Indicator 2.13. Criterion 2. Maintenance of Productive Capacity of Forest Ecosystems

advertisement
Criterion 2. Maintenance of Productive Capacity of Forest Ecosystems
National Report on Sustainable Forests—2010
Indicator 2.13.
Annual Harvest of Wood Products by Volume and as a Percentage of Net Growth or
Sustained Yield
Based on site productivity data measured during field inventories,
an estimate can be made of the productive potential of U.S. forests and how they relate to the current situation (fig. 13-4). This
Figure 13-1. Growth and removals of growing stock on
timber land by owner group and region, 1952–2006.
Public
25
Private
20
15
10
East
Net growth
East
Removals
West
Net growth
Region
2006
1976
1952
2006
1976
1952
2006
1976
0
1952
5
2006
Growth has exceeded removals on U.S. timber lands for
several decades (fig. 13-1), although the area of timber land
has remained relatively stable. The result has been a substantial
increase in the volume of growing stock on U.S. timber lands.
In the past decade, growth continued to exceed removals for
both publicly and privately owned timber lands in the East
(North and South Regions) and West (Rocky Mountain, Pacific
Coast and Alaska Regions). Trends in growth on timber land
since 1952 are attributable to several factors. In general,
positive growth trends reflect regrowth and maturation of
forests on lands that had been harvested before 1952. Investments in fire protection, landowner education, and silviculture
are also reflected in the trends. Changes in harvest patterns in
the 1990s resulted in growth and removals shifts by ownership
and region. Historically, most harvesting occurred on private
timber lands in the East and recent data show a further shift of
removals from public timber land in the West to private timber
land in the East as policies to reduce harvesting on public lands
in the West were implemented. Thus, growth has been exceeding removals by a wider margin in the West while the gap has
Since 1952, overall conifer volume has increased 23 percent
and broadleaf volume has increased 118 percent. The lower
percentage for conifers is reflective of higher demand for wood
products from these species. Growth exceeding removals in all
regions for both conifers and broadleaves is reflective of this
trend (figures 13-2 and 13-3).
1976
What does the indicator show?
Although this situation is significant, recent major planting
of conifers in the South are rapidly becoming of commercial
size and are expected to improve the situation in that region.
Current growth measures in the South may not fully reflect
anticipated growth on these planted forests. Currently, 91 per­cent of U.S. wood output is produced on private lands.
1952
This indicator compares net growth with wood harvest (removals) for products on timber land. This comparison is a frequently used method of assessing whether or not wood harvesting is
reducing the total volume of trees on forest available for wood
production. Growth is the net annual increase in the volume of
growing stock between inventories after accounting for effects
of mortality, but before accounting for the effects of harvest.
Removals are a measure of the average annual volume of growing stock trees harvested between inventories. Timber land is
assumed to be the subset of forest land on which some level of
wood harvesting is potentially allowed. So long as growth (net
of mortality) exceeds removals, the volume of trees on timber
land is considered sustainable. This measure, however, conveys
no information about quality, biodiversity, other attributes of
ecology, or management objectives, and it should be considered in conjunction with other indicators as part of an overall
analysis of objectives for forest ecosystem sustainability.
been decreasing in the East. Interpreting growth trend data in
the West, particularly on public land, can be complicated by the
reductions in harvesting and set asides of large areas of public
timber land into reserves since the mid-1970s, thus, apparent
declines may be an artifact of this situation.
Cubic feet (billions)
What is the indicator and why is it important?
West
Removals
Source: USDA Forest Service, Forest Inventory and Analysis
Last Updated June 2011 9
Net growth
Removals
1
Public
Public
25
25
Private
Private
National Report on Sustainable Forests—2010
20
20
2006
2006
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1976
1976
Figure 13-2. Net growth and removals of conifers in the
United States,
1952–2006.
East
East
WestNet growth West Removals
9NetEast
growth
9
Net growth
Net growthRemovals
East
West Removals
Removals
NetWest
growth
Removals
Removals Region
Net growth
Region
8
8
Cubic
Cubic
feet
feet
(billions)
(billions)
0
0
120 60
120 60
Potential growth
Potential growth
Current growth
Current growth
Current removals
Current removals
Current removals
Current removals
Cubic
feet
perper
acre
perper
year
Cubic
feet
acre
year
100 40
100 40
80 20
80 20
North
North
South
South
Rocky
Pacific
Rocky
Pacific
Mountain
Coast
Mountain
Coast
40
Region
40
Regionand Analysis
Source: USDA Forest Service, Forest Inventory
U.S.
U.S.
Total
Total
20
20
0
North June 2011
South
0 Updated
Last
North
South
Rocky
Mountain
Rocky
Mountain
Region
Region
Pacific
Coast
Pacific
Coast
U.S.
Total
U.S.
