INTRODUCTION
Forests, which cover about 30% of the Earth's land area, play a vital role in sustaining life by providing essential services at various levels However, climate change poses significant threats to these ecosystems and the communities that rely on them Projections indicate that by 2100, global temperatures could increase by 1.8 to 4 °C, leading to a sea level rise of 0.75 to 1.5 meters, primarily due to the greenhouse effect This phenomenon is driven by multiple factors, including industrial emissions, fossil fuel usage, and deforestation, which is a major contributor to greenhouse gas emissions—accounting for nearly one-fifth of the total Deforestation and forest degradation are not only prevalent in developing tropical regions but have also led to severe forest fires in developed countries like the USA and Russia, resulting in the loss of thousands of hectares of forest and agricultural land.
Forests play a crucial role in environmental protection and provide essential resources like timber and non-timber products, as demonstrated in various countries, including Vietnam Their capacity for carbon absorption is vital in mitigating greenhouse gas emissions and combating global climate change The value of carbon sequestration by forests can be monetized through mechanisms such as Payment for Environmental Services (PES) and the Reducing Emissions from Deforestation and Forest Degradation (REDD & REDD+) initiatives Additionally, the Clean Development Mechanism (CDM) further supports these efforts by promoting sustainable development through carbon trading.
Article 12 of the Kyoto Protocol under the UNFCCC recognizes that CO2 absorbed by plantations is a key measure for reducing greenhouse gas emissions in developing countries While the qualitative research on this topic and its environmental benefits represent a significant initial step globally, it remains a novel concept in Vietnam.
Recent advancements in Geographical Information Systems (GIS) and Remote Sensing (RS) technologies have significantly enhanced natural resource management Improved satellite imagery allows for large-scale image analysis, enabling the collection of critical data on forest above-ground biomass (AGB) and vegetation structure RS is effectively used to monitor and map vegetation biomass and productivity by measuring spectral reflectance However, while optical RS does not directly measure AGB, it is influenced by factors such as crown size, tree density, and shadow, particularly in infrared bands Consequently, RS data is now regarded as a reliable method for estimating spatial biomass in tropical regions Numerous studies have applied RS technology for biomass assessment, demonstrating its capability to gather extensive forest information cost-effectively and accurately through repetitive data collection.
Cuc Phuong National Park's forests are considered a national treasure, offering vital resources like grazing land, wildlife habitats, water, and timber for local communities Unfortunately, human activities are significantly threatening the quality and quantity of these forests Research is underway to measure carbon levels in this critical ecosystem.
Assessing biomass in mountain forest ecosystems is essential for understanding the value of forests based on their carbon accumulation capacity These mountainous regions serve not only as hubs for academic exchange but also as vital providers of ecological services that support tourism development Therefore, this study, titled "Using Geospatial Technology to Map Forest Carbon," aims to explore these critical aspects.
Cuc Phuong National Park" for the student graduation thesis
GOALS AND SPECIFIC OBJECTIVES
Goals
Evaluating the carbon stock in the forests of Cuc Phuong National Park provides essential scientific data that informs sustainable forest management and development strategies in Cuc Phuong commune, Ninh Binh province.
Scope of object
Studying the forests of Cuc Phuong national park in Cuc Phuong commune, Nho Quan district, Ninh Binh province.
Specific objectives
- Mapping the land cover in study site from SPOT 5 imagery
- Estimating and mapping the forest volume in study area from SPOT 5 imagery
- Estimating and mapping the aboveground carbon stock of forests in study area
- Proposing possible solutions to enhance forest management in study area
The forests carbon stock in Cuc Phuong Natinal Park at Cuc Phuong commune over 1,000,000 ton
STUDY AREA AND METHODOLOGY
Study area
The study site includes the forests of Cuc Phuong national park in Cuc Phuong Commune, Nho Quan District, Ninh Binh Provinces (Figure 3.1)
Figure 3.1.1: Map of study area: (a) Land border map of Vietnam (b) Ninh Binh
Province (c) Cuc Phuong commune, Nho Quan Districts
Cuc Phuong National Park, located between Ninh Binh, Hoa Binh, and Thanh Hoa provinces, spans a total area of 22,180 hectares, with 51.1% in Ninh Binh, 26.4% in Hoa Binh, and 22.5% in Thanh Hoa Positioned at 350 meters above sea level, the park experiences an average annual temperature of 20.6°C, with temperatures ranging from 19.9°C to 21.2°C, and receives annual precipitation between 1,800 mm and 2,400 mm The park's relative humidity averages 90%, promoting a diverse flora and fauna Its rich vegetation includes towering dipterocarp trees, semi-evergreen species, and a dense canopy of sclerophyllous evergreens, with over 2,000 vascular plant species, including several endangered and endemic varieties Additionally, Cuc Phuong is home to around 300 bird species, 65 mammal species, 37 reptile species, and 16 amphibian species, making it a vital ecological habitat.
