Translator Disclaimer
22 November 2019 Development of aboveground mangrove forests’ biomass dataset for Southeast Asia based on ALOS-PALSAR 25-m mosaic
Author Affiliations +
Abstract

Southeast Asia (SEA) has the largest mangrove forest area in the world, which plays an important role in the global carbon cycle and is helping to mitigate climate change. In order to manage the mangrove forests in SEA, their total biomass needs to be determined. However, development of a biomass dataset based on field survey is time consuming. An aboveground biomass (AGB) dataset of mangrove forests was developed for SEA based on ALOS PALSAR 25-m mosaic. Specifically, ALOS-PALSAR 25-m images were first retrieved for SEA from the Kyoto and Carbon Initiative projects and then converted from a digital number to a normalized radar cross-section format in decibels. Samples of mangrove forests in SEA were collected as regions of interest from ALOS PALSAR data based on visual interpretation using Landsat data and Google Earth imagery. A rule-based classification method based on mangrove backscattering characteristics was then used to classify mangroves and nonmangroves in the region. Subsequently, an empirical model was adopted to estimate the AGB of the mangrove forests and an AGB dataset was developed. The results indicate that the spatial distribution of mangrove forests over SEA is 5.1 million hectares, and the estimated average AGB is 140.5  ±  136.1  Mg  /  ha.

1.

Introduction

Mangroves are a unique and complex major component of coastal zones in the tropics and subtropics. They represent the transitional ecosystem at the confluence of sea, land, and fresh water. The main components of their vegetation are generally evergreen trees or shrubs that grow along the coastlines, estuaries, brackish water, or deltas.1 Mangroves provide important products and various ecosystem services.2 They not only play an important role in the sustainability of coastal ecosystems but also provide important socioeconomic and cultural benefits for coastal communities.1 They are also known as the most intense coastal carbon absorbers in the world and play a central and growing role in the global carbon cycle.3 According to Donato et al.,4 the amount of carbon storage per unit area of mangrove forests is five times larger than that estimated for temperate, boreal, and tropical terrestrial forests. The high growth rates of trees and plants, anaerobic soils, and stagnant water that can retard decomposition of biomass enable mangrove forests to store a large amount of carbon over the long term.5

According to Alongi and Dixon,6 mangroves are very productive ecosystems that have primary production levels equivalent to moist tropical forests and coral reefs. Mangroves only occupy 0.5% of the global coastal area, but they contribute 10% to 15% (24  TgCy1) of coastal sediment carbon storage and export 10% to 11% of particulate terrestrial carbon to the sea. Currently, their contribution to carbon sequestration is one of the important ways for conservation and restoration, and it further helps to reduce the greenhouse effect. This shows that mangroves play an important role in global climate change mitigation. In order to acquire and build strong awareness of global carbon conditions and the impacts of diminishing mangrove forests on climate change, it is very important to assess and quantify the spatial distribution of mangrove forests and carbon stocks, as well as the emission factors of greenhouse gases emitted from the major land use activities in mangrove forests. However, most countries, especially in Southeast Asia (SEA), do not have enough information about the total quantity of their mangrove biomass.5,7

From a global perspective, SEA has the largest mangrove area in the world estimated at more than 6.8 million hectares or around 34% to 42% of the total area of mangrove forests in the world.2 Mangrove forests in SEA are the most diverse in the world8 but have been suffering from human interference in destructive ways due to conversion of mangrove areas for agriculture, tourism, and housing.9 According to Giri et al.,10 the world’s mangrove forests are currently estimated to have been reduced by 50% from their previous state and, estimating by their current reduction rate, 30% to 40% of coastal wetlands and 100% of mangrove forests could be lost in the next 100 years if there is no improved management to conserve them. It is strongly recommended that mangrove biomass be mapped and monitored immediately. However, conducting field surveys for mangrove biomass and its productivity in the area has proved to be very difficult as it is expensive and time consuming by nature, caused by muddy soil conditions, the weight of wood,11,12 the vast area to cover, and tidal influences.

Remote sensing has been widely proven to be a fast, cost-effective, and efficient method for monitoring and mapping the highly spatial distribution and biomass of mangrove forests as well as their disturbance on regional and global scales.12,13 Optical imagery and synthetic aperture radar (SAR) data have the potential to predict the biomass of tropical forests.14 However, according to Häme et al.,14 the depth and thickness of the irregular atmospheric conditions in the area of interest make atmospheric correction of optical images difficult. On the other hand, radiometric correction of the topography of SAR data will require a very accurate digital elevation model unless the study location is relatively flat. In particular, in tropical areas with prevalent cloud cover, SAR data are better than optical data for multitemporal acquisition.

