A Primer on Estimating Biomass and Carbon Stocks in Tropical Forests: Brown 1997 Estimating Biomass.pdf
Estimating Biomass and Carbon Stocks in Tropical Forests
Biomass is the total amount of living organic matter in an ecosystem. It is a key indicator of ecosystem health, productivity, and biodiversity. Biomass also plays a crucial role in the global carbon cycle, as it stores and releases carbon through photosynthesis, respiration, decomposition, and combustion. Estimating biomass and carbon stocks in tropical forests is therefore essential for understanding their contribution to climate change mitigation and adaptation, as well as for designing and implementing effective conservation and management strategies.
Brown 1997 Estimating Biomass.pdf
Introduction
What is biomass and why is it important?
Biomass can be defined as the mass of living organisms per unit area or volume. It can be measured at different levels of biological organization, such as individual plants, species, communities, or ecosystems. Biomass can also be classified into different components, such as aboveground biomass (AGB), belowground biomass (BGB), deadwood, litter, or soil organic matter.
Biomass is important for several reasons. First, it reflects the structure, composition, and function of ecosystems, as it is influenced by factors such as climate, soil, disturbance, and human activities. Second, it provides various ecosystem services, such as food, fuel, fiber, medicine, habitat, and water regulation. Third, it affects the global carbon cycle, as it sequesters carbon from the atmosphere through photosynthesis and emits it back through respiration, decomposition, or combustion. Therefore, biomass is a key component of the terrestrial carbon sink, which offsets about 30% of the anthropogenic carbon emissions annually.
How to measure biomass in tropical forests?
Measuring biomass in tropical forests is challenging due to their high diversity, complexity, and spatial variability. There are different methods for estimating biomass and carbon stocks in tropical forests, which can be broadly categorized into three types: field-based methods, remote sensing methods, and modelling methods.
Field-based methods involve directly measuring or sampling the biomass of trees or other vegetation components in situ. These methods are usually based on allometric equations, which relate the biomass of a plant or a plant part to one or more easily measurable variables, such as diameter at breast height (DBH), height, or crown area. Field-based methods can also use biomass expansion factors (BEFs), which convert volume or basal area to biomass; root-to-shoot ratios (RSRs), which estimate the belowground biomass from the aboveground biomass; and carbon fractions (CFs), which convert biomass to carbon content.
Remote sensing methods use satellite or aerial imagery to estimate biomass over large areas. These methods rely on the relationship between biomass and spectral reflectance or radar backscatter of vegetation. Remote sensing methods can also use lidar (light detection and ranging) or radar (radio detection and ranging) to measure the vertical structure and canopy height of forests, which are correlated with biomass.
Modelling methods use mathematical equations or computer simulations to estimate biomass based on environmental variables, such as climate, soil, topography, disturbance, and management. These methods can be empirical, mechanistic, or hybrid. Empirical models use statistical techniques to derive the relationship between biomass and environmental variables from observed data. Mechanistic models use biophysical principles to simulate the processes of biomass production and allocation. Hybrid models combine both empirical and mechanistic approaches.
What are the challenges and uncertainties of biomass estimation?
Estimating biomass and carbon stocks in tropical forests is subject to various sources of error and uncertainty, which can affect the accuracy and reliability of the estimates. Some of the main challenges and uncertainties are:
Lack of data: There is a scarcity of field data on biomass and carbon stocks in tropical forests, especially in remote, inaccessible, or disturbed areas. This limits the development and validation of allometric equations, BEFs, RSRs, CFs, remote sensing algorithms, and models.
Variability: There is a high spatial and temporal variability of biomass and carbon stocks in tropical forests, due to factors such as species diversity, stand structure, age, site quality, disturbance, and management. This requires a large number of samples or observations to capture the heterogeneity and dynamics of biomass and carbon stocks.
Scaling: There is a difficulty in scaling up biomass and carbon estimates from plot to landscape to regional to global scales, due to differences in methods, definitions, units, and assumptions. This introduces errors and biases in the aggregation and extrapolation of biomass and carbon estimates.