Data sources
Studying the use inheritance to collect data with the following information:
In 2015, SPOT 5 satellite imagery was obtained for the study area from the Institute for Forest Ecology and Environment at the Vietnam National University of Forestry (IFEE) Configuration details of the SPOT 5 imagery are outlined in Table 1.
Table 3.2.1: Technical parameters and properties of the sensors used in this study
500–590 (green) 610–680 (red) 790–890 (NIR) 1580–1750 (mid IR)
Spatial Resolution (m) Pixel Size 1.5 × 1.5 (m/pxl)
Methodology
The proposed methodology is illustrated in Figure 3.2 Software eCognition Developer v8.9, MapInfo Professional 12.5 and ArcGIS desktop 10.2 were used
Figure 3.3.1: The processing steps of study
The study used eCognition v8.9 image analysis software for pre-analysis image SOPT 5 and NDVI for key of image classification eCognition v8.9 image analysis software
The eCognition v8.9 image analysis software facilitates image segmentation, which is the initial step in object-based image analysis (Baatz et al., 2004) Image segmentation involves partitioning an image into distinct regions according to specified parameters (Myint et al., 2008) These parameters typically take into account the homogeneity or heterogeneity of the regions (Pal et al., 1993).
The eCognition software utilizes multiresolution segmentation to classify homogeneous areas within forest boundaries, dividing them into smaller sub-plots for effective basin planning Key parameters include a scale parameter, shape set at 0.1, and compactness at 0.5 The software exports detailed boundary maps of the forest plots, providing essential statistical information such as the average value and standard deviation for each channel.
Normalized Different Vegetation Index (NDVI)
The study used NDVI to map the status of forest in Cuc Phuong National Park with 60 sample plots and 65 points in study area NDVI is calculated by following formula:
Where: Nir spectral reflectance values of the near-infrared channel (Band 3), the spectral values of channel Red (Band 2) in Spot 5
A survey conducted following the typical sample survey method, as outlined in Circular No 34/2009/TT-BNN, delineated plots of state forests and assessed the distribution of forest status subjects This study focused on 60 plots and established 65 points for comprehensive analysis in the region.
The study involves assessing the status of forest reserves and evaluating the accuracy of map interpretations, utilizing 20 plots and 22 points for testing accuracy The distribution of plots and secondary surveys varies across different forest states, influenced by the accessibility and conditions of the forests, resulting in differing numbers of plots and secondary surveys for each state.
The study area in Cuc Phuong National Park exhibits significant variability in topography and vegetation types, influencing the design of sample plots The chosen plot size of 25m x 40m balances accuracy, precision, time, and cost, while its appropriateness may vary with the vegetation type Utilizing GPS for high accuracy, measurements include tree diameter at breast height (DBH) and tree height, following specialized regulations for foreign industrial data collection After establishing a 1000m² sample plot, five 4m x 4m sub-plots are created to gather detailed information, including shrub data within these sub-plots.
Figure 3.3.3: The layout of distribution points
No Name X Y D13(Average) H(Average) Note
The study involves overlaying the analyzed plots and points onto a boundary map, while tagging land cover classes with their corresponding NDVI values Additionally, a statistical analysis of NDVI values for each land cover class is conducted to establish a key for image classification.
This thesis focuses on image classification for various objects within the study area, utilizing keywords related to forest conditions It aims to identify key forest reserves for effective image interpretation and mapping of the forest's status By grouping NDVI values for each object and employing reclassification in the analysis tool, the study enhances the understanding of forest conditions and reserves.