According to Henderson and Lewis,15 although sensors in the optical range of the electromagnetic spectrum have received the greatest attention and have been widely used, considerable effort has also been invested in the use of radar sensors. Backscatter radar is very sensitive to the dielectric properties (soil moisture and vegetation) and attributes of the geometric conditions (surface roughness) of objects on the surface of Earth.16 In many regions of the world (e.g., areas covered by clouds and/or lacking light), radar is the only sensor that can provide consistent and periodic data in a reliable manner. A radar sensor can obtain information in the electromagnetic spectrum of the bands K, X, C, L, and P (different wavelengths) with polarizations of Horizontal to Horizontal (HH), Vertical to Vertical (VV), Horizontal to Vertical (HV), or Vertical to Horizontal (VH), which have varying ranges and azimuth resolutions. Each of these wavelengths has unique characteristics related to reflection from forest stands. The X band interacts only with leaves and canopy cover surfaces; hence, it is very suited for information on the surface layer of the tree canopy. The C band can penetrate through the leaves and spread to small branches and other underlying objects. The L band, which has a higher penetration capability, can penetrate the surface layer and spread to the stems and main branches.17 The P band has the greatest penetration capability and can penetrate into the canopy covers. Therefore, backscatters from the L and P bands are the parameters most associated with the biophysical parameters of trees and are predominantly used for studies related to forest biomass.18 Long-wavelength radar sensors such as L and P bands have both the advantage of being very sensitive to forest biomass and the potential for observation with high spatial and temporal resolution for estimating and monitoring biomass.1921

The launch of the Japanese Space Exploration Agency’s (JAXA) Advanced Land Observing Satellite (ALOS) Phased Arrayed L-band SAR (PALSAR) in 2006, therefore, represented a milestone in the global observation, characterization, mapping, and monitoring of mangroves over a larger area. ALOS PALSAR has been proven to have the potential to estimate the aboveground biomass (AGB) of mangrove forests.11,22,2326 Many researchers have tried to identify the baseline mangrove forest in regional-global areas.2729 However, references for estimating the biomass of mangrove forests in regional-global areas are lacking, which is both a challenge and an opportunity.30 This study was conducted with the objective of developing an aboveground mangrove forests biomass dataset for SEA based on the ALOS PALSAR 25-m mosaic.

2.

Methodology

The methodology used in this study consisted of several steps. First, the study area was circumscribed and primary and secondary data were collected. Primary data were taken from ALOS PALSAR, and secondary data were obtained from land cover maps, Landsat image data, and high-resolution image data obtained from Google Earth. Data processing involved conversion of digital numbers (DN) to normalization of radar cross sections (NRCS). Training samples of the mangrove forests in the study area were delineated, backscattering characterization was performed by calculating the mean and standard deviation, mangrove forests were classified based on backscattering characterization, and the AGB of mangrove forests in SEA was estimated based on an empirical model.

2.1.

Study Site

According to Giesen et al.,2 the mangrove forests in SEA spread from the Irrawaddy delta in northwest Myanmar throughout the coasts of Thailand, Cambodia, and Vietnam, and over the archipelagos of the Philippine and Indonesian Archipelago from Papua island to Sumatra island. They cover more than 17,000 islands and stretch more than 6000 km from the east to the west and about 3500 km from the north to the south of SEA. The most extensive mangrove areas in SEA are found in Indonesia, followed by Malaysia and Myanmar. In this study, we collected regions of interest (ROIs) of mangrove forests on 46 sites that spread over SEA as presented in Fig. 1 and Table 1.

Fig. 1

Study sites based on ALOS PALSAR composite and distribution of 46 sites for collecting ROIs of mangrove forest.

JARS_13_4_044519_f001.png

Table 1

ROIs on 46 sites of mangrove forest.

