegetation consistently decreased over the decades, but unhealthy vegetation decreased during the 2001-2011 period. Conversion from healthy vegetation to unhealthy vegetation and unhealthy vegetation to healthy vegetation during 1991-2001 was similar. Such transform was much higher from unhealthy to healthy vegetation during 2001-2011. Transformation of healthy vegetation to unhealthy vegetation was remarkably higher during the 2011-2021 period. Further continuous change detection and classification algorithm (CCDC) showed a stable pattern over the study period without significant breakpoints. This study reveals the need for regular mangrove forest monitoring. The findings of this study can be used as a reference in the formulation and implementation of sustainable mangrove forest conservation and management.

Keywords: Sundarban, Forest vegetation, Change detection, Sustainable management.

Location: Sundarban mangrove forest of Bangladesh.

Start Date: 00-00-1991

End Date: 00-00-2021

Program Area: Conservation and Biodiversity

Commodity: Mangrove

Objectives:

The aim is multispectral satellite imagery and quantified spatiotemporal patterns of the Sundarban mangrove forest cover change to achieve this goal.

Methodology:

Sundarban is located at the south of the Tropic of Cancer and remains amid 89°00’E and 89°55’E longitude and 21°30’N and 22°30’N latitude. The study area is the part of Sundarban, situated at the side of southwestern Bangladesh interim the Baleshwar waterway in the East and the Harinbanga stream in the west head-to-head to the Bay of Bengal. The total area of the Sunderban was 15764.96 Sq. Km. Space includes plane isles enclosed with brackish swamplands and thick plantations interrupted by an arrangement of inlets, tidal tributaries, and bays traversed by irregular loud networks. It has a rainy season environment with elevated moisture of around 80.0 percent due to the abundant inlets and the end-to-end Bay of Bengal. The region is minded to whirlwind actions and has a sumptuous past of storms striking the area and governing to damage survivors and belongings. Outside mangrove plantations, which are the leading plants, the forest has great biodiversity. It is also home to a large number of animals, both earthly and aquatic. The biodiversity includes more than 300 species of vascular plants, 250 fishes, and 300 birds, besides numerous species of phytoplankton, fungi, bacteria, zooplankton, benthic invertebrates, mollusks, reptiles, amphibians, and mammals exist. Time series Landsat satellite images of 1991, 2001, 2011, and 2021 were used to analyze mangrove forest cover change. All the images were taken from the United States Geological Survey. The images are composed of Landsat Thematic Mapper (TM) and Operational Land Imager (OLI), respectively, having Path 137 and 138 and Row 45. The images were 30m spatial resolution with less than 1.0 percent cloud cover. Considering the seasonal deviation in the classification output images of a similar period of each year were used. Hence, we used images available in January and February. Surface reflectance (SR) band in 1991, 2001, 2011, and 2021 along with Enhanced Vegetation Index (EVI), Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), Modified Soil Adjusted Vegetation Index (MSAVI) were used to comprise multi-band imagery. Besides Landsat SR bands, vegetation indices were incorporated to prepare the multiband imagery. Vegetation indices (VIs) possess unique characteristics to portray the health of the vegetation. VIs are also helpful to delineate forest from other land covers so we incorporated VIs in the multiband imagery. During the multiband image creation, VIs were incorporated. Classification of satellite images engages the distribution of spectral signatures for several land use and land cover (LULC) classes. Spatial-temporal LULC map aids in identifying the variations in the area of various LULC by emphasizing the trend and pattern of LULC change of a specific place. QGIS software was used and applied the maximum likelihood technique for LULC classification. A complete of 20 sample signatures for each LULC were picked up individually for every satellite image. Therefore, four LULC categories were identified i.e. healthy vegetation, unhealthy vegetation, water bodies, and sandbar. Accuracy assessment is the procedure used to quantify the reliability of a classified image in terms of the real world. The standard accuracy assessment procedure is to construct an “error matrix”. Accuracy assessment essentially determines the quality of the information derived from remotely sensed data. Variation detection investigation Terrestrial transformation investigation was accomplished with the QGIS-SAGA module for the four intermissions: 1991-2001, 2001-2011, 2011-2021, and 1991-2021. This style permitted the researchers to denote the meaningful reforms in the study part. The pixel-based cross-tabulation investigation assisted the quantification of the size of land cover change from one class to alternative and corresponding areas. It also generated a new spatial stratum holding change classes and agreeing on area values. Implementation of nonstop change finding and cataloging algorithm on Sundarban Continuous change detection and classification algorithm (CCDC) is gaining popularity in land use and land cover change science due to its strength in detecting land conversion. The algorithm can detect the land cover change other than noise as the noise in satellite images are ephemeral. CCDC works in two major steps for instance identifying breakpoints to model stable temporal segments and finally assigning class labels to new temporal segments if any. CCDC algorithm was carried out to identify the breakpoints from archived Landsat time-series data. Breakpoints represent changes in land use and land cover. Landsat data from 1989 to 2021 of the study sites were assimilated for the CCDC algorithm.

Source of Information: Ann. Bangladesh Agric. (2020) 24 (2) : 15-32

Article Link (Online Journal): www.doi.org/10.3329/aba.v24i2.55781

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

The study reveals that the Sundarban earmarked forest in Bangladesh had not significantly altered from 1991 to 2021, and it is now stable. However, a temporal variation in areas under different land use classes was recorded. A consistent increase in water bodies but an inconsistent increase or decrease in healthy or unhealthy vegetation are observed in different decadal periods. The storms and cyclones may affect Sundarbans mangrove forest, but a quick regeneration of vegetative growth of the forest occurs that maintained the stability of the mangrove. The study quantified only the distribution and acreage of the healthy forest but not the species level. So, it can not suggest the status of species loss or shift, which is also vital for the sustainable existence of mangrove forests.

Knowledge Management: Journal