Abstract
The Phang-Nga Province on the Andaman Coast of Thailand is lined with mangrove forests. The mangroves had suffered losses to shrimp aquaculture since the 1970s which may have compromised the protective effects the mangroves offered during the tsunami of 2004. After the mangrove forest took the initial hit from the wave, the Thai people learned about the importance of this ecosystem beyond its food resources and knew that there would have been even more extensive damage and a larger death toll without it. Mangrove restoration is now an important on-going project to re-stabilize, strengthen and expand its extent. Forest regeneration is occurring naturally, but the cooperative effort to restore and replant the mangroves is speeding up the process. Young growth and restoration efforts are documented with in situ photography on Koh Prathong and along the Phang-Nga Province, but current satellite imagery was not available to show the extent of the progress. We aimed to measure the amount of mangrove forest prior to and after the tsunami to understand how much damage occurred. We used remote sensing imagery and processed the data in ENVI and ArcGIS.
Introduction
Thailand had never experienced a tsunami before the one that struck on December 26, 2004 from the Indian Ocean 9.0 magnitude earthquake (Figure 1). Located so close to the epicenter, the Andaman Coast of Thailand was highly vulnerable to the large wave (Figure 2) and the people had little time to react with the wave hitting in about 90 minutes (Figure 3). The Phang-Nga Province experienced the highest death toll - 5,990 - (Figure 4) and sustained the most damage within Thailand.
Many areas in Phang-Nga are protected under the conservation areas network, including the Ao Phang-Nga National Park, which covers an area of 4,000 ha and represents the largest tract of original primary mangrove forests remaining in Thailand (Giri et al., 2007). The IUCN recognizes the Andaman Coast of Thailand as an important mangrove location for the Phoenix paludosa, which is on the IUCN Red List as “Near Threatened” (Figure 5). Key species to this region are the Rhizophora apiculata (Figure 6) and the Nypa fruticans (nypa palm, Figure 7), both of which are listed on the IUCN Red List as “Least Concern” (Broekhoven et al., 2005; Rowland, 2010). The nypa palm is particularly valuable to the Thai culture for its use in house construction, cigarette wrappers, fruit, and numerous other uses. For protection against tsunamis, Rhizophora is the first line of defense as its root system allows it to grow further out into the ocean off the shore with the nypa palm second in line since it requires more soil stability to grow to maturity (Rowland, 2010).
Before the tsunami, the local populations were not aware of the important role the coastal mangroves would play in stabilizing the shorelines to protect the villages by reducing the devastating impact of such natural disasters (Giri et al., 2007). Prior to the tsunami, coastal mangroves were destroyed to make room for the rising industry of shrimp aquaculture and plundered as sources of food, medicine, fuel and building materials (Giri et al., 2007). The 2004 tsunami caused major damage to coastal towns and villages; mainland village Ban Talae Nok suffered 80% loss and Ban Pak Chok of Southern Koh Prathong suffered 100% loss of village structures and infrastructure with 33 and 45 lives lost, respectively (Rowland, 2010). However, as bad as the damage was, it would have been a lot worse with a greater loss of life if the mangroves had not absorbed some of the initial wave impact. Mangrove wetlands are particularly valuable in minimizing damage to property and loss of human life by acting as a barrier against tropical storms, such as typhoons, cyclones, hurricanes, and tsunamis (Barbier, 2006). In Phang-Nga, the most affected province, post-tsunami assessments suggest that large mangrove forests in the north and south of the province significantly mitigated the impact of the tsunami. The mangrove stands suffered damage on their seaside fringe, while reducing the tidal wave energy (Barbier, 2006). This event caused a new awareness and respect for the mangrove forests along the Andaman coast of Thailand and multiple reforestation projects (Figure 8, additional images at http://mangroveimages.
