Comparison of vegetation indices for estimating above-ground mangrove carbon stocks using PlanetScope image

doi:10.1016/j.rsma.2021.101730

Regional Studies in Marine Science

Volume 44, May 2021, 101730

https://doi.org/10.1016/j.rsma.2021.101730 Get rights and content

Abstract

Mangroves play a pivotal role in providing ecological benefits and services to reduce and adapt to climate change impact on coastal ecosystem. They are capable of absorbing carbon, which is crucial in controlling CO₂ levels in the atmosphere. This research aims to assess the accuracy of selected vegetation indices for estimating above-ground carbon (AGC) stocks of mangroves using PlanetScope images in Bedul, Banyuwangi, East Java Province, Indonesia. A semi-empirical approach was used to assess and map mangrove AGC, starting with applying the allometric equation to calculate field-measured species-specific AGC stocks. Regression analyses were applied to develop a relationship between field AGC and vegetation indices derived from PlanetScope Image, including Normalized Difference Vegetation Index (NDVI), Difference Vegetation Index (DVI), and Enhanced Vegetation Index (EVI). The Standard Errors of Estimates (SE) were 31.41, 32.93, and 31.63 tons/ha for DVI, EVI, and NDVI, respectively. Thus, carbon stocks estimation, including DVI as an independent variable, is considered more accurate than other vegetation indices tested in this research

Introduction

Global warming is one of the most feared problems, and carbon emissions are considered its strongest influencing factors (Bindu et al., 2018). Global climate change has been attributed to the loss of forest area due to advances in crop and livestock agriculture, forest fires, and resource overuse. Coupled with the release of greenhouse gases into the atmosphere by industrial activities and transportation development, it creates a daunting challenge for humans to survive (Smith et al., 2017, Grant et al., 2017, Pant et al., 2018). To overcome this circumstance, many have suggested enhancing the role of forests as CO₂ absorbers by devising natural and planted forest management systems (Brown, 1997). Coastal ecosystems such as mangroves and seagrass play a crucial part in providing benefits and services related to climate change impact reduction and adaptation (Howard et al., 2014). Mangroves are a group of vegetation species that can absorb carbon, an element controlling CO₂ concentrations in the atmosphere (Sidik et al., 2019). Mangrove forests in Indonesia have the most extensive area worldwide (Kusmana, 1996), making it tremendous potential for carbon absorption. The global mangrove area is about 16,530,000 hectares, with 42% of it found in Asia, 20% in Africa, and 15% in North and Central America (Giri et al., 2011).

Considering that mangrove forests often grow in hardly accessible terrain, it is necessary that constraints related to time, cost, and energy in mangrove studies to be minimized. Remote sensing satellites are now commonly used for estimating extension and dynamics, like detecting changes in mangrove areas (Sulaiman et al., 2013). In 2016, the latest remote sensing data product was released, namely PlanetScope. This high-resolution satellite imagery offers several advantages in data acquisition, including the ability to sense and record 200 million km² of the ground surface area daily with 130 complete satellite constellations. According to Planet Labs (2020), each PlanetScope satellite is 3U in size (CubeSat form factor $=$ 10 cm × 10 cm × 30 cm). It has a 98° inclination angle, with an orbital altitude of 475 km in a sun-synchronous orbit, allowing object sensing in the morning and during the day and, especially along the equator, at 09:30 $-$ 11:30 am. PlanetScope has a 12-bit radiometric resolution carrying four spectral bands (blue, green, red, and near-infrared/NIR), a 3 m spatial resolution, and a high temporal resolution (one day) at nadir (Planet Labs, 2020).

