Size ofthe Plot

A carbon inventory requires that both size and number of sample plots be decided. Plot size too has implications for the cost of carbon inventory or monitoring. Larger the plots, lower the variability between two samples. Therefore, plot size depends on the extent of variation among plots and the cost of measurement. According to Freese (1962), the relationship between plot size and its coefficient of variation (CV) is given by the following equation:

where P1 and P2 represent plot sizes and their corresponding coefficients of variation.

Increase in the plot size reduces variation among the plots and leads to smaller number of plots. Very often, the number of plots is selected based on expert judgment about the size of trees, size of project area and variations in stand density. Some illustrative numbers for different tree sizes and types of vegetation are given below. Diverse vegetation (natural vegetation) requires larger plots and homogenous vegetation (plantations of uniform age and density) requires smaller plots. In Table 10.2 Pearson et al. (2005b) suggest plot sizes for trees of different size.

Error in sampling that affects precision could occur because of variation in sample units and a large project area, measurement techniques or instruments, models or allometric equations and other errors. The relation between sampling error, population variance, and sample size is as follows (IPCC 2003):

• Increasing sample size increases precision

• Heterogeneous populations (i.e. those with large within-population variation) require larger samples to reach a given level of precision

• Where area proportions are to be estimated, sampling errors depend not only on sample size, but also on the proportion itself.

Table 10.2 Suggestion of plot size

Stem diameter Circular plot Square

10.8 Preparation for Fieldwork and Recording of Information

Estimation and monitoring of biomass in land-use systems involves measurement of plant-based parameters such as DBH and height of trees and weight of non-tree biomass. It is important to plan fieldwork well in advance for efficient use of staff and time, and necessary to procure all the background information before launching field studies. Field studies require:

• Trained staff

• Background information

• Instruments and materials for measurement

• Arrangement for collection of plant samples

• Formats for recording data

Trained staff Field studies require at least one trained person and one or two field assistants. The trained person takes the measurements and records them in the formats provided; the field assistants help in laying plots, holding the measuring device (tape, pole or scale), establishing the boundary and putting in corner pegs. It is always desirable that whosoever records the data in the field should also be the one to enter the data into the computerized database.

Background information Before embarking on field studies, it is important to obtain all the relevant background information, which will help in laying plots or taking measurements. Such background information could be obtained from the project office, land survey or forest departments, local government offices and local communities. It is particularly important to collect all the available maps and prepare a map showing the project area and boundary, baseline land-use systems and characteristics and the project scenario activities. The type of background information required includes:

• Projection of location maps showing latitude and longitude, topographic sheets, forest map and soil maps

• Names for land-use systems and their location and area

• Elevation, topography, broad soil type and rainfall

• Proximity to human settlements, roads, urban centres, markets

• Land tenure or ownership

• Livestock population and grazing locations

• Past land-use changes and features

• Data on afforestation, reforestation, soil and water conservation, i.e. programmes or activities implemented or proposed

• Sources of fuelwood and timber

• Socio-economic and demographic features

Instruments and materials for measurement The materials and instruments required for field studies on above-ground biomass estimation are given in the box below. The material required could most often be procured locally. The materials used should be of durable quality and the scales and measuring instruments validated or calibrated.

Arrangement for collection of plant samples Cloth and polythene bags are required for collecting plant samples for obtaining fresh as well as dry weights. A balance is required for taking fresh weights in the field. The dry weight is obtained by drying a sample of known weight in an oven to constant weight.

Data recording formats Formats vary for different plant forms such as trees, shrubs and herbs and need to be standardized for both field recording and for entering into the database. Sample formats are provided in Section 10.11.