Total
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1952
1952
1976
1976
2006
2006
1976
1976
South
South
Rocky
Rocky
Mountain
Mountain
Region
Region
Rocky
Mountain
Rocky
Mountain
Region
Pacific
Pacific
Coast
Coast
Pacific
Coast
Pacific
Coast
Alaska
Alaska
2006
2006
1976
1976
South
South
Alaska
Alaska
East
East
6,000
Region
6,000 Forest Service, Forest Inventory and Analysis
Source: USDA
Cubic
Cubic
feet
feet
(billions)
(billions)
60
60
80
80
North
North
North
North
West
West
4,000
4,000
Figure 13-5. Removals of growing stock for major forest products,
1952, 1976, and 2006.
2,000
Cubic
feet
(billions)
Cubic
feet
(billions)
Cubic
Cubic
feet
feet
per
per
acre
acre
per
per
year
year
100
100
Mountain
Coast
Potential growth Region
Current growth
Potential growth Region
Current growth
1952
1952
2006
2006
1976
1976
0
0
Figure
13-4. Potential
and
current Pacific
growth and
removals
North
South
Rocky
Alaska
Coast
North land by
South
Rocky
Alaska
on timber
region,Mountain
2006. Pacific
120
120
1952
1952
2
2
1952
1952
0
0
2006
2006
3
3
2006
2006
1
1
1976
1976
4
4
1976
1976
2
2
Removals
1952
1952
5
5
Net growth
1952
1952
3
3
2006
2006
2006
2006
1976
1976
1952
1952
2006
2006
Alaska
Alaska
6
6
1
1
1952
1952
2006
2006
1976
1976
1952
1952
2006
2006
1976
1976
1952
1952
2006
2006
1976
1976
1952
1952
2006
2006
1976
1976
1952
1952
Rocky
Pacific
Rocky
Pacific
Mountain
Coast
Mountain
Coast
Region
1
Region
1
Source: USDA Forest Service, Forest Inventory and Analysis
0
0
North
North
South
South
1976
1976
2
2
1952
1952
0
0
2006
2006
1
1
1976
1976
3
3
2
2
1952
1952
4
4
Growth continues to exceed removals on U.S. timber lands, as
it has for more than 50 years. Overall, domestic removals of
growing stock have declined from 15.8 to 15.5 billion cubic
feet since 2003. This decline is also reflected in the statistic
that conifers and broadleaf removals were 75 and 58 percent
of growth respectively in 2003, and currently these values are
Removals
65 and 49 percent respectively. Demand has Net
notgrowth
subsided, and
6
Net growth
Removals
6
imports continue to rise to meet U.S. wood needs (Indicators
6.28, 6.30,
5 and 6.32).
3
3
2006
2006
5
5
What has changed since 2003?
4
Figure 13-3.
Net growth and removals of broadleaves in
4
Net growth
Removals
the United States, 1952–2006.
4
4
1976
1976
Cubic
feet
(billions)
Cubic
feet
(billions)
6
6
Timber land is concentrated on private lands in the East and
public lands in the West. Recent studies indicate that 58 percent
of noncorporate private owners have harvested wood on their
land. Recent large divestitures of forest land by private corporate landowners, particularly forest industries, have left the
future viability of these lands for wood production less clear.
5
1952
1952
7
7
Removals
Removals
5
5
2006
2006
8
8
Net growth
Net growth
6
6
1976
1976
9
9
1952
1952
Cubic
Cubic
feet
feet
(billions)
(billions)
7
7
Saw and veneer logs and pulp wood are the dominant primary
wood products from U.S. timber lands, comprising 94 percent
of all wood removals, up from 75 percent in 1953 (fig. 13-5).
Cubic
feet
(billions)
Cubic
feet
(billions)
0
0
1952
1952
1952
1952
Cubic
feet
(billions)
Cubic
feet
(billions)
Cubic
Cubic
feet
feet
(billions)
(billions)
measure provides and estimate of the productive capacity of
15 based on maximum growth at the culmination of mean
forests
Public
Private
15
25
Public
Privateat
annual
increment.
Overall,
U.S.
timber
lands
are
growing
25
10
51 cubic
feet per acre per year, as opposed to a potential of 77
10
20
cubic feet. Thus, current growth is 66 percent of its estimated
20
maximum
potential. A clear capacity exists to sustain present
5
5
15
levels
of timber harvest from a pure wood volume standpoint,
15
even 0at current growth rates. Many reasons exist, however, as
0
10 why the potential growth may not be achieved. The main
to
10
reason is thatEast
the diverse objectives
of theWest
many different
owners
East
West
East
West
West
Removals
NetEast
growth
growth
of5 U.S. timber
lands mayRemovals
not
have theNet
maximization
of
wood
Removals
Net growth
Removals
Net growth
5
fiber
production as their primaryRegion
objective.
Region
2,000
6,000
6,000
0
0
4,000
4,000
East
East
1952
1952
2,000
2,000
0
0
West
West
1976
2006
1952
1976
2006
1976
2006
1952
1976
2006
Sawlogs/veneer
Pulpwood/composite
Sawlogs/veneer Product type
Pulpwood/composite
Product type
1952
1952
1976
2006
1952
1976
2006
1976
2006
1952
1976
2006
Sawlogs/veneer
Pulpwood/composite
Sawlogs/veneer Product type
Pulpwood/composite
Product type
Source: USDA Forest Service, Forest Inventory
and Analysis
2
Download