Mangrove cover dynamics and community zonation in Gulf of Kachchh – A study based on synergistic use of optical and microwave satellite data

Abstract

This research work has demonstrated efficacy of digital classification techniques using both optical as well microwave satellite data in inventory and monitoring of mangrove cover and community classes of the mangrove covered regions of the Gulf of Kachchh, Gujarat, India and has attempted to understand causes for changes in the mangrove habitats.The present work has utilized optical and microwave satellite data for studying mangrove cover dynamics and community zonation in the Gulf of Kachchh, Gujarat, India.The Gulf of Kachchhis a semi-enclosed basin situated in the north-west part of the Indian coast in the state of Gujarat. Occupying an area of approximately 7300km2, this region is characterized by the presence of shoals, channels, inlets, creeks, mudflats, islands, mangroves and coral reefs. The Gulf of Kachchh is a highly energetic macro-tidal system of the northeastern Arabian Sea. The tidal ranges in the Gulf of Kachchh reach upto 7.2 m. The region is characterized by arid/semi-arid climate with an average annual rainfall of 50 cm. The mangroves in the Gulf show spatial variability in extent, composition and condition based on dominant coastal processes, geomorphological setting and amount of protection from anthropogenic influences.Mangrove cover dynamics has been studied for five mangrove occupying regions of the Gulf viz., i) around Kori creek along north-west coast, ii) around Mundra along northern coast, iii) around Kandla along northern coast, iv) around Satsaida bet in north-eastern coast, and v) Marine National Park and Sanctuary (MNP&S) along southern coast of the Gulf of Kachchh.This study has observed that in all the five study regions mentioned above, Avicennia marinais the only dominant species, with smaller patches of Rhizophora mucronata, Rhizophora apiculata, Acanthus ilicifolius and Ceriops tagal. It has been possible to digitally classify mangroves as per density for all the regions. Based on existing knowledge about various classes in the inter-tidal zone, and understanding of spectral signatures, field visits, seven classes were chosen for supervised classification viz., Mangrove Dense (MD), Mangrove Sparse (MS), Mangrove Degraded (MDeg), Intertidal Mudflat (IM), Subtidal Mudflat (SM), Salt Encrustation (SE) and Sea Water (SW). Class separability has been evaluated using transformed divergence method, which involves computation of spectral distance between signatures of various classes by taking into account their statistical parameters such as mean, variance and covariance. Class separability value is well above the threshold 1700 (i.e., 1823.6 – 2000) for all seven classes used in this study. In fact mostly the range in near to 2000, except locations where attempt to discriminate mangrove species within MNP&S was made, the value is relatively lower (1823.6). Accuracy assessment for all the classified images shows overall accuracy range 89%-97.64% and corresponding Kappa values ranges 0.87-0.96. Temporal changes in extent and condition of mangrove classes during time frame 2011 and 2017 have been quantified based on the corresponding classified Landsat images. The results show that mangrove covered area in the entire Gulf of Kachchh has increased by 254.73 sq km during time frame 2011-2017. Mangrove covered area mapped is 963.62 sq km for 2011 and 1218.35 sq km for 2017 for the entire study area. Out of five regions, Satsaida bet and environs has shown 57.18 % increase, Kori creek and environs 18.81 % increase, Marine National Park and Sanctuary 10.70 % increase, Kandla and environs 7.33 % increase and Mundra and environs 5.98 % increase in mangrove cover. In general, there is decrease in area under degraded mangrove class and increase in area under sparse mangrove class. Primary reason of increase in mangrove cover is due to serious plantation efforts carried out by various agencies such as Gujarat Forest Department, Marine National Park (MNP) authorities, Gujarat Ecology Commission etc. in collaboration with local communities and organizations. There has been lot of developmental activities along the inter-tidal region of Gulf of Kachchh, major ones being development of Ports such as Mundra and massive expansion of salt industry, which led to destruction of mangroves at some locations prior to 2011, however as per this study, efforts and care being taken on mangrove plantation is resulting into positive results and there is substantial increase in mangrove cover during current decade. There has been an increase of about 48 sq km of mangroves around Kori Creek in 2017 compared to 2011. Dense mangroves have increased by about 18 sq km, sparse mangroves have shown an increase of about 27 sq km and mangroves under degraded class have increased by 2.5 sq km. The increase is more in the category of dense mangroves (63% increase) which reflects concomitant improvement of mangrove density in this region.It seems that due to lot of nutrients being brought out along with the sediments of Indus river and less anthropogenic activities due to its proximity to International border, growth of mangroves in this region is more due to natural processes rather than plantation efforts. Around Mundra, an increase of about 15 sq km of mangroves was observed during the period 2011-2017, largely because of approx. 