Accuracy assessment is a crucial step in the image classification process, which can be evaluated through positional or thematic accuracies This study utilized a matrix table, a widely used method for assessing classification agreement in remote sensing The classification precision was verified using 22 points—2 for Water, 5 for Plantation, 5 for Bamboo, and 10 for Others—along with 25 plots, comprising 5 plots for each forest type.
The study involved overlaying the forest allocation map with the forest status map to assess forest volume across various types Utilizing data from 60 previously established sample plots, the research calculated the forest volume and generated a comprehensive forest volume map.
(Source: The circular 34, Ministry of Agriculture and Rural Development)
The following formula were used to calculate carbon was assessed in five forest carbon pools, which is in accordance with the IPCC 2006 GL (Estrada 2011) These forest carbon pools are:
1 Aboveground vegetation: carbon stocked in live and standing vegetation (trees, shrubs, undergrowth and regeneration)
2 Belowground vegetation: carbon stored in roots
3 Dead wood: carbon stored in standing and fallen dead trees and shrubs
4 Litter: carbon in shed leaves and fine branches
5 Soil: carbon stored as soil organic matter
This article focuses on the carbon stock present in aboveground biomass, specifically examining aboveground vegetation Utilizing data from forest volume measurements alongside Biomass Conversion and Expansion Factors (BCEF), we apply a specific formula to derive accurate results.
Aboveground biomass carbon stock of trees in plot sp, in stratum i at time t=0; t
Merchantable volume of tree l of species j in plot sp in stratum i at time t=0, m3
Biomass conversion and expansion factor for conversion of merchantable volume to total aboveground tree biomass for tree species j; dimensionless Carbon fraction of biomass for tree species j;
L 1, 2, 3, … Nj,sp,i,t sequence number of individual trees of species j in sample plot sp in stratum i at time t
I 1, 2, 3, …M strata j 1, 2, 3 … S tree species t=0 0 years elapsed since start of the project activity
With =0.47 (Andreae and Merlet, 2001; Chambers et al., 2001; McGroddy et al., 2004; Lasco and Pulhin, 2003)
BCEF for nature forests in North Central Vietnam here:
(Sources: 2006 IPCC Guidelines for National Greenhouse Gas Inventories)
Evaluate carbon value for forest environmental services
Amount of CO2 absorbed = Carbon *44/12
- According to the Forest Science Institute of Vietnam
Carbon Price = Amount of CO2 * Price (USD/tonCO2)
- The price in the carbon dioxide:
• Low price: 5 USD / ton CO2
• High price: 11 USD / ton CO2
• $ 1 = 22300 VND exchange rate at the time of the study
RESULTS
Land cover classification map of Cuc Phuong National Park
The study utilized NDVI to categorize land cover across eight distinct classes, including water, bare land, plantation, grasses, restored forest, poor forest, medium forest, and rich forest GPS technology was employed to mark ten plots representing various surface covers, specifically focusing on restored forest, poor forest, medium forest, and rich forest, while excluding water.
8 points, Others (including bare land, grasses and residential) have 15 points, Plantation have
10 points and Bamboo forest have 10 points
Table 4.1.1: NDVI value for water
Table 4.1.2: NDVI value for others
Table 4.1.3: NDVI value for Bamboo forest based-object
Table 4.1.4: NDVI value for poor forest
Table 4.1.5: NDVI value for restored forest
Table 4.1.6: NDVI value for plantation based-object
(Sources: Cuc Phuong National Park - Studying using data and documents about forest volume of plantation at Cuc Phuong National Park managers in 2016)
Table 4.1.7: NDVI value for medium forest
Table 4.1.8: NDVI value for rich forest
Tables 4.1.1 to 4.1.8 reveal varying NDVI values, including minimum, maximum, and standard deviation for each land cover type Water exhibited an NDVI maximum of approximately -0.09021 and a minimum of around -0.39336 In contrast, rich forest recorded the highest NDVI values, ranging from 0.85263 to 0.78427.
