Transcript No Slide Title

Development of Yield Estimates of Short Rotation
Poplar and Mapping of U.S. Poplar Yield Potential
Bill Berguson, Dan Buchman, Mike Cunningham, Bernie McMahon, Ray Miller, Randy
Rousseau, Rich Shuren, Brian Stanton and Jeff Wright
National Energy Crops Mapping Project
- Requested by DOE (Bryce Stokes) to put current energy crop
production (including woody crops) into a national context
- Coordination of the project done by Laurence Eaton at ORNL
- Sun Grant/DOE Poplar Team provided data to construct a common
national dataset (the best available at the current time)
- Analysis done by Oregon State PRISM Climate group (Chris Daly
and Mike Halbleib) using gridded climate and soils data
- OSU developed crop production indices for regression against our
dataset to develop “backbone” for extension of poplar yield
nationally
Clarifying Notes - Stand Metrics
- Yield – all aboveground components without leaves in oven-dry condition
- Basal Area – the sum of the cross-sectional area of all trees expressed on an
areal basis (square feet per acre in our case)
- The product of BA and height is linearly related to total stand volume
Yield Data Sources
- Data types and immediate utility varied across regions
- Minnesota – used research yield block data of new clones on commercial
production acreage with extrapolation to age 10
- Pacific Northwest – Primarily irrigated dataset, decided to drop irrigated plots
in final analysis – rain-fed only sites
- Mid-South and South Alluvial – measured commercial plantations established
by MWV
- Southern Uplands (ArborGen data) – due to little data, we attepted to take
advantage of new data from clone tests on uplands (pine and agricultural sites)
in the region
Emphasis on potential yield using selected clones for each region
Minnesota Yield Data
- Yield blocks embedded in operational plantings on Verso Paper land
- Typical 7 X 7 tree blocks with 3 replications
- Interior 9 trees measured X 3 reps = 27 observations per clone/site
- Mean annual increment extrapolated to age 10 using average free-to-grow
incremental BA and height growth in recent history of each plot
Minnesota Yield Site Locations
Well-managed plantations
with no significant
fertilization history
All on agricultural sites of
moderate productivity
except Waseca (very highly
productive agriculture site)
Minnesota Yield Dataset
Site
Joppru
Sebeka
Kniels
Woelfel
Hemming
Hemming
Hemming
Hansen
Hansen
Hansen
Hansen
LEA Grand Rapids
LEA Grand Rapids
Schultz
Schultz
Waseca
Waseca
Average
YearEst
Spacing(sqft)
1996
1996
1999
2001
2005
2005
2005
2006
2006
2006
2006
2007
2007
2007
2007
2007
2007
Clone
64
64
70
70
80
80
80
100
100
100
100
64
64
100
100
25
25
76
10yrMAI(OD tons/Acre)
DN5
3.0
NM6
3.4
D124
3.4
D109
3.5
9732-19
4.0
9732-31
5.1
D113
4.9
98121022
3.6
D109
3.6
9732-31
3.8
D113
3.6
9732-11
5.1
9732-06
4.8
9732-11
3.7
9732-31
3.5
9732-11
3.7
9732-06
3.8
3.9
•
Relatively straightforward process with some concerns due to differences in spacing and rotation age
•
Most stands at or near rotation age
•
Used new hybrid clone set resulting from breeding program or select P. deltoides – some NM6, DN5
•
Yields generally higher than those reported in “Growth, Yield, and Disease Resistance of 7- to 12Year-Old Poplar Clones in the North Central United States” D.A. Netzer et.al. GTR-NC-229, 2002
Example of extrapolation method for MN sites
solid line is measured, dashed line is extrapolated
- Used recent historical “free-to-grow” steady state production rate to age 10
- Average gap of extrapolation is 3 years, average MN yield is 3.9 tons/ac/yr
Mid-South Alluvial Dataset
• Long-term commercial
plantation program with good
plantation management, no
fertilization throughout
rotation
• Randy R. hired contractor to
establish and measure growth
plots at eight sites and 30
clone/site combinations
• At total of 1,495 DBH
measurements on 55 plots
with subset of height (n=164)
• Numbers input and crunched
at UMD-NRRI and yields
estimated
Mid-South Alluvial Sites
(MO, KY, IL)
- Average spacing is 270 TPA or 13 X 13
feet spacing (potential at higher density
unknown currently)
- Data ultimately collapsed to eight sites
with the average of the clones occurring