- Fine measurement tape for measuring DBH (1 or 1.5 m)

- Rope and pegs for marking boundary and corner points

- Paint and brush for marking the point at which DBH is measured

- Aluminium tags for marking trees

- Global Positioning System (GPS)

- Clinometers for measuring tree height

- Slide calipers for measuring DBH of small stems

- Balance for weighing shrub, woody litter and herb layer biomass

- Cloth bags for sampling harvested or litter biomass for dry weight estimation

- Metallic frame for sampling for herb layer biomass (1 x 1m)

- Data recording sheets and pencils

10.9 Sampling Design

Sampling design aims at locating the sample plots in each of the selected stratum. The soil, topography, water availability and status of vegetation vary spatially within a land-use category, area proposed for the project activities or even in the area brought under a given project activity. Trees, biomass stock and growth rate are not distributed uniformly in a given project area or even for a given project activity, and the location of sampling plots could determine the biomass stock or growth rate estimates. The project staff may be biased in locating sample plots in spots with good tree or grass growth to obtain higher biomass stock values. Sampling techniques ensure unbiased selection of sites for laying out sample plots in the field. The main purpose of adopting a sampling design is to avoid bias in locating the sampling plots in land-use systems in both baseline scenario and project scenario. Different sampling designs for laying out sample plots for vegetation studies are as follows.

Selective sampling Selective, subjective or purposive sampling design is used for vegetation studies to assess the biomass carbon stock or roundwood production at some selected locations or as part of some projects. Although the time and effort required for locating and laying out sample plots selectively are minimal (Kangas and Maltamo 2006), the biomass estimate may not be reliable, reflecting a ground value that is not representative of the site. The error is likely to be high. Purposive sampling could be adopted, for example, to estimate the above-ground biomass of a project activity by laying plots close to and away from a village settlement to assess the impact of grazing or fuelwood extraction.

Simple random sampling To apply the simple random sampling technique, convert the project area into a large number of equal-sized grids. In this method, the sample plots are laid out randomly to avoid bias in locating the plots (Fig. 10.4). Random sampling ensures that each point or grid in the inventory area has an equal chance of being included in the sample. Further, the position of one plot has no influence on the position of other plots. Randomization makes it possible to obtain unbiased estimates of variability as well as the mean per unit area. However, randomized sampling layouts are not very convenient for field staff in locating the plots during periodical monitoring (Myers and Shelton 1980). Simple random sampling method is not often adopted out of consideration to the heterogeneity of the population or the project area because it is based on the premise that the population is homogenous, however, the method could be adopted when no prior information is available about the project area. All the area under project activity is therefore considered as one unit and the heterogeneity of soil, topography or other features is not considered. Stratified random sampling The features and benefits of stratification were described in Section 10.3. Stratification leads to efficient sampling and reduction of standard error. Each stratum can be considered as a subpopulation. In this technique, the project or activity area is stratified based on key features such as soil quality, topography, level of degradation and vegetation status and particularly the density and size of the trees. Area under each stratum is subdivided into a large number of equal-sized grids and the grids are numbered. The sample plots are chosen randomly among the grid numbers of each stratum, using the approach adopted for simple random sampling. The steps involved in stratified random sampling are as follows:

• Stratified random sampling becomes more effective with increasing homogeneity within each stratum.

• The approach is to implement the sampling procedure separately in each of the strata and then pool the information for a given project activity or land-use category.

• Stratified sampling avoids the possibility of large differences between strata contributing to the sampling error. The stratification of the sample leaves only the relatively small variation within each stratum to be reflected in the sampling error.

a. Random sampling b. Systematic sampling

Fig. 10.4 Layout for simple random sampling (left) and for systematic sampling (right)

Systematic sampling In systematic random sampling, the sample plots are placed at fixed intervals throughout the project area for a given activity. As the term implies, sample plots are not randomly distributed over the inventory area but arranged in a systematic pattern (Fig. 10.4). An important feature of systematic sampling and layout is that the position of the first plot, which is chosen at random, determines the positions of all the subsequent plots. According to Myers and Shelton (1980), the main advantage is that this approach is simple and can be adopted even in the absence of the maps. The regular spacing and systematic layout does tend to give convenient patterns for travel and fieldwork.