NoCountry/areaCoordinateROIs (pixels)
Myanmar
1.Lay Taung19°16′N 93°58′E8.027
2.Meinmahla Kyun-Irrawaddy delta15°51′N 64°56′E3.860
3.Tanintttharyi National Park11°53′N 98°42′E46.268
Thailand
4.Shi Phang Nga National Park08°58′N 98°22′E2.112
5.Phang Nga Ko Sriboya Krabi08°23′N 98°31′E6.339
6.Khlong Krabi07°54′N 99°04′E1.774
7.Suso Palian Trang07°11′N 99°39′E7.968
8.Puyu Mueng Satun06°40′N 99°56′E4.833
9.Pak Phanang Fangtawanok08°25′N 100°10′E5.217
Malaysia
10.Jebong Matang Perak04°49′N 100°36′E14.579
11.Pulau Kelang Selangor02°59′N 101°18′E7.923
12.Serkat Johor01°23′N 103°30′E3.303
13.Sarikei Sarawak02°07′N 111°23′E18.143
14.Pulau Selirong, Brunei04°50′N 115°07′E5.120
15.Kota Marudu Sabah06°33′N 116°46′E1.200
16.Beluran Sabah06°14′N 117°36′E11.049
Cambodia
17.Peam Krasaop Wildlife Sanctuary11°28′N 103°01′E3.166
18.Preak Piphot River10°40′N 103°52′E1.275
Vietnam
19.Thạnh An Cần Giò10°33′N 106°52′E13.967
20.Giao Thiện Giao Thuỷ Nam Định20°13′N 106°31′E390
21.Thái Thủy Thai Binh20°34′N 106°36′E730
22.Móng Cái Quảng Ninh21°27′N 107°55′E881
Philippines
23.Siargao island09°55′N 125°58′E1.108
24.Santa Margarita Samar18°21′N 121°35′E541
25.Abulug Cagayan18°21′N 121°35′E970
26.Dinas Zamboanga del Sur07°34′N 123°22′E716
Indonesia
27.Langsa Aceh04°31′N 98°01′E10.199
28.Bengkalis Riau02°01′N 101°36′E14.516
29.Indragiri Hilir Riau00°02′S 103°29′E40.262
30.Banyuasin South Sumatera01°58′S 104°32′E74.398
31.Pontianak West Kalimantan00°39′S 109°28′E36.615
32.Kotabaru South Kalimantan02°55′S 116°05′E25.237
33.Kutai Kartanegara East Kalimantan01°02′S 116°42′E16.728
34.Berau East Kalimantan02°04′N 117°39′E33.413
35.Nunukan East Kalimatan04°08′N 117°20′E33.699
36.Subang West Java06°13′S 107°45′E25
37.Cilacap Central Java07°41′S 108°54′E9.437
38.Badung Bali08°43′S 115°11′E1.275
39.Bombana Southeast Sulawesi04°31′S 122°5′E7.975
40.Muna Southeast Sulawesi04°44′S 123°4′E14.111
41.Sorong West Papua01°21′S 131°34′E38.666
42.Teluk Bintuni Papua02°9′S 133°36′E95.396
43.Teluk Bintuni Papua02°34′S 133°47′E118.867
44.Waropen Papua05°06′S 137°31′E256.657
45.Asmat Papua01°48′S 137°21′E91.473
46.Merauke Papua08°15′S 138°54′E34.947

2.2.

Data Collection

We used 25-m resolution PALSAR mosaic data generated by applying a high JAXA processing and analysis technique to images obtained from the Japanese L-band Synthetic Aperture Radars (PALSAR and PALSAR-2). Global 25-m resolution PALSAR/PALSAR-2 mosaic is a global SAR image made by SAR image mosaics of the backscattering coefficient measured by PALSAR/PALSAR-2. ALOS PALSAR data have been collected as part of the Kyoto and Carbon Initiative projects. As a result, we obtained a total of 60 ALOS PALSAR mosaic tiles for the SEA region. For secondary data to support the collection of training samples of mangrove areas, we used land cover maps from each country and Landsat data as well as high-resolution image data from Google Earth.

2.3.

Preprocessing

The preprocessing focused on converting the HH (Digital Number on Horizontal to Horizontal) and HV (Digital Number on Horizontal to Vertical) values to NRCS in decibels (dB) (i.e., σ° HH and σ° HV) using the following equations:31

Eq. (1)

σ°HH=10log10(DN2)CF,

Eq. (2)

σ°HV=10log10(DN2)CF,
where σ° is the backscattering coefficient and CF is the calibration factor. CF is equal to 83 for both HH and HV.

2.4.

Characterization

We collected mangrove samples data for the ROIs throughout the SEA region on ALOS PALSAR data based on visual interpretation using Landsat data and Google Earth images. As many as 46 ROIs of mangrove forests scattered in the SEA region were collected. The size of each ROI differed for each area of mangroves, as it depends on the size of the mangrove area in each region. After the ROI samples were obtained and the mean and standard deviation of the backscatter value for each sample were calculated. The equations for the mean and standard deviation are as follows:

Eq. (3)

μX=X¯=1ni=1nXi,

Eq. (4)

σ=1ni=1n(XiX¯)2,
where n is the number of samples taken, Xi is the value of the sample, and X¯ is the average of the samples.

2.5.

Classification

The classification of mangrove forests was performed based on the characteristics of the mangrove backscatter in each training sample. We used a rule-based classification method based on the mangrove backscattering characteristics to classify mangrove and nonmangrove areas in the region. The main parameters used in this rule-based classification were the mean and standard deviation of the backscatter value image of ALOS PALSAR. All pixels in the mangrove areas were identified using the following rule-based classification algorithm:12

Eq. (5)

If  μ_i-σ_ijband_iμ_i+σ_ij  then band_i=1,  otherwise band_i=0,
where μ and σ represent the mean and standard deviation of the grouped pixel samples (ROIs) from the sample data, respectively, i represents the band number of ALOS PALSAR, and j represents the positive real number of the selected standard deviation. The output was a binary image consisting of “mangrove” and “nonmangrove.”

2.6.