I concluded that NDVI ranging were clearly indicated in Table 4.9 (Data collected in Appendixes)
Table 4.1.9: NDVI range for each land cover class in Cuc Phuong National Park
No Land cover classes NDVI
The NDVI values for water in the study area ranged from -0.39336 to -0.09021, indicating strong light absorption in the infrared spectrum In contrast, various forest classes, including bamboo, restored, poor, medium, and rich forests, exhibited NDVI values between 0.53498 and 0.85263 The plantation class showed NDVI values ranging from 0.24363 to 0.53468, while other land types, primarily consisting of grass and small shrubs, had values between -0.09121 and 0.24263 The low NDVI for water is attributed to its high infrared absorption, while bare land and regenerating areas reflect more infrared light, resulting in higher NDVI values compared to water.
The NDVI values in Cuc Phuong National Park indicate that bamboo forests, restored forests, and varying qualities of natural forests (poor, medium, and rich) exhibit higher NDVI readings compared to other forest types This is attributed to the presence of natural trees and a lower density of shrubs and regeneration, resulting in relatively high NDVI values that, while notable, remain below those of untouched natural forests.
Figure 4.1.1: The status forest map in Cuc Phuong commune
To verify the accuracy of this study, I created five plots for each forest type and collected data points for water and other features using a GPS device The accuracy statistics are detailed in Table 4.1.10.
Table 4.1.10: Summarized accuracy of objects
Land cover Water Others Planta- tion
The results showed that error of interpretation on the status of forests is 16.25%, the accuracy of interpretation of the status of forests is 83.75%
The Table 4.1.11 showed the area of land cover classes in study area with eight classes with responding area class
Table 4.1.11: The status forest map in Cuc Phuong commune
No Land cover classes Area (ha) %
Table 4.1.11 reveals that rich forests dominate the area, covering approximately 9,447 hectares, which accounts for 72.07% of the total In contrast, water and bamboo classes are minimal, with areas of just 3.31 hectares and 2.15 hectares, representing 0.03% and 0.02%, respectively The medium forest categories, including poor and medium forests, comprise a combined area of around 3,225 hectares, or about 16.53% of the total, with poor forests at 1,095 hectares (8.32%), medium forests at 1,085 hectares (8.28%), and other forests at 945 hectares (7.21%) Conversely, plantation and restored forests occupy a mere 2% of the total area, with sizes of approximately 327 hectares and 203.74 hectares, respectively.
The forest volume map in Cuc Phuong National Park of Cuc Phuong commune
Table 4.2.1 presents the forest volume of Cuc Phuong National Park categorized by forest types The analysis specifically examines five classes: rich, medium, poor, restored, plantation, and bamboo forest, as the other classes, including Water and Other, do not have identifiable volumes.
Table 4.2.1: Forest volume of Cuc Phuong National Park
Class The average forest volume (m 3 /ha)
The data reveals that rich forests have the highest volume at 298.1 m³/ha, while restored forests show the lowest volume at approximately 41.4 m³/ha Additionally, plantation forests average 78 m³/ha, medium forests at 177 m³/ha, and poor forests at 79.3 m³/ha, indicating a consistent decrease in forest volume across these categories.
Figure 4.2.1: The forest volume map in Cuc Phuong commune
The map of forest carbon in Cuc Phuong National Park of Cuc Phuong commune
The study calculated the carbon dioxide absorbed in study site, which illustrated in following figure
Figure 4.3.1: The forest carbon map in Cuc Phuong National Park of
Figure 4.3.1 shows that the carbon stock in the study site reaches a peak of 133.1 tons in rich and medium forest areas, while water and bare land categories register the lowest carbon stock at 0 tons.
Table 4.3.1: Total forest carbon in Cuc Phuong National Park
The average forest volume (m 3 /ha)
As the result of Table 4.3.3 indicated the total forest carbon stock in Cuc Phuong
Cuc Phuong National Park has a total carbon stock of approximately 1,480,000 tons, with the rich forest accounting for the majority at 1,257,493.81 tons, representing 85.52% of the total The medium and poor forests contribute 117,383.84 tons and 69,394.54 tons, which are about 7.98% and 4.72% respectively Plantation forests add 17,982.83 tons, approximately 1.22% of the total carbon stock The restored forest has a minimal contribution of 8,126.96 tons, or 0.55% Other land types, such as water bodies, bare land, and bamboo forests, were not included in the carbon stock assessment, which focused solely on the volumes of rich, medium, poor, plantation, and restored forests.