Mangrove ecosystems are one of the objects detectable in remote sensing data because they are geographically situated at the interface between land and sea, resulting in unique sensing products compared to other vegetation objects (Faizal and Amran, 2005). Because remote sensing acquires information quickly and with a wide area of coverage, it can provide data required in mangrove carbon stock estimation (Hastuti et al., 2017). This method generally uses image transformation like vegetation indices to measure vegetation density levels, which are related to biomass, by applying a specific algorithm to a satellite image. The vegetation index transforms the spectral values of several remote sensing bands to accentuate those of vegetation objects. In practical terms, it is a mathematical transformation that involves several bands at once and produces a new, more representative image of vegetation objects (Danoedoro, 2012).

In this research, three vegetation indices (VIs) were selected: Normalized Difference Vegetation Index (NDVI), Difference Vegetation Index (DVI), and Enhanced Vegetation Index (EVI). These VIs were selected after factoring in numerous literature studies and research reporting their strengths and weaknesses in different circumstances (Wicaksono et al., 2011, Wicaksono, 2017, Kamal et al., 2016, Nguyen et al., 2019, Xia et al., 2020, Zhu et al., 2020, Aljahdali et al., 2021). Thus, it was intended to assess their accuracy in estimating above-ground carbon stocks from PlanetScope images in the Bedul Mangrove Area, Banyuwangi, East Java Province, Indonesia. The research differs from previous studies in remote sensing data used and area observed. Accordingly, it was also designed to investigate whether the three vegetation indices produced accurate estimates with such differences.

Section snippets

Research site

This research was conducted in the Bedul Mangrove Block, a mangrove forest part of the Alas Purwo National Park in Bloksolo Sub-village, Sumberasri Village, Purwoharjo District, Banyuwangi Regency. According to the minutes of the areal measurement on May 27, 1983, the national park covers 43,420 hectares of land. Lowland tropical rainforest ecosystem composes most of the park, while the mangrove forest area, situated at the Grajagan Bay, is currently about 1200 hectares. The Grajagan Bay is

Identified mangrove species characteristics

Table 2 presents the list of mangrove species found at several locations during the fieldwork. Data were collected from only 20 of the planned 40 points due to the mangrove forest’s harsh condition, making it difficult to access all sample sites. These 20 samples data were split up equally into two groups: ten samples were used for modeling carbon stocks, while the other ten were for accuracy testing. To obtain an optimum prediction from limited samples, we selected these sampling sets

Conclusion

There are 14 mangrove species found along the Segara Anak River in the Alas Purwo National Park. Among those species, Rhizophora sp. were found to be dominant in this area and contributed to the high values of AGC. Overall, PlanetScope image has a potential for mangrove AGC estimation and mapping. All the three vegetation indices resulted almost similar pattern of estimated AGC value distribution. However, based on the accuracy testing results, DVI has the lowest standard error of estimate

CRediT authorship contribution statement

Eva Purnamasari: Conceptualization, Methodology, Formal analysis, Writing - original draft. Muhammad Kamal: Validation, Interpreting, Writing - review & editing. Pramaditya Wicaksono: Methodology, Data curation, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was funded by the 2020 Penelitian Dasar (Fundamental Research) Grant scheme provided by the Ministry of Research and Technology/ National Research and Innovation Agency of the Republic of Indonesia (contract number 2979/UN1.DITLIT/DIT-LIT/PT/2020). The authors would like to thank (1) the Department of Geographic Information Science at the Faculty of Geography, Universitas Gadjah Mada, for providing research facilities and equipment, (2) the Alas Purwo National Park management for

References (51)