Estimation of Aboveground Biomass

The literature on estimation of the AGB of mangrove forests using PALSAR data is very small. Only a few studies11,2224 have established empirical relationships between L-band backscatter and the AGB of mangrove forests. The most significant difference between Takeuchi’s and Hamdan’s empirical relationships models is that Takeuchi’s empirical relationships model uses conversion through general allometric height-biomass relations. We refer to this model as the indirect model. The empirical model is expressed as follows:11

Eq. (6)

HH  (σ°)=3.6ln(tree height)23.7,

Eq. (7)

HV  (σ°)=4.4ln(tree height)24.9,

Eq. (8)

tree height=2.8ln(DBH)+0.4,

Eq. (9)

AGB=0.25DBH2.46,
where DBH is the diameter at breast height and AGB is the aboveground biomass. In contrast, Hamdan et al.23 directly related the backscatter value to the field biomass measurement, and thus we refer to it as the direct model. The empirical model is presented as follows:

Eq. (10)

HH(σ°)=0.472ln(AGB)12.326,

Eq. (11)

HV(σ°)=0.800ln(AGB)19.305.

3.

Results and Discussion

3.1.

Backscattering Characteristics of Mangrove Forests in Southeast Asia

We collected ALOS PALSAR data for the SEA region and determined the training samples as ROIs. Preprocessing was conducted by converting DN values to NRCS using Eqs. (1) and (2). In addition, the ROIs of 46 training samples were calculated to determine the backscattering characteristics of mangroves in the forms of mean and standard deviation values using Eqs. (3) and (4). The mean and standard deviations for each of the 46 ROIs can be seen in Fig. 2. Figure 2 shows the characterization of the backscattering coefficient of mangrove forests for the polarizations of HH and HV. The backscattering values of mangrove forests based on the HH polarization ranged from 10.88 to 6.65  dB, whereas the values based on the HV polarization were within the range 16.49 to 13.26  dB.32

Fig. 2

Means and standard deviations of backscattering values of mangrove forests based on ALOS PALSAR polarization on (a) HH and (b) HV.

JARS_13_4_044519_f002.png

Based on Fig. 2, each mangrove forest type has a wide range of backscattering values. The value depends on the physical and geometric conditions of the mangrove trees, as well as on environmental conditions such as weather dynamics, moisture, and the topography and altitude of tides.33 The backscatter value also depends on a specific backscatter signal (e.g., radar calibration and orthorectification),34,35 which is affected by the dielectric properties of the vegetation and ground surface.

3.2.

Spatial Distribution of Mangrove Forests Based on ALOS PALSAR

We classified mangrove forests based on their backscattering characteristics and topography data. The classification was enhanced by visual interpretation using Landsat images and Google Earth imagery as reference data. In this case, the classification method used rule-based classification as defined by Eq. (5). We classified mangrove forests and nonmangrove forests on an area by area basis. The parameters used in the rule-based classification were HH, HV, and topography data. To determine the threshold value, the mean and standard deviation of the backscattering values of HH and HV were used. The threshold values of the backscattering on HH and HV can be seen in Table 2.

Table 2

Threshold values of mangrove forests for HH and HV.

AreaThreshold values of backscattering ALOS PALSAR 25-m resolutions (in dB)
HHHV
Myanmar9.92<mangrove<5.0717.23<mangrove<11.72
Thailand12.29<mangrove<7.9218.14<mangrove<13.66
Malaysia11.76<mangrove<7.0617.20<mangrove<12.61
Cambodia10.28<mangrove<5.0916.64<mangrove<11.58
Vietnam10.50<mangrove<7.4219.03<mangrove<15.46
Philippines10.88<mangrove<6.5017.08<mangrove<13.02
Indonesia12.64<mangrove<5.3718.81<mangrove<11.47

Using rule-based classification, we determined the spatial distribution of the mangrove forests in the SEA region (Fig. 3). The area of the mangrove forests in the SEA region was calculated as 5,098,834  ha with an overall accuracy of 82% (Table 3). The latter was done based on a confusion matrix, which takes advantage of ground truth derived from some reference data such as land use and land cover maps and visual interpretation from Landsat and Google Earth. The estimated areas of mangrove forests in each country in comparison with other studies are presented in Table 4. According to the table, there are differences in the large areas of mangrove forests determined by this study and the other studies. These differences arise from the different methods employed to estimate the large areas of mangrove forests; the other studies estimated the large areas of mangrove forests using data derived from a statistical log book, which is an old data version.

Fig. 3

Spatial distribution of mangrove forests in Southeast Asia.

JARS_13_4_044519_f003.png

Table 3

Overall accuracy.

Reference data
MangroveNonmangroveTotalUser’s accuracy
Image classifiedMangrove1,197,076211,5501,408,62684.98%
Nonmangrove201,370685,374886,74477.29%
Total1,398,446896,9242,295,370
Producer’s accuracy85.60%76.41%
Overall accuracy = 82.01%

Table 4

Sizes of mangrove forest areas in comparison with other studies.