The Phang-Nga province [north of Phuket, south of Ranong and including Khao Lak, Khura Buri and Koh Prathong] of Southern Thailand is of particular concern as it is flat and vulnerable (as demonstrated by the slope in Figure 9) being located on the Andaman coast on the Indian Ocean side of the country where earthquake activity continues to disturb the Indonesian and Thai coasts. Wave destruction of vegetation is exemplified in the time series image of Khao Lak (Figure 10), where the red areas (vegetation) in 2003 have turned gray (barren) on December 31, 2004.
No study denies there was damage to the mangroves from the tsunamis, but the debate is how much wave damage the mangroves sustained, how much the mangroves were degraded before the tsunami, and how much the compromised forest decreased its effectiveness as a bio-shield. In this paper we are interested in the change of the amount of mangrove forest along the coast and compare the pre- and post-tsunami coastal mangrove vegetation. With the continuing threat of earthquakes in the Indian Ocean and the subsequent wave potential it is very important that the mangroves achieve their pre-tsunami state, if not surpass it.
Methods
Study Area
This study covers the Phang-Nga Province of the Andaman Coast of Thailand, which is on the West Coast, north of Phuket and south of Ranong and includes Koh Prathong Island (Figure 11). The Andaman Sea is part of the Indian Ocean between India, Thailand, and Indonesia, where the 9.0 magnitude earthquake struck. The Phang-Nga Province was one of the hardest hit regions by the 2004 tsunami. In this study we focus on the vulnerable area of Koh Prathong Island and the inlet areas between the island and Ko Koh Khao, Koh Ra, and the mainland.
Data
Three Landsat images were downloaded in geotiff format from Global Land Cover Facility in Maryland. The Landsat images show the study area in 1991, 2002, and 2005. After subsetting the images and geographically linking them in ENVI to uniformly display the study area, we created the time series image showing them side by side (Figure 12). The 1991 image is a Landsat 5 TM image and the images from 2002 and 2005 are both ETM+ taken by Landsat 7. All three images are in 30m resolution.
In ENVI we created NDVI images, from 2002 and 2005 geotiffs using bands 4, 5, and 3 (IR, near IR, and Red). Red, IR, and near IR bands were used because they clearly show wetland vegetation. We only used the two ETM+ images to create density slice images from band 5 with only green, blue and yellow ranges left alone and red edited into black. The earlier TM image was not comparable when manipulated by the Density Slice function. The blue and green colors show the extent of the mangrove trees pre- and post-tsunami (Figure 13a).
The Density Slices were turned into shape files and processed in ArcGIS, which eliminated most of the noise and background colors. The mangrove vegetation colors were edited to be green and blue to match the Density Slice results, which make the mangrove vegetation even more easily identifiable and quantifiable. We zoomed in into two specific areas (North and South edges of Koh Prathong and along the inlets to analyze the change in pixel distribution, comparing the data form 2002 and 2005. (Figure 13b).
An SRTM file was acquired from the Global Land Cover Facility and was processed in ENVI to create a slope model, which demonstrates the flatness and vulnerability of Koh Prathong and the surrounding areas to a tsunami (Figure 9).
Data Analysis
Visually comparing the density slices and ArcGIS images, we can clearly see that there was a change in the mangrove cover, shown in blue and green, from 2002 to 2005. Due to the lack of in situ data, we were not able to discern the difference between green and blue vegetation pixels. We are assuming that it is either the height difference, density, greenness, or the two different predominant Mangrove species (Rhizophora and Nypa fruticans).
Results
Comparing images from 1991 and 2002, there are some areas that suffered a loss of mangrove vegetation due to shrimp farming practices, however, it is less than expected and in some areas the mangrove vegetation even increased. When comparing the images from 2002 and 2005, there are areas with very evident mangrove vegetation loss – especially on the southern and northern tips of Koh Prathong Island and along the inlet of the ocean between Koh Prathong and Ko Koh Khao islands and mainland Thailand (Figures14a, 14b, 15a, 15b). More recent satellite images were not available, but studies done in situ suggest that reforestation is expanding the mangroves in these key areas.