Uncertainty propagation: There is a propagation of uncertainty from each step of the estimation process to the final result, due to errors or variability in the input data, parameters, or methods. This affects the confidence and precision of the biomass and carbon estimates.
Methods
Brown's (1997) approach for estimating biomass and carbon stocks
One of the most widely used methods for estimating biomass and carbon stocks in tropical forests is the approach proposed by Brown (1997), which is based on field measurements of trees and application of allometric equations, BEFs, RSRs, and CFs. The main steps of this approach are:
Select a representative sample of plots across the study area, with a minimum size of 0.1 ha for closed forests and 0.5 ha for open forests.
Measure the DBH of all trees with DBH 10 cm within each plot. For trees with buttresses, measure the DBH above the buttresses or use an appropriate correction factor.
Estimate the AGB of each tree using an allometric equation that is specific to the region, forest type, or species group. If no such equation is available, use a general equation for tropical forests.
Estimate the BGB of each tree using an RSR that is specific to the region, forest type, or species group. If no such RSR is available, use a default value of 0.24 for moist tropical forests or 0.37 for dry tropical forests.
Estimate the total biomass (TB) of each tree by adding the AGB and BGB.
Estimate the plot-level biomass by summing up the TB of all trees within each plot and dividing by the plot area.
Estimate the carbon stock of each tree by multiplying the TB by a CF that is specific to the region, forest type, or species group. If no such CF is available, use a default value of 0.47 for all tropical forests.
Estimate the plot-level carbon stock by summing up the carbon stock of all trees within each plot and dividing by the plot area.
Estimate the area-level biomass and carbon stock by averaging the plot-level estimates across all plots within the study area.
Allometric equations
Allometric equations are mathematical models that relate the biomass of a plant or a plant part to one or more easily measurable variables, such as DBH, height, or crown area. Allometric equations can be derived from destructive sampling (harvesting and weighing) or non-destructive sampling (using volume or density measurements) of plants. Allometric equations can be species-specific, genus-specific, family-specific, or general for a region or a forest type.
The most common form of allometric equation for estimating AGB of trees in tropical forests is: $$AGB = a \times DBH^b$$ where AGB is the aboveground biomass (kg), DBH is the diameter at breast height (cm), and $a$ and $b$ are regression coefficients that vary depending on the region, forest type, or species group.
Some examples of allometric equations for estimating AGB of trees in tropical forests are:
Biomass expansion factors
Biomass expansion factors (BEFs) are conversion factors that estimate the biomass of a tree or a forest stand from its volume or basal area. BEFs can be applied to trees or forest stands that have been measured using forest inventories or remote sensing. BEFs can be derived from allometric equations or from direct measurements of biomass and volume or basal area. BEFs can vary depending on the region, forest type, species group, or biomass component.
The general form of BEF for estimating AGB of trees in tropical forests is: $$BEF = \fracAGBV$$ where BEF is the biomass expansion factor (kg/m3), AGB is the aboveground biomass (kg), and V is the volume (m3).
Some examples of BEF values for estimating AGB of trees in tropical forests are:
For moist tropical forests in Africa, the mean BEF value is 1.22 kg/m3 (Brown et al., 1989).
For moist tropical forests in Asia, the mean BEF value is 1.12 kg/m3 (Brown and Lugo, 1992).
For moist tropical forests in Latin America, the mean BEF value is 0.99 kg/m3 (Brown and Lugo, 1992).
Root-to-shoot ratios
Root-to-shoot ratios (RSRs) are ratios that estimate the belowground biomass (BGB) of a tree or a forest stand from its aboveground biomass (AGB). RSRs can be applied to trees or forest stands that have been measured using allometric equations or biomass expansion factors. RSRs can be derived from destructive sampling (harvesting and weighing) or non-destructive sampling (using soil coring or excavation) of roots. RSRs can vary depending on the region, forest type, species group, or environmental conditions.
