Transcript CMM_4 - USAID LEAF

Section 4. Carbon Stock Measurement
Methods
4.3. Plot Design for Carbon Stock Inventory
USAID LEAF
Regional Climate Change Curriculum Development
Module: Carbon Measurement and Monitoring (CMM)
Name
Affiliation
Name
Affiliation
Deborah Lawrence, Co-lead
University of Virginia
Megan McGroddy, Co-lead
University of Virginia
Bui The Doi, Co-lead
Vietnam Forestry University
Ahmad Ainuddin Nuruddin
Universiti Putra Malaysia
Prasit Wang, Co-lead
Chiang Mai University,
Thailand
Mohd Nizam Said
Universiti Kebangsaan Malaysia
Sapit Diloksumpun
Kasetsart University, Thailand
Pimonrat Tiansawat
Chiang Mai University, Thailand
Pasuta Sunthornhao
Kasetsart University, Thailand
Panitnard Tunjai
Chiang Mai University, Thailand
Wathinee Suanpaga
Kasetsart University, Thailand
Lawong Balun
University of Papua New Guinea
Jessada Phattralerphong
Kasetsart University, Thailand
Mex Memisang Peki
PNG University of Technology
Pham Minh Toai
Vietnam Forestry University
Kim Soben
Royal University of Agriculture, Cambodia
Nguyen The Dzung
Vietnam Forestry University
Pheng Sokline
Royal University of Phnom Penh,
Cambodia
Nguyen Hai Hoa
Vietnam Forestry University
Seak Sophat
Royal University of Phnom Penh,
Cambodia
Le Xuan Truong
Vietnam Forestry University
Choeun Kimseng
Royal University of Phnom Penh,
Cambodia
Phan Thi Quynh Nga
Vinh University, Vietnam
Rajendra Shrestha
Asian Institute of Technology, Thailand
Erin Swails
Winrock International
Ismail Parlan
FRIM Malaysia
Sarah Walker
Winrock International
Nur Hajar Zamah Shari
FRIM Malaysia
Sandra Brown
Winrock International
Samsudin Musa
FRIM Malaysia
Karen Vandecar
US Forest Service
Ly Thi Minh Hai
USAID LEAF Vietnam
Geoffrey Blate
US Forest Service
David Ganz
USAID LEAF Bangkok
Chi Pham
USAID LEAF Bangkok
I
II
III
OVERVIEW: CLIMATE CHANGE AND FORESTCARBON
1.1
Overview: Tropical Forests and Climate Change
1.2
Tropical forests, the global carbon cycle and climate change
1.3
Role of forest carbon and forests in global climate negotiations
1.4
Theoretical and practical challenges for forest-based climate mitigation
FOREST CARBON STOCKS AND CHANGE
2.1
Overview of forest carbon pools (stocks)
2.2
Land use, land use change, and forestry (LULUCF) and CO2 emissions and sequestration
2.3
Overview of Forest Carbon Measurement and Monitoring
2.4
IPCC approach for carbon measurement and monitoring
2.5
Reference levels – Monitoring against a baseline (forest area, forest emissions)
2.6
Establishing Lam Dong’s Reference Level for Provincial REDD+ Action Plan : A Case Study
CARBON MEASUREMENT AND MONITORING DESIGN
3.1
IV
V
Considerations in developing a monitoring system
CARBON STOCK MEASUREMENT METHODS
4.1
Forest Carbon Measurement and Monitoring
4.2
Design of field sampling framework for carbon stock inventory
4.3
Plot Design for Carbon Stock Inventory
4.4
Forest Carbon Field Measurement Methods
4.5
Carbon Stock Calculations and Available Tools
4.6
Creating Activity Data and Emission Factors
4.7
Carbon Emission from Selective Logging
4.8
Monitoring non-CO2 GHGs
NATIONAL SCALE MONITORING SYSTEMS
By the end of this session, learners will be able to:

Determine minimum plots to be established in each
forest type and pool

Design plot shape and size for each forest type and
pool

Identify appropriate methods to locate sample plots.
Lecture (45 minutes)

Overview

Sampling in forest inventory

Plot shape and size

Number and distribution of plots

Three Activities ( 2 at 5 minutes each one 25 minutes)

References

Plot design for tree, non-tree
vegetation, litter, soil, and
deadwood pools

Impact of plot number; plot,
shape and size on carbon
stock inventory

Selection of plot locations

It is impractical to take measurements that would result in an
estimate of the entire mass of the vegetation in a landscape
The goal of a
carbon inventory
is to estimate the
mass of the
vegetation and soil
to determine
carbon stocks

Sampling in carbon inventories
involves repeated measurements
to estimate the mass of
vegetation in a known,
standardized area (“sampling
unit” or “plot”)

Standardized plots are used to
ensure all samples represent the
same area
PLOTS: WHICH
SHAPE? HOW
BIG? HOW
MANY?
1
VEGETATION TYPE
2
ACCURACY
3
PRECISION
4
TIME (COST) OF MEASUREMENT
5
What do accuracy and precision mean?