BaratiS. et al.
Comparison the accuracies of different spectral indices for estimation of vegetation cover fraction in sparse vegetated areas
Egypt J. Remote Sens. Space Sci.
(2011)
BhandariA.K. et al.
Feature extraction using normalized difference vegetation index (NDVI): A case study of Jabalpur city
Procedia Tech.
(2012)
HueteA. et al.
Overview of the radiometric and biophysical performance of the MODIS vegetation indices
Remote Sens. Environ.
(2002)
KamalM. et al.
Assessment of multi-resolution image data for mangrove leaf area index mapping
Remote Sens. Environ.
(2016)
KomiyamaA. et al.
Allometry, biomass and productivity of mangrove forest: A review
Aquat. Bot.
(2008)
TuckerC.J.
A spectral method for determining the percentage of green herbage material in clipped sample
Remote Sens. Environ.
(1980)
XiaQ. et al.
Evaluation of submerged mangrove recognition index using multi-tidal remote sensing data
Ecol. Indic.
(2020)
AlexanderD.L.J. et al.
Beware of R(2): Simple, unambiguous assessment of the prediction accuracy of QSAR and QSPR models
J. Chem. Inf. Model
(2015)
AljahdaliM.O. et al.
Monitoring mangrove forest degradation and regeneration: Landsat time series analysis of moisture and vegetation indices at Rabigh Lagoon, Red Sea
Forests
(2021)
AnandA. et al.
Use of hyperion for mangrove forest carbon stock assessment in Bhitarkanika Forest Reserve: A contribution towards blue carbon initiative
Remote Sens.
(2020)

AshtonE.C. et al.

Breakdown of mangrove leaf litter in a managed mangrove forest in peninsular Malaysia

Hydrobiologia

(1999)

BinduG. et al.

Carbon stock assessment of mangroves using remote sensing and geographic information system

Egypt J. Remote Sens. Space Sci.

(2018)

BrownS.

Estimating Biomass and Biomass Change of Tropical Forest: A PrimerFAO Forestry Paper 134

(1997)

CandraE.D. et al.

Above ground carbon stock estimates of mangrove forest using worldview-2 imagery in Teluk Benoa, Bali

IOP Conf. Ser. Earth Environ. Sci.

(2016)

Cerón-SouzaI. et al.

Hybridization and introgression in new world red mangroves, Rhizophora (Rhizophoraceae)

Am. J. Bot.

(2010)

DanoedoroP.

Pengantar Penginderaan Jauh Digital

(2012)

DharmawanI.W.S. et al.

Karbon Tanah dan Pendugaan Karbon Tegakan Avicennia marina (Forsk.) Vierh. di Ciasem, Purwakarta

J. Penelitian Hutan Konservasi Alam

(2008)

FaizalA. et al.

Effective vegetation index transformation model for the prediction of mangrove density Rhizophora mucronata. PIT MAPIN XIV ITS

(2005)

FranandaH. et al.

Komparasi Indeks Vegetasi Untuk Estimasi Stok Karbon Hutan Mangrove Kawasan Segoro Anak Pada Kawasan Taman Nasional Alas Purwo Banyuwangi, Jawa Timur

Majalah Ilmiah Globe

(2015)

GiriC. et al.

Status and distribution of mangrove forests of the world using earth observation satellite data

Global Ecol. Biogeogr.

(2011)

GoessensA. et al.

Is Matang Mangrove Forest in Malaysia sustainably rejuvenating after more than a century of conservation and harvesting management?

PLoS One

(2014)

GrantD.B. et al.

Sustainable Logistics and Supply Chain Management: Principals and Practices for Sustainable Operations and Management

(2017)

HairiahK. et al.

Pengukuran Karbon Tersimpan di Berbagai Macam Penggunaan Lahan

(2007)

HamdanO. et al.

Mangrove carbon stock assessment by optical satellite imagery

J. Trop. For. Sci.

(2013)

HanafiN. et al.

Pendugaan Cadangan Karbon Pada Sistem Penggunaan Lahan di Areal PT. Sikatan Wana Raya

Media SainS

(2012)

Cited by (19)

