FEM TIPS F

1
(No Transcript)
2
Climate change mitigation related to Tanzanian
forests Key factors for analysis and research
prioritizing
Ole Hofstad
3
Organisation of the presentation

• Mitigating climate change through REDD
• Monitoring
• Carbon accounting
• PES mechanisms
• Land-use change modelling
• Policy measures within the forest sector
• Other policies

4
Carbon stocks
5
GHG emissions
6
The importance of degradation
7
Monitoring forest ecosystems

• area and density
• technologies
• sampling
• accuracy
• frequency
• costs

8
The monitoring problem may be considered as two
separate components

1. estimating areas of different vegetation types
   (e.g. forest, woodland, savannah, cropland,
   etc.), and
2. estimating the average biomass density (tons/ha)
   in each vegetation type.

Cropland and burned bush in Northern
Mozambique (Photo E. H. Hansen)
9
Area estimates

• Areas may be measured on the ground, either by
  triangulation using surveying equipment, or GPS.
  These methods are both time consuming and
  expensive and best suited for small areas with
  very high precision requirements.
• Areas may be measured on aerial photographs. This
  is expensive if aerial photography is ordered for
  this particular use alone.
• Areas may be measured on satellite images based
  on reflected sunlight. Classification of
  vegetation types may be assisted by competent
  personnel, or be made unassisted by computer.
  Using satellite images is the preferred method in
  most modern applications for large areas of low
  unit value.

10
(No Transcript)
11
Biomass measurements

• Biomass density may be measured on temporary or
  permanent sample plots in the field. Trees (and
  bushes) are measured in various ways, e.g. stem
  diameter, height, crown diameter, etc. These
  measurements are transformed by allometric
  functions into estimates of volume or weight of
  individual trees or bushes.
• Biomass density may be estimated on the basis of
  crown cover measured on aerial photos.
• Biomass estimates may be based on data collected
  by the use of light emitted from an airborne or
  satellite laser, or
• from an airborne or satellite radar.

The three latter methods (photo, laser, radar)
require some sample plots on the ground where
trees are measured manually. Such data is
necessary in order to calibrate the remote
sensing data.
12
Combining area estimates with estiamated biomass
density
13
Air-borne laser
14
Remote sensing of biomass density in forests
Points of reflection distributed in space
15
Sampling

• Stratified sampling
• Sampling percentage
• Permanent plots
• Temporary plots

• Stratification
• Forest types rain forest (flooded or not),
  montane forest, seasonal green forest, open
  forest, shrub, savanna, etc., cropland, grazing
  land
• Agro-ecological zones, regions, districts
• Biomass density
• The smaller the reporting unit, the larger
  sampling percentage is required to give precise
  estimates

16
Proposed laser project

• 1. If FRA2010/NFI decides to measure ground plots
  either from FRA2010 tiles or along the lines
  formed by FRA2010 tiles (see map), we should
  consider offering to fly LiDAR along these lines
  of FRA2010 tiles in all, or parts of, Tanzania.
  If we fly all over Tanzania, it will imply flying
  a total distance of ca 9000 stripe-km, which will
  give a systematic sample of laser data for all of
  Tanzania. Calibrated with field data from below
  the flight corridors, one would be able to give a
  national biomass estimate for the whole of
  Tanzania in less than one year (given that field
  data are measured during the same period). We may
  even be able to break the estimate down into
  regional partial estimates.
• 2. In addition we should select one of the three
  "ecosystems" as an object for detailed studies,
  where we either fly wall-to-wall with LiDAR or
  fly stripes very close (as proposed in Brazil) in
  an area of 5-10,000 km2. In this area we must
  establish a set of separate sample plots on the
  ground. Observations from these plots will be
  used to calibrate LiDAR measurements of biomass.
  This set of data will serve two purposes
• 2a GEO/FCT sites
• 2b detailed studies of design of laser-mapping
  of biomass through sampling
• 2c ground validation of SAR-study. If we
  choose tropical rain forest as a case, this will
  be complementary to Brazil since we may find
  higher biomass density than in Amazonia.

17
Precision
Relationship between accuracy (Sm) and number of
plots (n) according to different patterns of
spatial variation Sm Standard error CV
Coefficient of variation

• For the REDD-activities in Tanzania, where a lot
  of different inventories will be performed, it
  will be of crucial importance to gain basic
  knowledge on patterns of spatial variation for
  biomass ha-1 (or volume or basal area ha-1) under
  different forest conditions and plot designs. A
  research project to approach these challenges
  could be performed along the following lines
•  
• Systematic review of previously performed
  inventories with respect to spatial variation
• Undertake inventories in selected study areas
  covering important vegetation types and inventory
  designs
• Perform theoretical inventory simulations in
  order to select optimal inventory strategies
  under different conditions and requirements

18
Frequency

• How often will new area estimates be presented?
• How often shall biomass estimates be updated?
• Rotation on permanent sample plots
• Repeated flights airplane or satellite (with
  camera, laser, or radar)
• Higher frequency, higher costs

19
CARBON IN FOREST

• IPCC Guidelines
• Three hierarchical tiers of methods that range
  from
• default data
• simple equations
• to the use of country-specific data and models to
  accommodate national circumstances.
• It is good practice to use methods that provide
  the highest levels of certainty, while using
  available resources as efficiently as possible.
• Combination of tiers can be used.