The general form of RSR for estimating BGB of trees in tropical forests is: $$RSR = \fracBGBAGB$$ where RSR is the root-to-shoot ratio (dimensionless), BGB is the belowground biomass (kg), and AGB is the aboveground biomass (kg).
Some examples of RSR values for estimating BGB of trees in tropical forests are:
For moist tropical forests in Africa, the mean RSR value is 0.24 (Mokany et al., 2006).
For moist tropical forests in Asia, the mean RSR value is 0.26 (Mokany et al., 2006).
For moist tropical forests in Latin America, the mean RSR value is 0.22 (Mokany et al., 2006).
Carbon fractions
Carbon fractions (CFs) are fractions that estimate the carbon content of a tree or a forest stand from its biomass. CFs can be applied to trees or forest stands that have been measured using allometric equations, biomass expansion factors, or root-to-shoot ratios. CFs can be derived from chemical analysis or from literature values. CFs can vary depending on the region, forest type, species group, or biomass component.
The general form of CF for estimating carbon stock of trees in tropical forests is: $$CF = \fracCB$$ where CF is the carbon fraction (dimensionless), C is the carbon stock (kg), and B is the biomass (kg).
Some examples of CF values for estimating carbon stock of trees in tropical forests are:
For moist tropical forests in Africa, the mean CF value is 0.47 (Brown et al., 1989).
For moist tropical forests in Asia, the mean CF value is 0.47 (Brown and Lugo, 1992).
For moist tropical forests in Latin America, the mean CF value is 0.47 (Brown and Lugo, 1992).
Other methods for estimating biomass and carbon stocks
Remote sensing
Remote sensing is the use of satellite or aerial imagery to estimate biomass and carbon stocks over large areas. Remote sensing can provide spatially explicit and temporally consistent information on forest extent, structure, and change. Remote sensing can also complement field-based methods by reducing the sampling error and uncertainty of biomass and carbon estimates.
Remote sensing methods for estimating biomass and carbon stocks in tropical forests can be classified into two types: passive and active. Passive remote sensing methods use the reflected or emitted electromagnetic radiation from the sun or the earth's surface to measure the spectral reflectance or thermal emission of vegetation. Active remote sensing methods use the transmitted and received electromagnetic radiation from a sensor to measure the radar backscatter or lidar return of vegetation.
Some examples of remote sensing methods for estimating biomass and carbon stocks in tropical forests are:
Optical imagery: Optical imagery uses the visible, near-infrared, and shortwave infrared bands of the electromagnetic spectrum to measure the spectral reflectance of vegetation. Optical imagery can be used to estimate biomass and carbon stocks by using vegetation indices, such as the normalized difference vegetation index (NDVI) or the enhanced vegetation index (EVI), which are correlated with biomass and carbon stocks. Optical imagery can also be used to classify forest types, detect forest cover change, and monitor forest degradation.
Radar imagery: Radar imagery uses the microwave band of the electromagnetic spectrum to measure the radar backscatter of vegetation. Radar imagery can be used to estimate biomass and carbon stocks by using the radar cross-section, which is related to the structural characteristics of vegetation, such as biomass, canopy height, and density. Radar imagery can also be used to penetrate clouds and smoke, which are common in tropical regions.
Lidar imagery: Lidar imagery uses the near-infrared band of the electromagnetic spectrum to measure the lidar return of vegetation. Lidar imagery can be used to estimate biomass and carbon stocks by using the canopy height, which is derived from the difference between the ground and canopy returns. Lidar imagery can also be used to measure the vertical structure and heterogeneity of forests, which are important for biodiversity and carbon storage.
Forest inventories
Forest inventories are systematic surveys of forest resources that collect information on forest area, volume, biomass, carbon, growth, yield, and quality. Forest inventories can provide accurate and reliable data on forest status and trends at national or subnational scales. Forest inventories can also support forest management and planning, as well as policy and decision making.