Mapping tree carbon density using sentinel 2A sensor on Google Earth Engine in Darjeeling Himalayas: Implication for tree carbon management and climate change mitigation
2024, Physics and Chemistry of the Earth
The Himalayan region is a most fragile ecosystem globally. Trees make up around 90 % of the global biomass carbon pool and previous studies have shown that tree carbon balance cannot be easily assessed by conventional methods. Considering trees as a backbone of the forest ecosystem, present study assessed the heterogeneity in tree carbon density using field-inventoried data and NDVI-based modelling with Sentinel 2 A imagery on Google Earth Engine. The specific aim of the study was to assess the spatial distribution of tree carbon density in the Darjeeling Himalayas using Sentinel 2 A. The object-based classification of forest area using a random forest algorithm showed a high accuracy (Kappa coefficient value of 0.92, OOB error 0.17). The regression model using NDVI as a predictor of tree carbon demonstrated a good fit (R² = 0.78) for predicting tree carbon density. Validation results show high accuracy of the regression model in predicting tree carbon density with a low RMSE of 9.39 Mg ha⁻¹ (R² = 0.80, % RMSE = 11.55 %). The classification of tree carbon density into five classes revealed that a significant proportion of the forest area (57.05 %) falls under moderate carbon density (50–75 Mg ha⁻¹). In Darjeeling Himalayas, majority of forests are under the carbon density between 50 and 75 Mg ha⁻¹. Improvement and conservation efforts must be directed for very low carbon density (01–25 Mg ha⁻¹) areas covering 0.05 %, and high carbon density (75–100 Mg ha⁻¹) covering 36.22 % of the forest area, respectively, to balance the overall carbon storage potential of the region.
Comparison of multi-source remote sensing data for estimating and mapping above-ground biomass in the West Usambara tropical montane forests
2023, Scientific African
Above-ground biomass (AGB) estimation is important to better understand the carbon cycle and improve the efficiency of forest policy and management activities. AGB estimation models, using a combination of field data and remote sensing data, can largely replace traditional survey methods for measuring AGB. There are, however, critical steps for mapping AGB based on satellite data with an acceptable degree of accuracy, such as choice of remote sensing data, the proper statistical modelling method, and remote sensing predictor variables, at known field locations. This study sought to identify the optimal optical and synthetic aperture radar (SAR) remote sensing imagery from five sensors (PlanetScope, Sentinel-2, Landsat 8 OLI, ALOS-2/PALSAR-2, and Sentinel-1) to model 159 field-based AGB values from two montane forests under semiparametric (Generalized Additive Model; GAM) and non-parametric (eXtreme Gradient Boosting; XGB) approaches using information from four groups of predictor variables (spectral bands/polarizations, vegetation indices, textures, and a combination of all). The study's results showed that PlanetScope (rRMSE = 69.19%; R² = 0.161) was the most precise optical sensor while ALOS-2/PALSAR-2 (rRMSE = 70.76; R² = 0.165) was the most precise amongst the SAR sensors. XGB models generally resulted in those with lower prediction errors as compared to GAMs for the five sensors. Models having textures of vegetation indices and polarization bands achieved greater accuracy than models that incorporated spectral bands/polarizations and vegetation indices only. The study recommends that PlanetScope and ALOS-2/PALSAR-2 remote sensing data using the XGB-based technique is an appropriate approach for accurate local and regional estimation of tropical forest AGB particularly for complex montane forest ecosystems.
Mangrove distribution and afforestation potential in the Red Sea
2022, Science of the Total Environment
Citation Excerpt :
Hence, to further improve the identification of the smallest mangrove stands, future studies might exploit sensors with even greater spatial resolutions, such as PlanetScope CubeSats (3–4 m pixels), SkySat (0.50 m pixels) or Maxar image data (0.15 m pixels) (McCabe et al., 2017). PlanetScope imagery has already yielded good results for mapping mangroves in local areas (Purnamasari et al., 2021) and might provide new insights on Red Sea mangrove population dynamics in the future. Considering the increase in the extent of mangroves since 2013 (27 % and 5.6 % using S2 and L8 data, respectively, in less than a decade), additional research is required to verify the drivers behind these changes.
Mangrove ecosystems represent one of the most effective natural environments for fixing and storing carbon (C). Mangroves also offer significant co-benefits, serving as nurseries for marine species, providing nutrients and food to support marine ecosystems, and stabilizing coastlines from erosion and extreme events. Given these considerations, mangrove afforestation and associated C sequestration has gained considerable attention as a nature-based solution to climate adaptation (e.g., protect against more frequent storm surges) and mitigation (e.g. offsetting other C-producing activities). To advance our understanding and description of these important ecosystems, we leverage Landsat-8 and Sentinel-2 satellite data to provide a current assessment of mangrove extent within the Red Sea region and also explore the effect of spatial resolution on mapping accuracy. We