论文部分内容阅读
Abstract: The analysis of shoreline positions in the most severely impacted area in the South of Thailand namely Phang Nga Province was conducted using time series remotely sensed data from LANDSAT-TM and ASTER acquired in 1991, 2003, 2005, 2006, and 2010. A density slice method was applied on a near infrared band to estimate the rate of accretion and erosion in the pre- and the post-2004 Indian Ocean tsunami. The impact of the 2004 tsunami caused a rate of erosion that varied from a few meters to a few kilometers inland. Six years after the catastrophic disaster, some of the severely damaged (> 300 m) had recovered more than 50% to the pre-tsunami positions. The accretion and erosion processes in the post-tsunami affected many locations along the coastal area more than the pre-tsunami (at a rate of ± 40 m). The variability in the shorelines at Nok Na, Pak Ko, Nam Khem, Bang Khaya Nai and Pramong villages in Takua Pa District affected the political boundary of the sub-district in Phang Nga. In addition, the changes in shoreline positions will affect the tourism of Phang Nga, changing the shape of beaches in front of some beautiful resorts and hotels and the erosion will cut into land parcels along the coastal areas. The method and results from this study can be used for coastal zone study and/or management.
dunes [1]. In Phang Nga, a famous upscale resort area north of Phuket was completely devastated by the waves that exceeded 10-m high in places and almost 80% of the death toll in Thailand was counted here. According to our surveys and information from the inhabitants, erosion and accretion processes of seashores at some locations appeared faster and in different direction/form compared to the past (before tsunami), affected with legal property of boundary definition and tourism at some locations. As a result, the positions of shoreline in Phang Nga in the pre- and post-2004 Indian Ocean tsunami were analyzed in this study. Instead of using only time-consuming and costly field measurements in the large tsunami-impacted areas, multi-temporal remote sensing (RS), geographic
The analysis focused on three districts, the most severely impacted on the western coast of Phang Nga: Tai Muang, Takua Pa, and Kuraburi, as shown in Fig. 1. The topography of Phang Nga consists of a hilly and mountainous area ranging from north to south and approximately 239.25 km of the coastal area in the west along the Andaman Sea. Tides in Phang Nga are semidiurnal with two high tides and two low tides each day, with an average tidal range of 1.55 m [2].
Technically, a shoreline is the intersection between the mean high water line and the shore. The line delineating the shoreline on nautical charts (sea maps) approximates this mean high water line [3]. In this study, water levels were concerned for estimating the affect of the tide level on the position of shorelines [4]. The level of water was predicted in meters above the lowest low water by two local stations of the Hydrographic Department in the Royal Thai Navy namely Ao Tap Lamu and and Khuraburi in Phang Nga. Based on acquisition time of the images used in this study, water levels during 10:30-11:04 am (seven
2005 by the Thematic Mapper (TM) and the Enhanced Thematic Mapper Plus (ETM+) sensors respectively were gathered in this study. The acquisition time lag between ASTER and LANDSAT is about 30 minutes. LANDSAT scenes were downloaded from the Global Land Cover Facility (www.landcover.org) for our research purposes and ASTER scenes were supported from GEO Grid, the National Institute of Advanced Industrial Science and Technology (AIST), Japan. 4.2 Methods
In this study, the shoreline detection techniques applied to visibly discernible shoreline features are image processing or semi-automatic interpretation from satellite data. The selected geo-coded ASTER and LANDSAT images were co-registered or adjusted to the correct scale (15 m spatial resolution). Extraction of the shoreline on specific tidal flats has rarely been investigated in depth. The reflectance of the flats’ tidal mud is affected by various parameters including particle size, moisture content, local slope, and turbidity of seawater. A differentiate between water and land was examined in each spectral band, a ratio vegetation index called Normalized Difference Vegetation Index (NDVI), and a false-color composite of ASTER and LANDSAT images. An appropriate input band which has the highest contrast between water, and land was selected for further analysis [5]. The NDVI formula is presented in Eq. (1) [6]:
In this study, multi-temporal remotely sensed data from LANDSAT-TM and ASTER acquired in 1991, 2003, 2005, 2006 and 2010 were analyzed using a density slice method that was applied on a near infrared band to estimate the rate of accretion and erosion in the pre- and the post-2004 Indian Ocean tsunami. The analysis was focused on the most severely impacted area by the tsunami, Phang Nga Province, located in southern Thailand. The changes in the shoreline positions of 2003 compared with those of 1991 presented a rate of erosion of 20-154 m, and the most severe erosion occurred at Pramong Village in Takua Pa District. In 2003, the accretion was significant at Nok Na Village on Kho Khao Island (185 m) compared to its shoreline position in 1991. Two months after the catastrophic tsunami, in 2005, the rate of erosion was 18-652 m, and locations with erosion rate of more than 300 m were identified as severely damaged. In 2006, most of the eroded shoreline positions had recovered more than 50-70% to the pre-tsunami positions, especially the positions with a rate of erosion of less than 100 m. While most of the shoreline positions with a rate of erosion of more than 300 m in 2005 had more than 60% recovered in 2010 and were resumed closer to the positions in the pre-tsunami (2003). In 2010, six years after the disaster, significant variation in the shoreline was observed in Takua Pa District, the shorelines at Nok Na, Pak Ko, and Nam Khem villages had changed their alignments. In the absence of a wave breaker and a management plan at these locations, the continuous erosion that changes the shoreline positions might also affect the political boundary of Phang Nga
