Improvements to the representation of forest land in the UK GHG inventory and implications for KP accounting

Paul Henshall, Robert Matthews (and Gwen Buys, CEH)

Forest Mensuration, Modelling and Forecasting

Centre for Sustainable Forestry and Climate Change

JRC LULUCF Workshop, Arona, 6th May 2014

12/05/2014 2

• Input data on forests

• Modelling

• Representation of management

• How the results have changed

• Implications for KP accounting (CP1 and CP2).

A description of improvements

to the GHG calculations for forest land

What we will aim to cover

12/05/2014 3

Input data: forest area and species

0.0

100.0

200.0

300.0

400.0

500.0

600.0

700.0

800.0

900.0

Nor

way

spr

uce

Sitk

a sp

ruce

Cor

sica

n pine

Sco

ts pin

e

Lodg

epole

pine

Eur

opea

n larc

h

Japa

nese

larc

h

Dou

glas

fir

Gra

nd fir

Nob

le fir

Wes

tern

red

ceda

r

Wes

tern

hem

lock

Oak

Bee

ch

Syc

amor

e, a

sh, b

irch

Not

hofa

gus

Pop

lar

Are

a (

kh

a)

Post-1920 planting Area

Pre-1921 planting Area

Previously:

• Post-1920 forest areas (pre-1921 areas assumed to be ‘in balance’)

• All conifer area represented by Sitka spruce

• All broadleaf area represented by beech.

Updated:

(based on NFI from 1990s)

12/05/2014 4

Input data: growth rates

Previously:

• Sitka spruce all with average potential growth rate of 12 m3 ha-1 yr-1 apart from Northern Ireland (14)

• Beech all with average potential growth rate of 6 m3 ha-1 yr-1.

0

50

100

150

200

250

2 4 6 8 10 12 14 16 18 20 22 24

Yield class (m3 ha-1 yr-1)

Are

a (

kh

a)

Western hemlock

Western red cedar

Noble fir

Grand fir

Douglas fir

Japanese larch

European larch

Lodgepole pine

Scots pine

Corsican pine

Sitka spruce

Norway spruce

0

50

100

150

200

250

300

350

400

450

500

2 4 6 8 10 12 14 16 18 20 22 24

Yield class (m3 ha-1 yr-1)

Are

a (

kh

a)

Poplar

Nothofagus

Sycamore, ash, birch

Beech

Oak

Updated:

(based on records for Public Forest Estate)

12/05/2014 5

Input data: age distribution

Previously:

• Effectively the new planting projected forward (no independent data, example shown for Wales)

0

1

2

3

4

5

6

1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020

Year of planting

Are

a (

kh

a)

Conifer

Broadleaf

12/05/2014 6

Analysis of forest age distribution

0

5

10

15

20

25

30

35

40

45

50

55

60

65

0 50 100 150 200

Age class

Are

a (

kh

a)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

0 50 100 150 200

Age class

Are

a (

kh

a)

Updated: Reconciling new planting with NFI age distributions

Conifer Broadleaf

0

1

2

3

4

5

6

1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020

Year of planting

Are

a (

kh

a)

Conifer

BroadleafFC data on

“new planting”

12/05/2014 7

Which gives this profile for new planting …

0

5

10

15

20

25

30

35

401

50

0

15

11

15

22

15

33

15

44

15

55

15

66

15

77

15

88

15

99

16

10

16

21

16

32

16

43

16

54

16

65

16

76

16

87

16

98

17

09

17

20

17

31

17

42

17

53

17

64

17

75

17

86

17

97

18

08

18

19

18

30

18

41

18

52

18

63

18

74

18

85

18

96

19

07

19

18

19

29

19

40

19

51

19

62

19

73

19

84

19

95

20

06

Are

a k

ha

Year

UK - CFlow Planting InputsBroadleaf

Conifer - org soil

Conifer - min soil

0

5

10

15

20

25

30

35

40

15

00

15

11

15

22

15

33

15

44

15

55

15

66

15

77

15

88

15

99

16

10

16

21

16

32

16

43

16

54

16

65

16

76

16

87

16

98

17

09

17

20

17

31

17

42

17

53

17

64

17

75

17

86

17

97

18

08

18

19

18

30

18

41

18

52

18

63

18

74

18

85

18

96

19

07

19

18

19

29

19

40

19

51

19

62

19

73

19

84

19

95

20

06

Are

a k

ha

Year

UK - CARBINE Planting InputsBroadleaf

Conifer - org soil

Conifer - min soil

Updated:

Previously:

12/05/2014 8

Input data: rotations

Previously:

• Sitka spruce all age 59 years (minor variations), all beech 91 years.

