Approach to and Steps in Stratification

A stratification procedure for the baseline as well as the project scenario, which requires identical steps, is presented in this section:

Step 1: Define the project boundary as described in Section 10.3. Step 2: Obtain maps of the project area and overlay the different maps representing, for example, land-use systems, soil and topography under the baseline scenario.

Step 3: Overlay the project activities on land-use systems in the baseline scenario, such as degraded forest land or grassland. Step 4: Identify the key differentiating features for stratification of land-use systems in the baseline scenario that are likely to impact carbon stocks: ° Current land-use such as open access grazing, controlled grazing, fuel-

wood extraction and rainfed cropping ° Soil quality good, moderate and low ° Topography levelled, sloping, hilly terrain

Step 5: Collect all the information available from secondary sources as well as from participatory rural appraisal (Chapter 8). Step 6: Stratify the area under the baseline scenario:

° Delineate areas under different project activities. ° Overlaying the delineated areas with key features of land-use systems that are critical for estimating baseline carbon stocks. ° Mark the strata to be brought under different project activities spatially on the project map.

Step 7: Stratify the area under the project scenario:

° Locate the project activities on the baseline scenario strata spatially. ° Mark spatially the different strata representing different project activities, land-use systems and other features; however, each stratum is homogeneous within itself.

10.3.5 Application of Remote Sensing and GIS for Stratification

A variety of remote sensing data are available, which can be used for defining project area or land-use category as described in Chapter 8. The data products can be in the form aerial photography or satellite imagery. This means that different data sets may cover different time series, a point significant for developing a baseline for above-ground carbon. More information on remote sensing and its use in carbon inventory is given in Chapter 14. Several important criteria govern the selection of remote sensing data and data products for defining the project area or land-use category (IPCC 2006).

• Adequate categorization of Jand use The data should help in distinguishing between different types of land use; for example, grassland and forest land are seldom a problem to monitor, but natural forest and degraded forest can be difficult, depending on the remote sensing product.

• Appropriate spatial resolution The data should be available at an appropriate resolution, since the spatial resolution determines how well the data can be classified into different levels of land-use classes. For most project-based assessments, a fine resolution (25 m or lower) is needed to categorize different land uses.

• Appropriate temporal resolution Because land use may change with time, the data should offer adequate temporal resolution as well for estimating land-use conversions. To be able to assess the process of change, data need to be analysed across a given time span. Depending on geographical location and vegetation, it is also important to take seasonality of vegetation into account. Usually, peak vegetation periods are the ones most easily used.

• Ground-truthing To validate the data from remote sensing, they need to be compared against data obtained by other means. Interpretation of remote sensing data is cross-checked with empirical data on vegetation in the land-use systems to be assessed. For above-ground biomass, actual biomass inventory data are very useful but management plans, land-use maps or data from participatory rural appraisal can also be used.

Geographical information system (GIS) can be used not only for interpreting actual remote sensing data but also for synthesizing the data collected over the project area. Further, it is a perfect system for storing and adding data over time. Several user-friendly programs are available to be used on regular PCs.

10.4 Selection of a Method for Estimation of Above-Ground Biomass: the "Plot Method"

Several methods are presented for estimating above-ground biomass in Chapter 9, of which the "plot method" is the one most commonly used. The method, versatile, cost-effective and applicable to baseline as well as project scenario, is described in detail here. The "plot method" is used in preparing a forest inventory and estimating biomass in grassland, crop productivity and timber and fuelwood production. "Plot method" is also among the methodologies approved by the Clean Development Mechanism for afforestation and reforestation projects ( under the Kyoto Protocol. The method involves selecting plots of an appropriate size and number, laying them randomly in the selected strata, measuring the indicator parameters (e.g. tree DBH, height or grass production), using different approaches such as allometric functions to calculate the biomass and extrapolating the value to per hectare and for the total project area. These sample plots could also be used for assessment of biodiversity, land degradation and soil fertility improvements.

10.5 Selection of Appropriate Frequency of Measurement for the Above-Ground Biomass Pool

The frequency of measurement and monitoring of above-ground biomass pool depends on the land-use system, soil quality, species and management systems (refer to Chapter 4 for details). The frequency is different for the baseline and for the project scenario and also depends on the biomass stock and its rate of growth. Frequency of monitoring has implications for carbon inventory due to the effort and cost involved:

• Baseline scenario The frequency will depend on the rate of above-ground biomass growth, which is likely to be low for most baseline scenario situations. Thus, the above-ground biomass could be monitored once in 3-5 years. The frequency of monitoring for avoided deforestation projects, with high biomass stock under baseline scenario, could also be 3-5 years.

• Project scenario The frequency will be determined by the type of project activity and the rate of growth of above-ground biomass. Fast growing species, such as those grown intensively for bio-energy plantations, may require frequent monitoring.

It is important to decide on the frequency of monitoring above-ground biomass stocks so that resources can be planned for and allocated accordingly.