No.AreaMangrove area (ha)
This studyOther studies
1.Brunei587117,100 in 199236
2.Cambodia77,43272,835 in 1997,36 83,600 in 200236
3.Indonesia3,280,0313,493,110 in 198836
3,244,018 in 200937
4.Malaysia587,613587,269 in 199536
564,606 in 200338
5.Myanmar486,840452,492 in 199636
6.Philippines266,170127,610 in 199036
256.185 in 200039
7.Thailand233,334244,085 in 200036
228.158 ha in 200740
8.Vietnam159,901252,500 in 198336
157,000 in 200541

3.3.

Estimation of Aboveground Biomass of Mangrove Forests

We applied two models to estimate the AGB of mangrove forests: the indirect model [Eqs. (6)–(9)] and the direct model [Eqs. (10) and (11)]. According to Sun et al.,42 HV polarization of the RADAR data is the most sensitive for estimation of AGB. We estimated the average AGB of mangrove forests in the SEA region as 140.5±136.1  Mg/ha. Some plots of the spatial distribution of the AGB of mangrove forests are presented in Fig. 4. However, on the low and flat topography, the backscatter value of mangrove forest is affected by tidal height. The deviation of backscatter value on the low and flat topography is around 1.6 dB for HV.33 This study cannot identify information of tidal height because data are given from global 25-m resolution PALSAR and PALSAR-2 mosaics, so in this case, the AGB of mangrove forest that is estimated can be low or high.

Fig. 4

Spatial distribution of the AGB of mangrove forests in the SEA region; (a) Suso Palian region, Thailand, (b) Jebong Perak region, Malaysia, (c) PeamKrasaop Wildlife Sanctuary, Cambodia, and (d) ThanhAn Can Gio, Vietnam.

JARS_13_4_044519_f004.png

Since the lack of ground measurement of the ABG of the mangrove forest on overall SEA still needs further research to validate our results, we have compared our results to those obtained by other studies (Table 5 and Fig. 5). We observed that the results based on the direct model are closer to those estimated by other studies than the indirect model. However, the direct model has a coefficient of determination (R2) of 0.427 with a residual error of 61.32  Mg/ha.

Table 5

Comparison of AGB estimation of mangrove forests using the indirect and direct models with other studies.

No.LocationMajor speciesIndirect model (Mg/ha)Direct model (Mg/ha)Other studies (Mg/ha)
1.Perak, MalaysiaRhizophora apiculata250.5101.7299.423
2.Aceh, Indonesia96.546.811.6843
3.Chumpon, ThailandRhizophora apiculata807.4231.42166
4.Pang-nga, ThailandMix494.1166.310844
5.Kuala Selangor, MalaysiaBruguiera parviflora411.33146.3144.4745
6.MuiCamau, VietnamAvicennia alba
Rhizophora apiculata
167.274.290.2+15.846
7.West Kalimantan, IndonesiaAvicennia marina
Rhizophora apiculata
Rhizophora stylosa
Rhizophora alba
Rhizophora mucronata
Bruguiera gymnorrhiza
Xylocarpus granatum
467.1159.9159.1+69.55
8.Papua, IndonesiaAvicennia marina
Rhizophora apiculata
Rhizophora stylosa
Rhizophora alba
Rhizophora mucronata
Bruguiera gymnorrhiza
Xylocarpus granatum
924.3252.56213.8+129.85
9.The PhilippinesAvicennia officinalis1.342319.3297.247

Fig. 5

Comparison of AGB derived from backscattering of HV on ALOS PALSAR.

JARS_13_4_044519_f005.png

According to Suzuki et al.,48 one limitation of SAR in estimating AGB is that the backscatter intensity becomes saturated when the AGB volume exceeds a certain critical value. For example, estimation of the AGB of the mangrove forests in Matang Malaysia, where the AGB<100  Mg/ha, gives the highest coefficient of determination (R2) and the smallest root mean square error (RMSE), whereas an AGB>100  Mg/ha has a lower R2 and a higher RMSE.23 In addition, the saturation level of AGB for Sonneratia caseolaris and Kandelia obovata is above 100  Mg/ha.23 According to Ghasemi et al.,49 the saturation problem can be solved by applying an interferometry technique. However, this method has not been widely tested; thus, it is still not clear whether it can be used for AGB estimation of all types of mangrove forests.

We also have compared our result with Landsat data, where the actual optical images of the Landsat data with textural and spectral characteristics of the canopy and leaves are the main features used to distinguish among mangrove communities.35,50 According to Fig. 6 the results of the classification of mangroves and nonmangroves on ALOS PALSAR imagery are almost the same as those produced by Landsat imagery. However, in this study, not only can be classified mangrove and non-mangrove, but also we have estimated above ground biomass of mangrove forest.

Fig. 6

Comparison of AGB estimated based on ALOS PALSAR and Landsat imagery data in (a) Suso Palian region, Thailand, (b) Jebong Perak region, Malaysia, and (c) Peam Krasaop Wildlife Sanctuary, Cambodia.