The disadvantages include: (i) the regular spacing of sample units might coincide with some cyclic fluctuation in the vegetation being sampled; (ii) dependence on the location of the first sampling unit; and (iii) the difficulty in estimating the variability of the population from systematic samples.

10.10 Location and Laying of Sample Plots

This section presents an approach to locating and laying out sample plots in the field in different land-use systems. The criteria in locating the plots are as follows:

• Plots located must be representative of the land-use system.

• Plots must be located in an unbiased way within the land-use system, except those for estimating leakage (Chapter 6).

• Plots should be accessible to investigators for measurement and monitoring.

The selected number of plots needs to be located and laid in the carbon inventory area in an unbiased manner, given the variations in soil, topography, vegetation, etc. Sample plots are required to be located and laid out during the project-development phase as well as project-monitoring phase. The main approach involves: (i) fixing the number of plots for each stratum or project activity; (ii) selection of sampling design; and (iii) location of the sample plots in the carbon inventory area converted to grids in each sampling stratum. The sample plots can be laid without any bias as follows:

Step 1: Select and stratify the project area or the area under each activity. Step 2: Obtain a map of the total project area and convert it into grids of appropriate size depending upon the area under a project activity or baseline land-use system. The grids could be 10 x 10 m to 100 x 100 m. It is desirable to make the grid size larger than the size of the sample plots. Further, the number of grids is usually several times the number of sample plots. Step 3: Numbers the grids 1 to n, where n is the total number of grids. Step 4: Select the number of sample plots for each land-use system under the baseline scenario and project activity stratum by using the methods described in Section 10.7.3.

Step 5: Select the sampling design: simple random sampling, stratified random sampling or systematic sampling. Step 6: Locate the sample plots in the carbon inventory area using the sampling design adopted (using the steps described in the following section).

(i) Simple random sampling

• Randomly pick as many grid numbers as the number of sample plots, using a table of random numbers or by drawing lots. For instance, if five tree plots are to be selected, pick five random numbers.

• Ensure that the randomly drawn plots do not all fall into a single cluster, which is rare.

• Locate tree plots in the grids selected in the field with respect to some permanent visible landmark and mark the boundary of each tree plot or use GPS.

• Prepare and store a map with all the details, including the location of sample plots marked on it. If GIS is available, it can be very useful.

(ii) Stratified random sampling

• Stratify the land-use system or project activities into homogeneous units.

• Adopt for each stratum the procedure given for simple random sampling.

• Repeat the procedure for laying plots for the next stratum and continue until all the strata are covered.

(iii) Systematic sampling

• Stratify the land-use system into a number of homogenous units.

• Obtain a map showing the grids in each sampling stratum and estimate the total number of grids for each stratum (N): e.g. 200 grids with a total project area of 40 ha (which is worked out below as an illustration). The plot numbers and the locations of the sample plots are marked on the grid map of the carbon inventory area.

• Calculate the sampling interval "k" by using the following equation:

k = N/n where k = sampling interval of grids or plots = 200/5 = 40, N = total number of grids representing a given strata (200) and n = number of sample plots (quadrats) to be selected (e.g. 5).

• Draw a random number smaller than k (smaller than 40 in this example), say 25.

• Select and mark the first grid based on the random number.

• The first sampling grid number is 25.

• The second sampling grid or plot = sampling interval k (40) + first sampling grid (25) = 65.

• The third sampling grid or plot = sampling interval k (40) + second sampling grid (65) = 105.

• Repeat the procedure for the remaining number of sample plots.

Marking the plot in the field The plot numbers and the locations of the sample plots are marked on the grid map of the carbon inventory area. These grid numbers have to be located in the field for long-term periodical monitoring of vegetation. The following steps could be considered to facilitate the process: Step 1: Use the carbon inventory project area maps with sample plots marked on the grid map along with their geographic coordinates (latitude and longitude).