Forest inventories methods for estimating biomass and carbon stocks in tropical forests can be classified into two types: design-based and model-based. Design-based methods use a probabilistic sampling design to select a representative sample of plots across the study area, and then apply field-based methods (such as allometric equations, BEFs, RSRs, or CFs) to estimate biomass and carbon stocks within each plot. Model-based methods use a non-probabilistic sampling design to select a sample of plots across the study area, and then apply modelling methods (such as regression models, k-nearest neighbors, or artificial neural networks) to estimate biomass and carbon stocks within each plot.
Some examples of forest inventory methods for estimating biomass and carbon stocks in tropical forests are:
National Forest Inventory (NFI): NFI is a comprehensive and periodic inventory of forest resources at the national level. NFI usually uses a design-based method with a systematic or stratified sampling design. NFI can provide consistent and comparable data on forest area, volume, biomass, carbon, and other variables across countries and regions.
Forest Resource Assessment (FRA): FRA is a global assessment of forest resources conducted by the Food and Agriculture Organization (FAO) every five years. FRA uses a model-based method with a purposive sampling design. FRA can provide harmonized and updated data on forest area, volume, biomass, carbon, and other variables across countries and regions.
Reducing Emissions from Deforestation and Forest Degradation (REDD+): REDD+ is an international mechanism that provides financial incentives for developing countries to reduce greenhouse gas emissions from deforestation and forest degradation, and to enhance forest carbon stocks. REDD+ uses a combination of design-based and model-based methods with a nested sampling design. REDD+ can provide accurate and transparent data on forest area change, biomass change, carbon emissions, and carbon removals across countries and regions.
Modelling
and carbon stocks in tropical forests based on environmental variables, such as climate, soil, topography, disturbance, and management. Modelling can provide mechanistic and dynamic understanding of the processes and drivers of biomass and carbon stocks in tropical forests. Modelling can also provide projections and scenarios of future biomass and carbon stocks under different environmental and socio-economic conditions.
Modelling methods for estimating biomass and carbon stocks in tropical forests can be classified into three types: empirical, mechanistic, or hybrid. Empirical models use statistical techniques to derive the relationship between biomass and carbon stocks and environmental variables from observed data. Mechanistic models use biophysical principles to simulate the processes of biomass production and allocation. Hybrid models combine both empirical and mechanistic approaches.
Some examples of modelling methods for estimating biomass and carbon stocks in tropical forests are:
Chave et al. (2005) model: This is an empirical model that estimates AGB of trees in tropical forests from DBH, wood density, and height. The model uses a power-law function with coefficients that vary depending on the forest type (dry, moist, or wet) and the continent (Africa, Asia, or America). The model is based on a large dataset of biomass measurements from 2,410 trees across 27 countries.
CASA model: This is a mechanistic model that estimates NPP (net primary production) and NEP (net ecosystem production) of tropical forests from climate, soil, and satellite data. The model uses a light-use efficiency approach to simulate the photosynthesis and respiration of vegetation and soil. The model is calibrated and validated with field measurements of carbon fluxes from eddy covariance towers.
Geoscience Laser Altimeter System (GLAS) model: This is a hybrid model that estimates AGB of tropical forests from lidar data and ancillary data. The model uses a regression tree approach to relate the lidar-derived canopy height to the field-measured AGB, and then applies a spatial interpolation technique to extrapolate the AGB estimates over large areas. The model is based on a global dataset of lidar observations from the ICESat satellite and field measurements of AGB from 3,000 plots.
Results and Discussion
Comparison of different methods and their accuracy and reliability
The different methods for estimating biomass and carbon stocks in tropical forests have different strengths and weaknesses, which affect their accuracy and reliability. A comparison of the main advantages and disadvantages of each method is presented in Table 3.
Table 3: Comparison of different methods for estimating biomass and carbon stocks in tropical forests Method Advantages Disadvantages --- --- --- Field-based - Direct and accurate measurement of biomass and carbon stocks - Can capture the diversity and complexity of tropical forests - Can provide detailed information on forest structure, composition, and function - Time-consuming and labor-intensive - Costly and logistically challenging - Limited spatial coverage and representativeness - Subject to sampling error and uncertainty Remote sensing - Rapid and consistent measurement of biomass and carbon stocks -