establish that Sentinel-2 provides a more precise spatial record of extent and subsequently use these data together with a maximum entropy (MaxEnt) modeling approach to: i) map the distribution of Red Sea mangrove systems, and ii) identify potential areas for future afforestation. From these current and potential mangrove distribution maps, we then estimate the carbon sequestration rate for the Red Sea (as well as for each bordering country) using a meta-analysis of sequestration values surveyed from the available literature. For the mangrove classification, we obtained mapping accuracies of 98 %, with a total Red Sea mangrove extent estimated at approximately 175 km². Based on the MaxEnt approach, which used soil physical and environmental variables to identify the key factors limiting mangrove growth and distribution, an area of nearly 410 km² was identified for potential mangrove afforestation expansion. The factors constraining the potential distribution of mangroves were related to soil physical properties, likely reflecting the low sediment load and limited nutrient input of the Red Sea. The current rate of carbon sequestration was calculated as 1034.09 ± 180.53 Mg C yr⁻¹, and the potential sequestration rate as 2424.49 ± 423.26 Mg C yr⁻¹. While our results confirm the maintenance of a positive trend in mangrove growth over the last few decades, they also provide the upper bounds on above ground carbon sequestration potential for the Red Sea mangroves.
Estimation of aboveground mangrove carbon stocks from WorldView-2 imagery based on generic and species-specific allometric equations
2022, Remote Sensing Applications: Society and Environment
Citation Excerpt :
Moreover, Wicaksono (2017) used several multispectral and hyperspectral images, including WorldView-2, ALOS AVNIR-2, ASTER, Landsat 8 OLI, and Hyperion hyperspectral images, resulting in a standard error of estimation of 11.39–14.23 kg C/m2. Furthermore, Purnamasari et al. (2021) estimated the mangrove AGC by applying the DVI, NDVI, and EVI indices to PlanetScope images. In this case, DVI was the best vegetation index for estimating mangrove AGC, with a standard error of 31.41 ton C/ha.
The aboveground mangrove carbon stock can be estimated and mapped from remote-sensing data using allometric equations. However, the distinct effects of applying the generic or species-specific allometric equations remain unclear in the mangrove environment. The aims of this study are to (1) describe the relationship between field biomass and image pixels using generic and species-specific allometric equations, (2) estimate and map the mangrove carbon stock at the generic and species-specific levels, and (3) compare the accuracy and distribution of the resulting mangrove carbon stock maps. We tested the performance of these allometric equations using a resampled WorldView-2 image (6 m pixel size). The study site was in Clungup Mangrove Conservation (CMC), Malang, East Java and included natural mangrove forests and well-planted mangroves. We used the simple ratio (SR) vegetation index to estimate and map the aboveground carbon stock (AGC) of the mangroves based on the WorldView-2 image. Overall, the results indicate that the use of a generic allometric equation estimates mangrove AGC more accurately than the species-specific equations. However, we found a low coefficient of determination (R²) for both models: 0.2175 for the generic model and 0.1801 for the species-specific one. This may be due to the mangrove objects’ high variation in internal spectral reflectance at the tested image pixel size. The total estimated mangrove AGC obtained was 3,834.15 tons with a standard error of 0.12 for the generic model and 4,927.90 tons with a standard error of 0.17 for the species-specific model. Our findings substantiate the importance of the selection of allometric equations for mangrove AGC studies based on remote-sensing images.
Factors involved in the degradation of mangrove forests in Iran: A mixed study for the management of this ecosystem
2022, Journal for Nature Conservation
In Iran, mangroves are located in the south of the country, where they fulfill essential ecosystem functions, but some parts of them have been destroyed for various reasons in recent years so that it seems necessary to have a management plan consistent with the conservation of this ecosystem. But, it needs knowledge of the trend of and reasons for the degradation of these forests. So, the primary idea of this research was to discover the factors involved in the degradation of these forests and how to manage them. So, the research first used the remote sensing technique to determine the rate of their degradation over the last 20 years. Then, using the Delphi technique, 45 factors were identified for the degradation of these forests and they were classified into five categories including climatic and environmental, anthropogenic, socioeconomic, psychological, and policymaking and legal. Finally, the analytic hierarchy process (AHP) was used to rank these factors within two management paradigms of remediability and preventability so that the ecosystem can be soundly managed by classifying these factors. Overall, the results can provide policymakers with new insights into forest management and the policies presented here can contribute to the sustainability of this ecosystem.
Mangrove mapping and monitoring using remote sensing techniques towards climate change resilience
2024, Scientific Reports