dunes [1]. In Phang Nga, a famous upscale resort area north of Phuket was completely devastated by the waves that exceeded 10-m high in places and almost 80% of the death toll in Thailand was counted here. According to our surveys and information from the inhabitants, erosion and accretion processes of seashores at some locations appeared faster and in different direction/form compared to the past (before tsunami), affected with legal property of boundary definition and tourism at some locations. As a result, the positions of shoreline in Phang Nga in the pre- and post-2004 Indian Ocean tsunami were analyzed in this study. Instead of using only time-consuming and costly field measurements in the large tsunami-impacted areas, multi-temporal remote sensing (RS), geographic
The analysis focused on three districts, the most severely impacted on the western coast of Phang Nga: Tai Muang, Takua Pa, and Kuraburi, as shown in Fig. 1. The topography of Phang Nga consists of a hilly and mountainous area ranging from north to south and approximately 239.25 km of the coastal area in the west along the Andaman Sea. Tides in Phang Nga are semidiurnal with two high tides and two low tides each day, with an average tidal range of 1.55 m [2].
Technically, a shoreline is the intersection between the mean high water line and the shore. The line delineating the shoreline on nautical charts (sea maps) approximates this mean high water line [3]. In this study, water levels were concerned for estimating the affect of the tide level on the position of shorelines [4]. The level of water was predicted in meters above the lowest low water by two local stations of the Hydrographic Department in the Royal Thai Navy namely Ao Tap Lamu and and Khuraburi in Phang Nga. Based on acquisition time of the images used in this study, water levels during 10:30-11:04 am (seven
2005 by the Thematic Mapper (TM) and the Enhanced Thematic Mapper Plus (ETM+) sensors respectively were gathered in this study. The acquisition time lag between ASTER and LANDSAT is about 30 minutes. LANDSAT scenes were downloaded from the Global Land Cover Facility (www.landcover.org) for our research purposes and ASTER scenes were supported from GEO Grid, the National Institute of Advanced Industrial Science and Technology (AIST), Japan. 4.2 Methods
In this study, the shoreline detection techniques applied to visibly discernible shoreline features are image processing or semi-automatic interpretation from satellite data. The selected geo-coded ASTER and LANDSAT images were co-registered or adjusted to the correct scale (15 m spatial resolution). Extraction of the shoreline on specific tidal flats has rarely been investigated in depth. The reflectance of the flats’ tidal mud is affected by various parameters including particle size, moisture content, local slope, and turbidity of seawater. A differentiate between water and land was examined in each spectral band, a ratio vegetation index called Normalized Difference Vegetation Index (NDVI), and a false-color composite of ASTER and LANDSAT images. An appropriate input band which has the highest contrast between water, and land was selected for further analysis [5]. The NDVI formula is presented in Eq. (1) [6]:
In this study, multi-temporal remotely sensed data from LANDSAT-TM and ASTER acquired in 1991, 2003, 2005, 2006 and 2010 were analyzed using a density slice method that was applied on a near infrared band to estimate the rate of accretion and erosion in the pre- and the post-2004 Indian Ocean tsunami. The analysis was focused on the most severely impacted area by the tsunami, Phang Nga Province, located in southern Thailand. The changes in the shoreline positions of 2003 compared with those of 1991 presented a rate of erosion of 20-154 m, and the most severe erosion occurred at Pramong Village in Takua Pa District. In 2003, the accretion was significant at Nok Na Village on Kho Khao Island (185 m) compared to its shoreline position in 1991. Two months after the catastrophic tsunami, in 2005, the rate of erosion was 18-652 m, and locations with erosion rate of more than 300 m were identified as severely damaged. In 2006, most of the eroded shoreline positions had recovered more than 50-70% to the pre-tsunami positions, especially the positions with a rate of erosion of less than 100 m. While most of the shoreline positions with a rate of erosion of more than 300 m in 2005 had more than 60% recovered in 2010 and were resumed closer to the positions in the pre-tsunami (2003). In 2010, six years after the disaster, significant variation in the shoreline was observed in Takua Pa District, the shorelines at Nok Na, Pak Ko, and Nam Khem villages had changed their alignments. In the absence of a wave breaker and a management plan at these locations, the continuous erosion that changes the shoreline positions might also affect the political boundary of Phang Nga