Step 2: Locate the sample grids on the ground using GPS points from the map or using any permanent visible landmark in the field. Step 3: Mark the corners of the sample quadrats on the ground using pegs or any other permanent marking arrangements for long-term periodical monitoring. To avoid any special treatment to the permanent sample plots, it may be desirable to hide the corner points of the quadrats. Step 4: Use GPS positions of the quadrat corners for long-term periodic visits to avoid any bias in treatment of the vegetation in sample plots. The boundary line of the plot should be marked with a rope or coloured chalk powder during measurement.

Marking of tree, shrub and herb quadrats Tree sample plots or quadrats are normally several times larger than the shrub quadrats, which are several times larger than the herb or grass quadrats:

• Measure and mark the corners and boundary of the tree quadrats in the field.

• Mark the shrub quadrats within each of the tree quadrats, normally at two opposite corners, keeping two shrub plots per tree quadrat.

• Mark the herb or grass quadrats within the shrub quadrats at the opposite corners, keeping two herb plots per shrub plot.

Location and layout of the tree, shrub and herb quadrats could be along the following lines (Fig. 10.5).

a) Rectangular plot

Herb plot

Shrub plot Tree plot b) Circular plot

Herb plot

Shrub plot Tree plot b) Circular plot

c) Strip plot

Fig. 10.5 Shape or type of sampling plots: (a) rectangle, (b) circle and (c) strip c) Strip plot

Fig. 10.5 Shape or type of sampling plots: (a) rectangle, (b) circle and (c) strip

Table 10.3 Size and number of plots for different land-use systems or project activities under baseline and project scenario. (From Pearson et al. 2005b.)

Trees

Shrub

Herb/grass

Soil

Size of

No. of

Size of

No. of

Size of

No. of

Size of

No. of

Land use system

plot (m)

plots

plot (m)

plots

plot (m)

plots

plot (m)

plots

Natural forest or

50 x 40

5

5 x 5

10

1 x 1

20

1 x 1

20

heterogeneous

50 x 50

4

5 x 5

10

1 x 1

20

1 x 1

20

vegetation

Plantations with

50 x 20

5

5 x 5

8

1 x 1

16

1 x 1

16

homogenous

or

vegetation or

40 x 25

uniform species

distribution

and density

Savannah or

50 x 40

5

5 x 5

10

1 x 1

20

1 x 1

20

grassland or

rangeland with

few trees

Degraded forest

50 x 40

5

5 x 5

10

1 x 1

20

1 x 1

20

or barren or

fallow land

An illustrative sample size and number of tree, shrub and herb plots for different land-use systems are given in Table 10.3. Such sample sizes are often adopted.

10.11 Field Measurement of Indicator Parameters

Estimation of carbon stock in biomass or its growth rate requires measurement of indicator parameters such as tree height and GBH. These parameters are measured in the field through a sampling design. Field measurements for biomass carbon assessment are required at:

• Project development phase for baseline scenario land-use systems and proposed project scenario activities

• Project monitoring phase for baseline scenario land-use systems and implemented project scenario activities

Above-ground biomass is estimated in any typical land-based projects for trees, shrubs and herbs/grasses. The biomass of these plant forms is measured using the following steps:

Step 1: Decide on the sample size; locate and mark the sample plots for trees, shrubs and herbs on the ground (Section 10.7 and 10.11). Step 2: Select the parameters for tree, shrub and herb biomass (Section 10.6) and procure all the material required for field studies.

Step 3: Measure the parameters for trees, namely species, height, DBH and status or features.

Step 4: Measure the parameters for shrubs, namely height, DBH and weight of the woody and non-woody biomass. Step 5: Measure parameters for herbs/grass, namely species, number of plants, weight of the plants in the sample plots. Step 6: Record all the parameters in the standard format for trees, shrubs and herbs/grass.

Steps 1 and 2 have already been described in earlier sections. Steps 3-6 are described in Sections 10.11.1-10.11.5. These steps largely focus on measurement of different parameters as indicators of plant biomass.

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