0

50

100

150

200

250

300

350

400

450

500

Less than

30

30-39 40-49 50-59 60-69 70-99 100-119 120-149 150+

Rotation (years)

Are

a (

kh

a)

Poplar

Nothofagus

Sycamore, ash, birch

Beech

Oak

Western hemlock

Western red cedar

Noble fir

Grand fir

Douglas fir

Japanese larch

European larch

Lodgepole pine

Scots pine

Corsican pine

Sitka spruce

Norway spruce

Updated:

• Based on growth rates and reconciliation of ages with new planting.

12/05/2014 9

• Species-specific:

• Forest growth functions (published FC models)

• Wood density estimates (published source)

• Biomass expansion factors for branches, foliage and roots (new analysis of available data)

• Allocation factors for harvested wood.

Refined modelling: CARBINE

Some examples …

12/05/2014 10

Assumptions about management

Previously:

• All in management, thinned and clearfell.

Updated:

(based on records for Public Forest Estate and Woodland Grant Scheme records provided by country representatives)

0

100

200

300

400

500

600

700

Nor

way

spr

uce

Sitk

a sp

ruce

Cor

sica

n pine

Sco

ts pin

e

Lodg

epole

pine

Eur

opea

n larc

h

Japa

nese

larc

h

Dou

glas

fir

Gra

nd fir

Nob

le fir

Wes

tern

red

ceda

r

Wes

tern

hem

lock

Oak

Bee

ch

Syc

amor

e, a

sh, b

irch

Not

hofa

gus

Pop

lar

Are

a (

kh

a)

Not in management

Continuous

Thin/fell

No thin/fell

12/05/2014 11

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

1975 1980 1985 1990 1995 2000 2005 2010

Year

Vo

lum

e p

rod

uc

tio

n (

millio

n m

3)

Predicted

Reported

Management/production ‘reality check’

About „double‟ About +30%

Example for Wales

12/05/2014 12

• Gross:net area

• In production:not in production (private sector)

• Thin:no-thin (private sector, historical for all)

• Production ‘held back’ in private sector?

• Growth rate assumptions (private sector)

• New planting:pre-1920 forests (FC:private sector)

• Under-reporting in production statistics

Why is predicted production higher than reported?

Management/production ‘reality check’

12/05/2014 13

Input data (again): deforestation

Previously:

Updated:

Results

12/05/2014 15

Afforestation

-3.5

-3

-2.5

-2

-1.5

-1

-0.5

0

0.5

1985 1990 1995 2000 2005 2010 2015 2020

Year

Em

issio

n/r

em

oval

(+/-

) M

tCO

2

Previously

Updated

12/05/2014 16

Deforestation

0

0.2

0.4

0.6

0.8

1

1.2

1990 1995 2000 2005 2010 2015 2020

Year

Em

issio

n/r

em

oval

(+/-

) M

tCO

2 Previously

Updated

12/05/2014 17

Forest Management

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

1990 1995 2000 2005 2010 2015 2020

Year

Em

issio

n/r

em

oval

(+/-

) M

tCO

2 Previously

Updated

Implications for KP accounting

12/05/2014 19

• FM removals now much bigger than previously reported

• However, FM removals are capped in CP1.

• The cap is related to the level of FM removals in the base year, but was fixed at that level at the time the details were negotiated, and doesn’t get changed now

• Therefore, removals accounted due to FM will not change in CP1

• Accounted removals due to afforestation lower than previously suggested (down by about 10% or 300 ktCO2, yr

-1)

• Accounted emissions due to deforestation higher than previously suggested (up by about 85% or 500 k tCO2, yr

-1).

First commitment period

12/05/2014 20

• At present, the main issue is the implication of the changed FM results for the FMRL.

• FM removals now much bigger than previously reported, and projected level now very likely to be significantly different to when the FMRL was originally set.

• It is likely that a Technical Correction to the FMRL will be required…

Second commitment period

12/05/2014 21

Forest Management

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

1990 1995 2000 2005 2010 2015 2020

Year

Em

issio

n/r

em

oval

(+/-

) M

tCO

2 Previously

Updated

Original

projection

New

projection

Basis for calculating

Technical Correction?

12/05/2014 22

• Main challenge: There is a new National Forest Inventory (NFI) for GB

• The base year for the new NFI is ~2013

• This NFI includes directly and transparently-calculated estimates of tree carbon stocks

• We will need to decide whether to refer directly to these carbon stock estimates or use them for validation of model results

• There could be implications for back-calculation (i.e. for period 1990 to 2012)

• Several other challenges: Integrating calculations for non-CO2 GHGs; soil GHG emissions/removals related to AR and particularly D; better reconciliation with actual wood production?

Future challenges

Thank you