JARS_13_4_044519_f006.png

4.

Conclusions

In this study, an AGB dataset of mangrove forests in SEA was developed. The spatial distribution of the mangrove forests in SEA was found to be 5.1 million ha with an overall accuracy of 82%. The estimated average AGB of the mangrove forests in SEA was found to be 140.5±136.1  Mg/ha. The direct model was also determined to be more accurate for estimation of the AGB of mangrove forests than the indirect model. However, saturation in estimating AGB needs to be reduced by integrating multisource data, including different wavelengths for optical data and microwaves. Thus, in future studies, development of more appropriate procedures and approaches is still required to reduce uncertainty and also the conducting and collecting of field survey measurements is required for validation.

Acknowledgments

The authors would like to thank the Ministry of Research, Technology, and Higher Education of the Republic of Indonesia (RISTEKDIKTI) for all their support. Also the authors would like to thank LPPM-ITENAS, JAXA-JAPAN, and Anggun Tridawati from Lampung University for their support.

References

1. 

M. N. Suratman, “Carbon sequestration potential of mangroves in Southeast Asia,” Managing Forest Ecosystems: The Challenge of Climate Change, 297 –315 Springer, New York (2008). Google Scholar

2. 

W. Giesen et al., Mangrove Guidebook for Southeast Asia, Wetlands International, Bangkok (2007). Google Scholar

3. 

C. Nellemann, Blue Carbon, A Rapid Response Assessment, 80 United Nations Environment Programme, Norway (2009). Google Scholar

4. 

D. C. Donato et al., “Mangroves among the most carbon-rich forests in the tropics,” Nat. Geosci., 4 (5), 293 –297 (2011). https://doi.org/10.1038/ngeo1123 Google Scholar

5. 

D. Murdiyarso et al., “The potential of Indonesian mangrove forests for global climate change mitigation,” Nat. Clim. Change, 5 1089 –1092 (2015). https://doi.org/10.1038/nclimate2734 Google Scholar

6. 

D. M. Alongi and P. Dixon, “Mangrove primary production and above and below-ground biomass in Sawi Bay, Southern Thailand,” Phuket Mar. Biol. Center Spec. Publ., 22 31 –38 (2000). Google Scholar

7. 

R. A. Houghton et al., “Importance of biomass in the global carbon cycle,” J. Geophys. Res. Biogeosci., 114 G00E3 (2009). Google Scholar

8. 

W. Giesen and S. Wulffraat, “Indonesian mangroves part I: plant diversity and vegetation,” Trop. Biodivers., 5 (2), 11 –23 (1998). Google Scholar

9. 

R. Tabuchi and S. B. Unyavejchewin, “Development of mangrove stands in Southeast Asia: with special reference to the west coast of the Malay Peninsula,” Global Environ. Res., 215 –221 (2013). Google Scholar

10. 

C. Giri et al., “Status and distribution of mangrove forests of the world using earth observation satellite data,” Glob. Ecol. Biogeogr., 20 (1), 154 –159 (2011). https://doi.org/10.1111/j.1466-8238.2010.00584.x Google Scholar

11. 

W. Takeuchi et al., “Above ground biomass mapping of mangrove forest in Vietnam by ALOS PALSAR,” in 3rd Int. Asia-Pac. Conf. Synth. Aperture Radar (APSAR), 1 –3 (2011). Google Scholar

12. 

A. Aslan et al., “Remote sensing of environment mapping spatial distribution and biomass of coastal wetland vegetation in Indonesian Papua by combining active and passive remotely sensed data,” Remote Sens. Environ., 183 65 –81 (2016). https://doi.org/10.1016/j.rse.2016.04.026 Google Scholar

13. 

A. Held et al., “High resolution mapping of tropical mangrove ecosystems using hyperspectral and radar remote sensing,” Int. J. Remote Sens., 24 (13), 2739 –2759 (2003). https://doi.org/10.1080/0143116031000066323 IJSEDK 0143-1161 Google Scholar

14. 

T. Häme et al., “Improved mapping of tropical forests with optical and SAR imagery, part II: above ground biomass estimation,” IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., 6 (1), 92 –101 (2013). Google Scholar

15. 

F. M. Henderson and A. J. Lewis, “Radar detection of wetland ecosystems: a review,” Int. J. Remote Sens., 29 (20), 5809 –5835 (2008). https://doi.org/10.1080/01431160801958405 IJSEDK 0143-1161 Google Scholar

16. 

N. M. Yusoff et al., “Phenology and classification of abandoned agricultural land based on ALOS-1 and 2 PALSAR multi-temporal measurements,” Int. J. Digital Earth, 10 (2), 155 –174 (2017). https://doi.org/10.1080/17538947.2016.1216615 Google Scholar

17. 