Mapping the Carbon Dioxide absorbed
The study calculated the carbon dioxide absorbed in study site, which illustrated in following figure
Figure 4.4.1: The CO 2 absorbed map in Cuc Phuong commune Table 4.4.1: The CO 2 absorbed in Cuc Phuong National Park at Cuc Phuong Commune
With formula: Amount of CO2 absorbed = Carbon *44/12 The figure 4.4.1 indicated the CO2 absorbed in forest at Cuc Phuong commune with the highest value at 488.04 ton/ha
27 for rich forest and the lowest value is 146.26 ton/ha (represent 144.87 ton/ha) for restored forest
Evaluate carbon dioxide value for forest environmental services
Table 4.4.2: Evaluate carbon dioxide value for forest environmental services
CO 2 USD/ton CO 2 per ha
Ton/ha Low price High price
Figure 4.4.2: Figure carbon absorb value of different forest types in CPNP
USD/ton CO2 per ha USD/ton CO2 per ha
The carbon absorption capacity and associated prices of various forest types reveal significant differences Rich forests lead in carbon absorption, priced at $2,440.17 per hectare, followed by medium forests at $1,982.75 per hectare and poor forests at $1,161.60 per hectare In contrast, plantation and restored forests have the lowest values, at $1,008.15 and $731.20 per hectare, respectively.
The primary factor contributing to the differences in carbon absorption capacity among forest types is their uneven age distribution Variations in growth index and density further influence their ability to sequester carbon effectively.
4.3 Proposing solutions to improve the management of forests in Cuc Phuong commune
Cuc Phuong National Park in Hanoi faces significant challenges regarding forest volume and carbon stock, including inadequate land use planning and ineffective management plans The park contains a considerable area of mature forest that is experiencing degradation, alongside a large land area that increases the risk of erosion during heavy rains To address these issues, it is essential to implement solutions aimed at enhancing forest quality and optimizing carbon stock management and carbon dioxide absorption in Cuc Phuong Commune.
Strengthening cooperation in forest management involves clearly defining the rights and obligations of individuals or organizations assigned to land and forests The state must ensure that those responsible for managing forest resources are held accountable for any legal violations Additionally, rigorous inspections and strict enforcement are necessary to address unlawful encroachments and improper land use conversions.
The application of information technology into management and monitoring changes in forest resources processes such as organizations and units who own protection forest should
29 participate fully in managing and monitoring changes in forest resource by using database system on the computer at the local and the whole province
To foster sustainable ecotourism and enhance forest protection, it is essential to encourage investment through both domestic and international policy support Additionally, prioritizing the education and awareness of the younger generation about the vital role forests play in human life and the environment is crucial for long-term conservation efforts.
A comprehensive study was conducted to map the forest carbon in Cuc Phuong National Park, specifically in Cuc Phuong commune This research produced three distinct maps: one illustrating forest volume, another depicting forest carbon, and a third showing the carbon dioxide absorption in the study area.
The first map detailing the forest volume of Cuc Phuong National Park, covering a study area of 13,109.84 hectares, categorizes the land into eight classes: water, plantation, bamboo forest, restored forest, poor forest, medium forest, rich forest, and others However, the focus of the study is primarily on five forest types with the highest volume value, particularly the rich forest, while the water, bamboo, and other categories do not significantly contribute to forest volume Conversely, the restored forest exhibits the lowest volume value due to its status as a replantation and recovery area.
The thesis presents a comprehensive mapping of carbon stock in the study area, revealing a total forest carbon of approximately 1,480,000 tons Notably, the highest carbon density recorded is 133.10 tons per hectare, surpassing the average of 130 tons per hectare and the total of 1,023,726 tons of carbon documented in Kim Boi District, Hoa Binh Province, as reported by Assoc Prof Dr Tran Quang Bao.
2013, identified the carbon stock of forest classes in Kim Boi District, Hoa Binh province
This thesis presents a comprehensive mapping of carbon dioxide absorption in Cuc Phuong National Park, revealing a significant total CO2 absorption in Cuc Phuong Commune, with peak values reaching approximately 488.04 tons per hectare.