Discussion
Comparisons
It was estimated that 168,682 hectares of mangrove forests areas existed in 1993 and that more than 50% of mangrove forests were lost during 1961-1993 (Yousif, et al., 1997). Various activities carried out in mangrove forest areas such as shrimp farming, tin-mining activities, mangrove over-exploitation, industrial area and settlements leading to the reduction of mangrove forests (Yousif, et al., 1997).
In some areas, mangrove forests hit by the Indian Ocean tsunami suffered severe damage from breaking and uprooting. Recent findings suggest that the continued destruction and degradation of mangrove forests over the past few decades has decreased the protective capacity of mangrove forest ecosystems and reduced their ability to rebound from natural disasters (Giri et al., 2007). Although there are no critically endangered species in our study area, mangrove forests have been declining in biological diversity and economic value with many flora and fauna vulnerable, near-threatened, threatened, endangered or critically endangered (Giri et al., 2007).
The study by the IUCN found that compared to other ecosystems, the tsunami had relatively little impact on the mangrove forests: a band of a few meters depth into the mangroves can be observed where debris is deposited and where some trees or branches are broken. No signs of any serious uprooting of mangrove trees were observed, except on one shallow sandbank in between Koh Prathong and Koh Ra, where a number of trees, including mangrove trees have been washed away (Broekhoven, et al., 2005). This is a relative statement and we can identify areas in our images where other vegetation suffered more extreme damage, but the IUCN report does not argue against the damage to mangroves, it just does not quantify the extent.
Limitations
There are limitations on completing accurate studies on these areas due to the lack of in situ data to measure and quantify the amount of trees and biomass. Satellite imagery allows us to analyze and compare vegetation cover but does not offer precise species and density breakdown. Many changes are visually identifiable as change but not what the change is. Many blue areas turned green and many yellow areas disappeared entirely. It is unclear if the color variations are due to species variety, density differences, greenness values, etc. Onsite recording is necessary to clarify. IUCN identified species (Table 1) have been verified in person, but not the extent or exact locations of each.
Future Research
Mangrove forests need to be properly quantified in situ to compare with satellite imagery. Newer images need to be analyzed to verify the assumption of forest regeneration and reforestation. Evidence of broken and dead trees still exist in the inlet areas and the distance can be measured to the old growth left standing to calculate the amount of loss sustained by the tsunami wave.
References
Papers cited
Barbier, Edward B. (2006) Natural barriers to natural disasters: replanting mangroves after the tsunami. Front Ecol Environ, 4(3), 124–131
Broekhoven,Guido, Jerker Tamelander, Maria Osbeck, Somsak Soonthornnawaphat and Mattias Rust. (2005) Koh Prathong – Post Tsunami Resource and Livelihood Impact Assessment. IUCN.
Chang, Stephanie E., Beverley J. Adams, Jacqueline Alder, Philip R. Berke, Ratana Chuenpagdee, Shubharoop Ghosh and ColetteWabnitzc. (2004) Coastal Ecosystems and Tsunami Protection after the December Indian Ocean Tsunami. Earthquake Spectra, 22(S3), S863–S887
Giri, C., Z. Zhu, L. L. Tieszen, A. Singh, S. Gillette and J. A. Kelmelis. (2007) Mangrove forest distributions and dynamics (1975—2005) of the tsunami-affected region of Asia. Journal of Biogeography, 35, 519-528.
Rowland, Janice A. (2010) Personal Research Studies via Andaman Discoveries Travel Study Program.
Yousif, Somjai Sremongkontip, Ali Hussain and Liza Groenindijk. (1997) Mangrove Forest Rehabilitation and Management in Pang NGA Bay, Southern Thailand Using Remotely Sensed Data And GIS 1. The international institute for aerospace survey and Earth Science (ITC).
Images Used
IUCN: http://www.iucnredlist.org/
NASA - EQ: http://earthobservatory.nasa.
NASA - Khao Lak: http://earthobservatory.nasa.
NOAA: http://nctr.pmel.noaa.gov/
Rowland: Rowland, Janice A. (2010) Personal Research Studies via Andaman Discoveries Travel Study Program
UNEP: http://www.grid.unep.ch/