S. Darmawan et al., “An investigation of age and yield of fresh fruit bunches of oil palm based on ALOS PALSAR 2,” IOP Conf. Series: Earth Environ. Sci., 37 012037 (2016). https://doi.org/10.1088/1755-1315/37/1/012037 Google Scholar

18. 

S. Sinha et al., “A review of radar remote sensing for biomass estimation,” Int. J. Environ. Sci. Technol., 12 (5), 1779 –1792 (2015). https://doi.org/10.1007/s13762-015-0750-0 Google Scholar

19. 

E. T. A. Mitchard et al., “Using satellite radar backscatter to predict above-ground woody biomass: a consistent relationship across four different African landscapes,” Geophys. Res. Lett., 36 (23), 1 –6 (2009). https://doi.org/10.1029/2009GL040692 GPRLAJ 0094-8276 Google Scholar

20. 

T. Le Toan et al., “The BIOMASS mission: mapping global forest biomass to better understand the terrestrial carbon cycle,” Remote Sens. Environ., 115 (11), 2850 –2860 (2011). https://doi.org/10.1016/j.rse.2011.03.020 Google Scholar

21. 

Y. Yu and S. Saatchi, “Sensitivity of L-band SAR backscatter to aboveground biomass of global forests,” Remote Sens., 8 (6), 522 (2016). https://doi.org/10.3390/rs8060522 Google Scholar

22. 

A. C. Morel et al., “Estimating aboveground biomass in forest and oil palm plantation in Sabah, Malaysian Borneo using ALOS PALSAR data,” For. Ecol. Manage., 262 (9), 1786 –1798 (2011). https://doi.org/10.1016/j.foreco.2011.07.008 FECMDW 0378-1127 Google Scholar

23. 

O. Hamdan, H. Khali Aziz and I. MohdHasmadi, “L-band ALOS PALSAR for biomass estimation of Matang Mangroves, Malaysia,” Remote Sens. Environ., 155 69 –78 (2014). Google Scholar

24. 

P. Tien Dat, Y. Kunihiko and B. DieuTien, “Biomass estimation of Sonneratia caseolaris (l.) Engler at a coastal area of Hai Phong city (Vietnam) using ALOS-2 PALSAR imagery and GIS-based multi-layer perceptron neural networks,” GISci. Remote Sens., 54 329 –353 (2016). https://doi.org/10.1080/15481603.2016.1269869 Google Scholar

25. 

P. Tien Dat and Y. Kunihiko, “Aboveground biomass estimation of mangrove species using ALOS-2 PALSAR imagery in Hai Phong city, Vietnam,” J. Appl. Remote Sens., 11 (2), 026010 (2017). https://doi.org/10.1117/1.JRS.11.026010 Google Scholar

26. 

V. Sasan et al., “Improving accuracy estimation of forest aboveground biomass based on incorporation of ALOS-2 PALSAR-2 and sentinel-2A imagery and machine learning: a case study of the Hyrcanian Forest Area (Iran),” Remote Sens., 10 172 (2018). https://doi.org/10.3390/rs10020172 Google Scholar

27. 

R. Lucas et al., “Characterisation and monitoring of Mangroves using ALOS PALSAR dataEorc. Jaxa. Jp.,” http://www.eorc.jaxa.jp/ALOS/en/kyoto/phase_1/KC-Phase1-report_Lucas_WT.pdf Google Scholar

28. 

T. Nathan et al., “Distribution and drivers of global mangrove forest change, 1996±2010,” PLoS One, 12 (6), e0179302 (2017). https://doi.org/10.1371/journal.pone.0179302 POLNCL 1932-6203 Google Scholar

29. 

B. Pete et al., “The global mangrove watch: a new 2010 global baseline of mangrove extent,” Remote Sens., 10 1669 (2018). https://doi.org/10.3390/rs10101669 Google Scholar

30. 

P. Tien Dat et al., “Remote sensing approaches for monitoring mangrove species, structure, and biomass: opportunities and challenges,” Remote Sens., 11 230 (2019). https://doi.org/10.3390/rs11020230 Google Scholar

31. 

M. Shimada et al., “PALSAR radiometric and geometric calibration,” IEEE Trans. Geosci. Remote Sens., 47 (12), 3915 –3932 (2009). https://doi.org/10.1109/TGRS.2009.2023909 IGRSD2 0196-2892 Google Scholar

32. 

S. Darmawan et al., “Characterization and spatial distribution of mangrove forest types based on ALOS-PALSAR mosaic 25-m resolution in Southeast Asia,” IOP Conf. Ser.: Earth Environ. Sci., 37 012035 (2016). https://doi.org/10.1088/1755-1315/37/1/012035 Google Scholar

33. 

S. Darmawan et al., “Impact of topography and tidal height on ALOS PALSAR polarimetric measurements to estimate aboveground biomass of mangrove forest in Indonesia,” J. Sens., 2015 1 –13 (2015). https://doi.org/10.1155/2015/641798 Google Scholar

34. 