View all citing articles on Scopus

View full text

Comparison of vegetation indices for estimating above-ground mangrove carbon stocks using PlanetScope image

Abstract

Introduction

Section snippets

Research site

Identified mangrove species characteristics

Conclusion

CRediT authorship contribution statement

Declaration of Competing Interest

Acknowledgments

Egypt J. Remote Sens. Space Sci.

Procedia Tech.

Remote Sens. Environ.

Remote Sens. Environ.

Aquat. Bot.

Remote Sens. Environ.

Ecol. Indic.

Beware of R(2): Simple, unambiguous assessment of the prediction accuracy of QSAR and QSPR models

J. Chem. Inf. Model

Monitoring mangrove forest degradation and regeneration: Landsat time series analysis of moisture and vegetation indices at Rabigh Lagoon, Red Sea

Forests

Use of hyperion for mangrove forest carbon stock assessment in Bhitarkanika Forest Reserve: A contribution towards blue carbon initiative

Remote Sens.

Breakdown of mangrove leaf litter in a managed mangrove forest in peninsular Malaysia

Hydrobiologia

Carbon stock assessment of mangroves using remote sensing and geographic information system

Egypt J. Remote Sens. Space Sci.

Estimating Biomass and Biomass Change of Tropical Forest: A PrimerFAO Forestry Paper 134

Above ground carbon stock estimates of mangrove forest using worldview-2 imagery in Teluk Benoa, Bali

IOP Conf. Ser. Earth Environ. Sci.

Hybridization and introgression in new world red mangroves, Rhizophora (Rhizophoraceae)

Am. J. Bot.

Pengantar Penginderaan Jauh Digital

Karbon Tanah dan Pendugaan Karbon Tegakan Avicennia marina (Forsk.) Vierh. di Ciasem, Purwakarta

J. Penelitian Hutan Konservasi Alam

Effective vegetation index transformation model for the prediction of mangrove density Rhizophora mucronata. PIT MAPIN XIV ITS

Komparasi Indeks Vegetasi Untuk Estimasi Stok Karbon Hutan Mangrove Kawasan Segoro Anak Pada Kawasan Taman Nasional Alas Purwo Banyuwangi, Jawa Timur

Majalah Ilmiah Globe

Status and distribution of mangrove forests of the world using earth observation satellite data

Global Ecol. Biogeogr.

Is Matang Mangrove Forest in Malaysia sustainably rejuvenating after more than a century of conservation and harvesting management?

PLoS One

Sustainable Logistics and Supply Chain Management: Principals and Practices for Sustainable Operations and Management

Pengukuran Karbon Tersimpan di Berbagai Macam Penggunaan Lahan

Mangrove carbon stock assessment by optical satellite imagery

J. Trop. For. Sci.

Pendugaan Cadangan Karbon Pada Sistem Penggunaan Lahan di Areal PT. Sikatan Wana Raya

Media SainS