• Living biomass
• Trees, bushes, herbs and grass
• Above ground
• Roots
• Ded wood
• Logging residues
• Ded branches, roots and more
• Soil

20
LIVING BIOMASS

• Biomass expansion factor (BEF/BF)
• E.g. IPCC default value 0.44 tons Dry Matter /
  m3 fresh volume
• Biomass equation
• Allometric functions for whole trees or fractions
  like stem, branches and roots.
• E.g. Biomass above ground
• B 0.3623 dbh1.382 h0.64
• B - 4.22412 0.56 dbh2
• Field measurements and laboratory measurement of
  wood density are required.

21
Land-use changes to achieve REDD
22
Leakage
23
Global trade in forest products
Main trade flows of tropical roundwood 2007.
(million m3) Buongiorno, J., D. Tomberlin, J.
Turner, D. Zhang, S. Zhu 2003. The Global Forest
Products Model Structure, Estimation, and
Applications.
24
Source Jayant Sathaye, Lawrence Berkeley
National Laboratory, California
25
(No Transcript)
26
(No Transcript)
27
Land-use model
28
Land-use models at village or watershed level

• Namaalwa, J., P. L. Sankhayan O. Hofstad 2007.
  A dynamic bio-economic model for analyzing
  deforestation and degradation An application to
  woodlands in Uganda. Forest Policy and Economics,
  9 (5)479-95.
• Sankhayan, P. L., M. Gera O. Hofstad. 2007.
  Analysis of vegetative degradation at a village
  level in the Indian Himalayan state of Uttarkhand
  a systems approach by using dynamic linear
  programming bio-economic model. Int. J. Ecology
  and Environmental Sciences 33(2-3) 183-95.
• Hofstad, O. 2005. Review of biomass and volume
  functions for individual trees and shrubs in
  southeast Africa. J. Tropical Forest Science,
  17(1)413-8.
• Namaalwa, J., W. Gombya-Ssembajjwe O. Hofstad
  2001. The profitability of deforestation of
  private forests in Uganda. International Forestry
  Review 3 299-306.
• Sankhayan, P. L. O. Hofstad 2001. A
  village-level economic model of land clearing,
  grazing, and wood harvesting for sub-Saharan
  Africa with a case study in southern Senegal.
  Ecological Economics 38 423-40.
• Hofstad, O. P. L. Sankhayan 1999. Prices of
  charcoal at various distances from Kampala and
  Dar es Salaam 1994 - 1999. Southern African
  Forestry Journal, 18615-18.
• Hofstad, O. 1997. Woodland deforestation by
  charcoal supply to Dar es Salaam. J.of
  Environmental Economics and Management, 3317-32.
  

29
Tanzanian land-use and forest sector trade models

• Kaoneka, A.R.S. 1993. Land use Planning and
  quantitative modelling in Tanzania with
  particular reference to agriculture and
  deforestation some theoretical aspects and a
  case study from the West Usambara mountains.
  Dr.Scient. Thesis, Agriculture University of
  Norway, Aas.
• Monela, G. S. 1995. Tropical rainforest
  deforestation, biodiversity benefits and
  sustainable land use Analytical of economic and
  ecological aspects related to the Nguru
  Mountains, Tanzania. Dr. Scient. Thesis,
  Department of Forestry, Agricultural University
  of Norway.
• Ngaga, Y.M. 1998 Analysis of production and trade
  in forestry products of Tanzania. Dr.Scient.
  Thesis, Agriculture University of Norway, Aas.
• Makundi, W. R. 2001. Potential and Cost of Carbon
  Sequestration in the Tanzanian Forest Sector.
  Mitigation and Adaptation Strategies for Global
  Change, 6(3-4)335-53.
• Ngaga, Y. M. B. Solberg 2007. Assessing the
  Suitability of Partial Equilibrium Modelling in
  Analyzing the Forest Sector of Developing
  Countries Methodological Aspects with Reference
  to Tanzania. Tanzania Journal of Forestry and
  Nature Conservation, 7611-27.
• Monela, G. C. J. M. Abdallah 2007. External
  policy impacts on Miombo forest development in
  Tanzania. In Dubé, Y. C. F. Schmithüsen
  (eds.) Cross-sectoral policy developments in
  forestry.
• Monela, G. C. B. Solberg 2008. Deforestation
  and agricultural expansion in Mhonda area,
  Tanzania. In Palo, M. H. Vanhanen (eds.)
  World forests from deforestation to transition?

30
Policy measures

• General policies
• Good governance (legal system, transparency,
  corruption)
• Energy
• Agriculture
• Transport
• Sector specific measures
• PES (monitoring, verification)
• Projects (administrative costs, foreign
  assistance)
• Land ownership and user rights
• Cost effectiveness and efficiency (Cost-Benefit)

31
Schematic view of a REDD PES system
32
(No Transcript)