J. J. Van Zyl, “The effect of topography on radar scattering from vegetated areas,” Int. Geosci. Remote Sens. Symp., 2 (1), 1132 –1134 (1992). https://doi.org/10.1109/IGARSS.1992.578363 Google Scholar

35. 

C. Kuenzer et al., “Remote sensing of mangrove ecosystems: a review,” Remote Sens., 3 878 –928 (2013). https://doi.org/10.3390/rs3050878 Google Scholar

36. 

“Status and trends in mangrove area extent worldwide,” FAO, Rome, (2003). Google Scholar

37. 

C. Kusmana, “Distribution and current status of mangrove forests in Indonesia,” Mangrove Ecosystems of Asia: Status, Challenges and Management Strategies, 1 38 –59 Springer, New York (2013). Google Scholar

38. 

A. Latiff, I. Farida-Hanum, “Mangrove ecosystem of Malaysia: status, challenges and management strategies,” Mangrove Ecosystems of Asia; Status, Challenges and Management Strategies, 1 1 –18 Springer, New York (2013). Google Scholar

39. 

J. B. Long and C. Giri, “Mapping the Philippines’ mangrove forests using Landsat imagery,” Sensors, 11 (3), 2972 –2981 (2011). https://doi.org/10.3390/s110302972 SNSRES 0746-9462 Google Scholar

40. 

N. Pumijumnong, “Mangrove forests in Thailand,” Mangrove Ecosystems of Asia: Status, Challenges and Management Strategies, 62 –77 Springer, New York (2013). Google Scholar

41. 

R. Mcnally, A. McEwin and T. Holland, “The potential for mangrove carbon projects in Vietnam,” (2011). Google Scholar

42. 

G. Sun, K. J. Ranson and V. I. Kharuk, “Radiometric slope correction for forest biomass estimation from SAR data in the Western Sayani Mountains, Siberia,” Remote Sens. Environ., 79 (2–3), 279 –287 (2002). https://doi.org/10.1016/S0034-4257(01)00279-6 Google Scholar

43. 

D. M. Yuwono, Pemetaan Mangrove Sumatera, Provinsi Aceh, Sumatera Utara dan Riau, Badan Infromasi Geospasial (BIG), Bogor (2013). Google Scholar

44. 

P. Phongsuksawat, “Determination of tree biomass in the mangrove habitat study area, Changwat Phangna,” Kasetsart University, Thailand, (2002). Google Scholar

45. 

M. Hossain et al., “Net primary productivity of Bruguiera parviflora (Wight & Arn.) dominated mangrove forest at Kuala Selangor, Malaysia,” For. Ecol. Manage., 255 (1), 179 –182 (2008). https://doi.org/10.1016/j.foreco.2007.09.011 FECMDW 0378-1127 Google Scholar

46. 

N. T. Tue et al., “Carbon storage of a tropical mangrove forest in MuiCa Mau National Park, Vietnam,” Catena, 121 119 –126 (2014). https://doi.org/10.1016/j.catena.2014.05.008 CIJPD3 0341-8162 Google Scholar

47. 

A. C. Abino, J. A. A. Castillo and Y. J. Lee, “Assessment of species diversity, biomass and carbon sequestration potential of a natural mangrove stand in Samar, the Philippines,” For. Sci. Technol., 10 (1), 2 –8 (2014). https://doi.org/10.1080/21580103.2013.814593 Google Scholar

48. 

R. Suzuki, Y. Kim and R. Ishii, “Sensitivity of the backscatter intensity of ALOS/PALSAR to the above-ground biomass and other biophysical parameters of boreal forest in Alaska,” Polar Sci., 7 (2), 100 –112 (2013). https://doi.org/10.1016/j.polar.2013.03.001 Google Scholar

49. 

N. Ghasemi, M. R. Sahebi and A. Mohammadzadeh, “A review on biomass estimation methods using synthetic aperture radar,” Int. J. Geomatics Geosci., 1 (4), 776 –788 (2011). Google Scholar

50. 

G. Winarso et al., “Mangrove above ground biomass estimation using combination of Landsat 8 and ALOS PALSAR data,” Int. J. Remote Sens. Earth Sci., 12 (2), 85 –96 (2015). https://doi.org/10.30536/j.ijreses.2015.v12.a2687 Google Scholar

Biographies of the authors are not available.

© The Authors. Published by SPIE under a Creative Commons Attribution 4.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.
Soni Darmawan, Dewi K. Sari, Wataru Takeuchi, Ketut Wikantika, and Rika Hernawati "Development of aboveground mangrove forests’ biomass dataset for Southeast Asia based on ALOS-PALSAR 25-m mosaic," Journal of Applied Remote Sensing 13(4), 044519 (22 November 2019). https://doi.org/10.1117/1.JRS.13.044519
Received: 25 April 2019; Accepted: 1 November 2019; Published: 22 November 2019
JOURNAL ARTICLE
15 PAGES


SHARE
Advertisement
Advertisement
Back to Top