17 sq km of increase in the sparse mangrove category, which is probably result of plantation activities. However, around1sq km of dense mangroves and 0.69 sq km of degraded mangroves were destroyed in this region due to expansion of port-related activities. In Kandla region, there has been an increase of roughly 19 sq km of sparse mangroves during 2011 and 2017, which is also probably result of plantation activities. However, the study has also observed decline of about half sq km of dense mangroves in this region because of port-related activities and construction of saltpans. The mudflats around Satsaida bet showed an increase of 145.62 sq km of mangrove during the mapping period of 2011-2017, which is around 57.18 % increase among the five regions studied. Here, the increase is mainly because of increase of 201 sq km of sparse mangroves. However, the study also observed reduction of approx. 20 sq km of dense mangroves due to construction of saltpans. The area under ‘mangrove degraded’ category declined by about 36 sq km as their condition improved and they were categorized as sparse mangroves in 2017. Mangrove cover in the Marine National Park and Sanctuary (MNP&S), located along southern coast of the Gulf of Kachchh is 190.52 sq km in 2011 and 217.85 sq km in 2017. There has been an increase in mangrove cover of about 27.23 sq km (10.69% among five regions studied). The increase is mainly in Avicenniasparse mangrove class (~ 25 sq km). The increase is observed mainly on islands such as Pirotan, Mundeka-Dideka, Kalubhar, Dhani and on coastal belts adjoining Balachadi, Jodiya, near Hansthal Creek and Sikka. Cause of increase in mangrove cover is primarily due to plantation efforts by authorities of the MNP&S. These plantations were carried out under various schemes such as Cher Plantation, Coastal Border Plantation etc. An attempt has been made to evaluate potentials of C-band HH RISAT-1 MRS (Medium Resolution ScanSAR mode) data in conjunction with Resourcesat-2 LISS-IV (Linear Imaging Self Scanner-IV) datafor identifying mangrove communities in Jindra-Chhad Island complex located within Marine National Park and Sanctuary (MNP&S), with a primary focus on developing approach for improving discrimination of mangrove communities by synergistic use of microwave and optical data.Three different approaches were used to combine the SAR and optical data to explore the synergistic potential of the SAR data with optical data, for discriminating different mangrove communities in the study region. The study has carried out supervised classification for mangrove community zonation using Maximum Likelihood Classifier i) using LISS-IV data, ii) using RISAT-1 SAR data, iii) Integrating RISAT-1 SAR and three LISS-IV bands, iv) Merging RISAT-1 SAR and LISS-IV using IHS method and v) Integrating RISAT-1 to the band ratios derived from LISS-IV bands and has subsequently evaluated the results qualitatively as well quantitatively. All approaches resulted into classified images showing seven classes viz., Avicennia Dense (AD), Avicennia Sparse (AS), Rhizophora-Ceriops Dense (RCD), Intertidal Mudflat (IM), Hightidal Mudflat (HM), Sand and Sea. Quantitative evaluation of classified images was done through class separability analysis using transformed divergence method and accuracy assessment. The results indicate that addition of RISAT-1 SAR significantly improves the class separability. All the combinations employing optical and SAR data, have shown increased class separability compared to either LISS-IV or RISAT-1 when used alone. In addition, when all the four channels (two visible, one NIR and one microwave) are integrated and used together, the separability and accuracy obtained is highest. Replacing hue with SAR data yielded image that provided better class separability value (1847.51) than replacing either intensity (1811.78) or saturation (1829.72).The band ratio approach which segregated the nonvegetative component from the vegetative one in the image, yielded image with relatively lower class separability value (1867.27) than those obtained by integration method, but this approach provided relatively higher class separability value than those obtained by IHS method.The study demonstrated that the synergistic use of RISAT-1 C-band MRS and Resourcesat-2 LISS-IV data improves mangrove community discrimination. The mangrove communities discriminated are Avicennia Dense, Avicennia Sparse, and Rhizophora-Ceriops Dense. Among the different approaches of merging SAR with optical data, maximumseparability among mangrove community classes could be obtained by integrating SAR data with Red, Green and NIR bands of optical data. The work provides the latest estimates for mangroves in the Gulf of Kachchh. In addition, the community zonation done through synergistic use of microwave and optical data provides new approach for mangrove studies using disparate datasets.