A stepwise approach to reference levels

A stepwise approach to reference levels

Louis Verchot

THINKING beyond the canopy

•  Activity data – types of deforestation and forest degradation •  Emission factors – carbon loss per unit area for a specific

activity •  Drivers – to describe how much DD are caused by each

specific change activity •  Data on existing national monitoring capacities

Business as usual and national capacity


Ac#vity data

Approach 1 Approach 2 Approach 3 Data on forest change (or emissions) following IPCC approaches

TOTAL LAND-‐USE AREA, NO DATA ON CONVERSIONS BETWEEN LAND USES Example: FAO FRA data

TOTAL LAND-‐USE AREA, INCLUDING CHANGES BETWEEN CATEGORIES Example: Na:onal level data on gross forest changes through a change matrix (i.e. deforesta:on vs. reforesta:on), ideally disaggregated by administra:ve regions

SPATIALLY-‐EXPLICIT LAND-‐USE CONVERSION DATA Example: data from remote sensing

Table 9: Activity data on the national level can be estimated from the different approaches as suggested by the IPCC GPG:


Three levels of emission factors •  Tier 1 methods are designed to be the simplest to use,

for which equations and default parameter values (e.g., emission and stock change factors) are provided by IPCC Guildelines.

•  Tier 2 can use the same methodological approach as Tier 1 but applies emission and stock change factors that are based on country- or region-specific data

•  Tier 3, higher order methods are used, including models and inventory measurement systems tailored to address national circumstances, repeated over time, and driven by high-resolution activity data and disaggregated at sub-national level.


Deforestation/degradation drivers for each continent

-‐39%

-‐35%

-‐13%

-‐11%

-‐2%

-‐57% -‐36%

-‐2% -‐4% -‐1%

-‐41%

-‐37%

-‐7%

-‐10% -‐7%

26%

62%

4% 8%

70%

9%

17%

4%

67%

20%

6% 7%

Forest degrada#on driver Deforesta#on driver

AMERICA AFRICA ASIA

Deforesta#on

Degrada#on


Changes of Deforestation Drivers: Important for assessing historical deforestation

Using national data from 46 countries: REDD-related data and publications

Time

Phase1 Pre

Transition

Phase4 Post

Transition

Phase2 Early

Transition

Phase3 Late

Transition

Fore

st C

over

(%)


Deforestation Drivers

n  Agriculture (commercial) is 45%, agriculture (local/subsistence) 38%, mining 7%, infrastructure 8%, urban expansion 3% and only agriculture make up 83% of total

n  Ratio of mining is decreasing and urban expansion is relatively increasing over time

Deforested-‐area ra:o of deforesta:on drivers

Deforested area

0

100

200

300

400

500

600

700

pre early late post

Urban expansion

Infrastructure

Mining

Agriculture (local-‐slash & burn) Agriculture (commercial)

km2

0%

20%

40%

60%

80%

100%

pre early late post

Distribu:on of 46 countries -‐ Pre: 7, early: 23, late: 12, post: 4

(subsistence)

Criteria for comparing country circumstances and strategies

Criteria for comparing country circumstances and strategies


RLs using regression models –  Simple, easy to understand and test new variables –  But, data demanding –  Predic:ng deforesta:on in a period: Pt – Pt+1, based on deforesta:on in the previous period Pt-‐1 – Pt and a set of other factors (observed at :me t).

–  Using structure (coefficients) from the es:mated regression equa:on to predict deforesta:on in period Pt+1 – Pt+2, based on observed values at :me t+1

10

2000 2009 2005 2004 2010

Es:mated/Predicted deforesta:on Historical deforesta:on

Predic#ve model, based on structure from regression model

Regression model

Tier 1 case for 4 countries using FAO FRA data

0

500

1,000

1,500

2,000

2,500

3,000

3,500

1985 1990 1995 2000 2005 2010 2015 2020 2025

Fore

st C

sto

ck (

Mt)

Year

Cameroon

0 2,000 4,000 6,000 8,000

10,000 12,000 14,000 16,000 18,000

1985 1990 1995 2000 2005 2010 2015 2020 2025

Fore

st C

sto

ck (

Mt)

Year

Indonesia

0

300

600

900

1,200

1,500

1985 1990 1995 2000 2005 2010 2015 2020 2025

Fore

st C

sto

ck (

Mt)

Year

Vietnam

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

1985 1990 1995 2000 2005 2010 2015 2020 2025

Fore

st C

sto

ck (

Mt)

Year

Brazil


Step 2: Brazil Predict deforesta#on rates for legal Amazon 2005-‐ 2009

12

Category Regression coefficient

Deforesta#on rate (2000-‐2004) 0.395 Trend variable -‐0.136 -‐0.145 Deforesta#on dummy -‐0.373 -‐0.773 Forest stock 2.18 4.756 Forest stock squared -‐1.8 -‐3.826 Log per capita GDP -‐0.034 -‐0.13 Agric GDP (%GDP) 0.28 0.28 Popula#on density 0.081 -‐0.81 Road denisty 0.039 0.076

R2 0.831 0.789 N 3595 3595


Step 2: Vietnam Predict deforesta#on rates 2005-‐ 2009

13

Category Regression coefficient

Deforesta#on rate (2000-‐2004) 01.464 Trend variable -‐0.006 0.003 Deforesta#on dummy -‐0.011 -‐0.031 Forest stock 0.067 0.260 Forest stock squared -‐0.189 -‐0.463 Popula#on density -‐1.177 1.036 Road denisty 0.004 -‐0.001

R2 0.515 0.052 N 301 301


Conclusions •  Historical def. is key to predict future deforesta:on

–  Coefficients below one → simple extrapola:on can be misleading

•  Some evidence of forest transi:on (FT) hypothesis –  Robustness of FT depends on the measure of forest stock

•  FT supported when forest stock is measured rela:ve to total land area, otherwise mixed results emerge

•  Other na:onal circumstances have contradictory effects

•  Contradictory rela:onships may be linked to data quality and interrela:ons of econ. & ins:tu:ons differ

14

MRV capacity gap analysis

MRV capacity gap in rela:on to the net change in total forest area between 2005 and 2010 (FAO FRA)

-‐3000

-‐2000

-‐1000

0

1000

2000

3000

Very large Large Medium Small Very small

Net cha

nge in fo

rest area since 1990

(1000h

a)

Capacity gap


§  53% use site specific biomass equations

§  24% had methods for belowgound C

§  41% had methods for dead wood and litter

§  Most projects will use IPCC defaults for soil-C

We surveyed 17 REDD+ demonstration projects

Thank you

Education

A stepwise approach to reference levels