Accuracy of a relative measure of the
exactness of the value of an inferred
variable for a population.

Precision is the closeness of
agreement between independent
results of measurements obtained
under stipulated conditions.

Various plot shapes can be used (circles, rectangles)
Shape: maximize efficiency and minimize measurement
error
 Need to decide whether tree is inside or outside of plot
 Circles: lowest edge – to – area ratio

Area of shape must be known and measured

So if square/rectangle used-must ensure 90°angles

Plot sides must be straight
Answers:
3m
9m
0m
4m
12m
4m

Plot size needs to balance

Rule of thumb: area large enough to sample ~10
individuals
Relatively Uniform Forest – e.g. Plantation

Trees are similar size range/homogeneous

Planting lines are visible
Relatively Uniform Forest – e.g. Plantation

Need to avoid setting square plots up along planting lines

Must choose appropriate size plot
Non-uniform sized trees - Natural Forest
 In most natural forests:


Large trees - few, very spread out

Small trees – many, close together
Plot design needs to adequately sample all sizes
Non-uniform sized trees - Natural Forest



Plots can be single fixed size or “nested” plots containing
smaller sub-units
Nested plots are efficient for forests with trees of different
sizes
Rule of thumb – each nested plot should include ~10 trees

Non-uniform sized trees - Natural Forest
Source: Winrock International, 2012
Standing dead wood, palms, bamboo, lianas, shrubs

As for trees, large
single size or nested
plots are used to
capture spatial
variability of carbon
stocks in large-scale
non-tree vegetation

For efficiency, these
vegetation types may
be measured in the
same plots as trees

On the following slide there are two plot designs for sampling
shrubs. Pick which one is better and write down why you
think it is a better design
Which plot size (A or B) would best sample the
shrubs in this landscape?
A
B
B is correct because plots should be big enough to
capture > 10 individuals
Herbaceous vegetation, litter, lying
dead wood, shrubs

Smaller plots are adequate and
efficient for capturing spatial
variability in small scale
vegetation, litter, soil.

Herbaceous vegetation
and litter is sampled in
small circular or square
plots

Typically plot size of ~1m2
is adequate and efficient

Fallen dead wood is typically
measured using a transect

A transect length of 100 m per
plot is usually adequate and
efficient
Both these photos need sources or to be replaced

Soil carbon is measured by collecting core samples from
the top 30 cm of soil


Depends on targeted level of precision in inventory,
common targets are:

±10% of the mean at 90% CI

±5% of the mean at 95% CI
The more variable the carbon stocks in pool or stratum,
the more plots needed to achieve the targeted level of
precision in carbon stock estimates

Calculate number of plots needed to achieve targeted
level of precision in estimate of dominant carbon pool
(i.e. trees) by the following equation:

Smaller plot size  greater variation in carbon stocks
between plots

Rule of thumb: Number of non-tree plots proportional to
number of tree plots

May result in large variance, but overall carbon in non-tree
pools is small compared to tree pool

To maintain statistical rigor, plots must be located
without bias

Location of tree plots should either be random or located
using a fixed grid

Divide students in to groups and provide each group will
with a forest cover map with a study area delineated

Have the students choose locations for plots in the study
area by two different methods (random and systematic
distribution) ( If available you could provide preliminary
data and have students calculate the correct number of
plots as well)

Total time: 25 minutes

For non-tree carbon pools, sub-plot locations may be
based on the location of the tree plot

Sub-plots should be outside of tree plots

Measurements at subsequent inventories should occur in
new location


Temporary

For one-time measurements and preliminary data

All non-tree pools
Permanent

For measurements repeated over time

Statistically more efficient for measuring incremental
growth of trees
Nested
plots: Tree
pool,
standing
dead wood,
palms,
lianas,
bamboo,
large shrubs
Transect: Lying dead wood
Sub plots:
 Herbaceous vegetation and small shrubs (~4)
 Litter (~4)
 Soil (~4 carbon content + ~2 bulk density)

The size and shape of the sample plots is a trade-off
between accuracy, precision, time, and cost for
measurement;

The most appropriate size and shape may also be
dependent on the vegetation type found in the sampling
area;

Nested plots are efficient for forests with trees of different
sizes;

Plots can be distributed randomly or systematically.
1. Diaz, D. and M. Delaney, 2011. Carbon stock assessment guidance
2. Goslee, K. and S. Petrova, 2012. LEAF Technical Training on Reference Level
Development. USAID and LEAF;
3. Phuong VT et al, 2012. Guidelines on destructive measurement for forest
biomass estimation (For Technical Staff Use). UN-REDD Vietnam Program
4. Walker, S. et al, 2012. Standard operation procedures for terrestial carbon
measurement. USAID, LEAF and WINROCK international;
5. Pearson, TRH, SL Brown, RA Birdsey, 2007. Measurement Guidelines for
the Sequestration of Forest Carbon, USDA
6. Pearson, T., S. Walker and S. Brown, 2005. Sourcebook for land use, land
use change and forestry project. BioCF and WINROCK International.