Cuc Phuong National Park boasts a high carbon volume and significant carbon dioxide absorption, highlighting its ecological importance This natural forest is characterized by its rich biodiversity, primarily consisting of evergreen broadleaf trees and limestone-rich primary forests Compared to other studies, the findings emphasize the necessity of protecting Cuc Phuong National Park to preserve its unique wildlife and plant species for future generations.
In conclusion, the study revealed that utilizing remote sensing technology and high-resolution satellite images significantly enhances forest investigation, achieving an impressive interpretation accuracy of 83.75% in mapping forest type changes and quality The analysis of SPOT 5 satellite images identified key land cover categories in Cuc Phuong commune, including rich, medium, poor, restored, bamboo, plantation forests, water, and others The total study area encompasses 13,109.84 hectares, with an estimated total forest carbon stock of 1,480,000 tons.
The study reveals that the rich forest spans an impressive 9,447.62 hectares, accounting for 72.07% of the total area, and boasts a volume of 298.1 m³/ha, which corresponds to 1,257,493.81 tons of carbon, representing 85.52% of the total carbon stock In contrast, the restored forest exhibits the lowest values, with a volume of only 203.74 m³/ha and a carbon stock of 8,126.96 tons, making up just 0.55% of the total.
The carbon stock and CO2 absorbed for rich forest in forest at Cuc Phuong commune with the highest value around 133.1 ton/ha and 488.04 ton/ha, respectively
The study faces several challenges, including insufficient time and resources for extensive field data collection, as well as my limited expertise To address these issues and enhance outcomes, future research should focus on gathering more comprehensive information for planning, expanding the number of sample plots across various forest types, and improving users' proficiency in geospatial technology application.
Adams, R.M., C.C Chang, B.A McCarl, and J.M Callaway 1993 Sequestrating Carbon on Agricultural Land: A Preliminary Analysis of Social Cost and Impacts on Timber Markets Contemporary Policy Issues 11(January)
Dixon, R.K., J.K Winjum, K.J Andrasko, J.J Lee, and P.E Schroeder 1994 Integrated LandUse Systems: Assessment of Promising Agroforest and Alternative Land-Use Practices to Enhance Carbon Conservation and Sequestration
Duan, P.V, Khoa P.V 2014 Testing a method for establishing the watershed’s forest inventory map using satellite image SPOT5
Marland, G 1988 The Prospect of Solving the CO2 Problem Though Global Reforestation U.S Department of Energy, Office of Energy Research, Carbon Dioxide Division
Moulton, R and K Richards 1990 Costs of Sequestration Carbon Through Tree Planting and Forest Management in the United States U.S Department of Agriculture Forest Service General Technical Report WO-58
Parks, P.J and I.W Hardie 1995 Least-Cost Forest Carbon Reserves: Cost Effective Subsidies to Convert Marginal Agricultural Land to Forest Land Economics 71
Phutchard Vicharnakorn, Rajendra P.Shrestha, Masahiko Nagai, Abdul P Salam, Somboon Kiratiprayoon 2014 Carbon Stock Assessment Using Remote Sensing and Forest Inventory Data in Savannakhet, Lao PDR
Kauppi and Sedjo, along with other lead authors, contributed to the Intergovernmental Panel on Climate Change (IPCC) in 2001, specifically in the Third Assessment Report (TAR) Working Group III, Chapter 4 This chapter focuses on the technical and economic potential of strategies to enhance, maintain, and manage biological carbon reservoirs, as well as the implications of geo-engineering solutions.
The Kyoto Protocol, established in 1997, is an international treaty under the United Nations Framework Convention on Climate Change (UNFCCC) aimed at combating climate change by reducing greenhouse gas emissions For the full text of the Kyoto Protocol, you can visit the official UNFCCC website at https://unfccc.int/resource/docs/convkp/kpeng.html, accessed on October 10, 2023.
REDD Methodological Module 2009 Estimation of Carbon stocks and changes in carbon stocks in the above-ground biomass carbon pool
Roger A Sedjo August 2001, Forest Carbon Sequestration: Some Issues for Forest Investments
No Commune X Y D13(Average) H(Average) Note