10.6 Identification of the Parameters to be Measured for Estimating the Above-Ground Biomass Pool

The goal of measurement and monitoring is to estimate the stocks of above-ground biomass or its rate of growth on per hectare basis as well as for the total project area. This requires identification and selection of a key set of indicator parameters. The parameters to be selected depend on the method adopted; those required for the "plot method" are presented here. The most commonly used parameters are as follows:

(i) Name ofthe species The first parameter to be recorded is the plant form, namely tree, shrub, herb or liana, followed by the name of the species. Among trees, species differ in shape, size, rate of growth and wood density. It is also important to estimate the density of trees (number per unit area) of each species in the sample plots and per hectare. Names of species are important even for non-tree plant forms such as shrubs, herbs and grass. Biomass for tree species is estimated as volume or weight per tree, which can be extrapolated to per hectare based on the density and distribution of each species. While recording the species name and number, it is desirable to record other features such as:

° Regeneration naturally regenerated or planted ° Status oftree crown percent damaged or full crown ° Status of the tree living, dead and standing, or dead and fallen

(ii) Diameter or girth at breast height for trees Size, usually measured in terms of diameter or girth at breast height (DBH or GBH), is one of the most important parameters and represents the volume or weight of a tree, which can be converted to biomass per unit area (tonnes/hectare or tonnes/hectare/year). The diameter and height can be used for estimating the volume by simple equations; DBH values can also be used in allometric functions to estimate volume or biomass per tree or per hectare. Usually, DBH is easy to measure in the field and, by appropriate marking, the measurements can be repeated over time. The breast height in DBH is normally taken to be 130 cm above the ground. The measurement techniques are described later in the chapter (Section 10.11).

(iii) Height of trees Next to DBH, height is the most important indicator of the volume or weight of a tree and used in many allometric functions along with DBH. Measuring the height of tall trees, especially those with overlapping canopies, requires instruments and may introduce errors.

(iv) Indicatorparameters fornon-tree species Height and DBH are not measured for non-tree species such as herbs and grasses; biomass is estimated in terms of weight per unit area by actually harvesting and weighing all the herbs and grasses in the sample plots.

The parameters to be monitored for estimating above-ground biomass are listed in

Table 10.1.

Table 10.1 Parameters for estimation

Carbon pool

Parameters to be recorded


Name of the species

biomass of trees and

DBH (cm)


Height (m)

Origin: regenerated or planted

Extent of crown: full crown or percent

crown damaged

Status: dead or living


Name of the species

biomass of herb or

Density (number/plot)


Fresh weight of herb layer biomass (g/m2)


Dry weight of herb layer biomass (g/m2)

10.7 Selection of Sampling Method and Sample Size

Sampling includes deciding on the number, size and shape of the plots, a step often ignored by project developers and managers because of the perceived complexity of the methods. Two approaches can be considered for measurement and estimation of carbon in land-use systems, namely complete enumeration and sampling. Complete enumeration, measurement and monitoring of all trees and non-tree plants in different land-use systems is time-consuming, very expensive and not even necessary to get a reliable estimate of biomass. A carbon inventory based on appropriate sampling can yield reliable estimates at a limited cost and human effort. Thus, the main goal of sampling is to get a reliable estimate with minimal cost. Sampling methods include simple random sampling, stratified random sampling and systematic sampling. This section presents the principles of sampling, the accuracy and precision needed, the methods for choosing sample size and shape of the plots and practical steps to be followed. Sampling is crucial to measuring and monitoring carbon stock changes. Several books (Johnson et al. 2000; Shiver and Borders 1996; De Vries 1986; Wenger 1984) and guidebooks (MacDicken 1997; Pearson et al. 2005b; FAO 2005; are available to assist sampling, including IPCC Good Practice Guidance (IPCC 2003).

10.7.1 Sampling Principles

Sampling enables conclusions to be drawn about an entire population by observing only a portion of it. Sampling theory provides the means for scaling up information from the sample plots to the whole project area or even to a regional and national level (IPCC 2003). Thus, measurements of indicators of carbon stocks made on a small set of sample plots can be extrapolated to per hectare, for the strata and the whole project area or the land-use category. Field sampling is needed for all methods of carbon inventory -plot method, harvest method and even remote sensing techniques - that require ground-based data from sample sites for interpretation and verification. Standard sampling theory relies on random selection of a sample from the population so that each unit of the population has an equal probability of being included in the sample.

Accuracy and precision Sampling involves two common statistical concepts, namely accuracy and precision (IPCC 2003; Pearson et al. 2005b). Accuracy is a measure of how close the sample measurements are to actual values. Inaccurate or biased measurements will move the average away from the actual value. Precision is a measure of how well a value is defined. In the case of carbon inventory, precision shows how closely the results from different sampling points or plots are grouped. Figure 10.3 shows:

(a) Points are close to the centre and therefore accurate but widely spaced and thus not precise.

Fig. 10.3 Accuracy and precision

Fig. 10.3 Accuracy and precision

(b) Points are closely grouped and therefore precise, but far from the centre and thus not accurate.

(c) Points are close to the centre as well as tightly grouped, and thus both accurate and precise.

Accuracy and precision reflect how well the measurements estimate the true value of tree variables such as diameter, height and area covered by a stand of trees. An unbiased estimate will depend on repeated measurements being similar (precise) and averaging close to the true value (accuracy). A carbon inventory requires measurements that are both accurate, or close to the population values, and precise or closely grouped. Sampling involves selecting plots that are appropriate in size, number and location, which contributes to accuracy and precision. The level of precision required for carbon inventory has direct implications for inventory costs. The level of precision should be determined at the beginning of a project, and could vary from ± 5% to ± 20% of the mean. The lower the precision, the lower the confidence that the change in carbon stocks over time is real and due to project activity. The chosen level of precision will determine sample sizes for each project activity (Pearson et al. 2005b).

Confidence interval The representativeness of the estimate, or precision, is indicated by the confidence interval. Normally a 95% confidence interval is used, which implies that 95 times out of 100, the estimated value lies within the limits of twice the standard deviation.

Project Management Made Easy

Project Management Made Easy

What you need to know about… Project Management Made Easy! Project management consists of more than just a large building project and can encompass small projects as well. No matter what the size of your project, you need to have some sort of project management. How you manage your project has everything to do with its outcome.

Get My Free Ebook

Post a comment