at those sites
- Due to wide spacings, a 9 year rotation
age assumed with data either used
directly for those sites greater than age 9
or extrapolated based on MAI:age
relationship across the dataset and ratio
of current to predicted
Age 4 superior clone in DOE/Sun Grant Consolidated
Clone Test (photo courtesy of Randy Rousseau)
Mid-South Alluvial Dataset
Site
Angelo
Angelo
Angelo
Angelo
Island 3
Island 3
Island 3
Island 3
Island 3
Island 3
Island 3
EstYear
Age
Tree/Acre
AvgHt(ft)
MDbh(in)
255
104
10.5
240
98
10.2
250
121
10.1
225
95
10.2
270
86
9.0
250
72
8.5
250
86
8.9
290
67
7.8
290
65
7.3
290
70
8.0
250
67
7.6
Ba/Acre
MAI
(sqft)
(BA/acre/yr)
152.6
13.9
135.0
12.3
138.9
12.6
125.9
11.4
119.1
10.8
99.3
9.9
108.9
12.1
95.8
12.0
84.2
10.5
100.5
12.6
78.6
9.8
MAI
tons/acre (tons/acre/yr)
85.7
7.8
71.6
6.5
90.6
8.2
64.8
5.9
55.0
5.0
38.8
3.9
50.5
5.6
34.7
4.3
29.4
3.7
37.9
4.7
28.3
3.5
1999
1999
1999
1999
1999
2000
2001
2002
2002
2002
2002
Clone
11WV316
11WV335
11WV369
11WV94
11unknown
10unknown
9unknown
8WV316
8WV90
8WV94
8WV99
Wolf Island
Island 3
Shelby Wilson
Shelby Wilson
Shelby Wilson
Wise
Wolf Island
2002
2003
2003
2003
2003
2003
2003
8unknown
7WV98
7WV90
7WV98
7WV99
7WV316
7WV370
300
270
290
295
280
270
280
67
58
59
61
60
80
73
7.3
6.6
6.5
6.9
7.1
8.5
8.4
88.3
64.8
65.8
75.4
77.8
105.7
106.9
11.0
9.3
9.4
10.8
11.1
15.1
15.3
31.9
20.2
21.0
24.9
25.4
45.6
42.0
4.0
2.9
3.0
3.6
3.6
6.5
6.0
Wolf Island
Ice Grain
Ice Grain
WMA
WMA
WMA
Wolf Island
2003
2004
2004
2004
2004
2004
2004
7WV98
6WV335
6WV90
6WV413
6WV90
6WV99
6unknown
270
270
270
275
290
275
265
78
47
46
65
57
68
66
7.7
6.0
5.6
8.0
6.1
7.8
6.7
87.9
53.6
46.8
96.5
59.7
90.5
64.6
12.6
8.9
7.8
16.1
9.9
15.1
10.8
36.8
13.6
11.7
33.9
18.2
33.2
23.0
5.3
2.3
1.9
5.6
3.0
5.5
3.8
Wolf Island
2004
6WV413
275
65
7.5
84.0
14.0
29.5
4.9
Wolf Island
Island 3
Peck
Wolf Island
2004
2005
2005
2005
6WV98
5WV90
5WV98
5WV98
270
270
265
279
76
65
62
64
8.5
7.5
7.8
6.7
105.7
82.7
87.3
68.3
17.6
16.5
17.5
13.7
43.3
28.9
29.0
23.7
7.2
5.8
5.8
4.7
Southeast Uplands Dataset
Very difficult to find single-clone blocks
over a sufficient time period using
material selected from tests in the region
Used ArborGen’s unique series of clone
tests across the Southeast ranging from
sites in GA, SC, NC and AL
Development of Reference Yield Curves for Clone
Tests on Southern Uplands
- Lack of data on selected clones planted on upland sites in the Southeastern US in
yield blocks on representative sites
- Growth from some published studies showed very low growth rates which were not
consistent with observations across the ArborGen’s network of trials in the area
- Any existing yield data is in private hands and unavailable (lost/discarded after
program termination)
- Challenge – how to make use of ArborGen’s clone trial network data to estimate
yield potential
- Can we use measured clone trial height of top clones as indicator of yield potential?
Development of Reference Yield Curves for Clone
Tests on Southern Uplands
- Decided to evaluate long-term datasets of poplar growth in regions particularly if
spacing varied in the study and data were collected over a rotation
- Through help of Rich Shuren and Brian Stanton, we were provided with two longterm datasets with variable spacing
- 1984 study
- 1990 study
- USFS-Stoneville – Roger Krinard
- Nelder plot spacing trial
- Large block test
Selected the GreenWood Resources 1984 and Krinard Nelder Plot datasets as these
included a range of spacings over a sufficient period with tree height and diameter
Evaluated expected stand basal area and height growth through time to develop a set
of expected stand BA and height with known final rotation MAI
Used expected height to compare against measured clone trial height growth to
estimate stand production
GreenWood 1984 Study
(Thanks to Rich and Brian)
GreenWood Resources 1984 study
- Westport - unirrigated
- Spacing treatments in a 5 X 5 matrix of 3, 4, 6, 8 and 12 feet width
reciprocal combinations for a total of 25 treatments
- Spacing ranged from 3X3 feet to 12 X 12
- Three replications, 25 tree/plot, 1,876 observations
- Diameter measured age 2 through 8 annually
- Height measure age 3 through 7
- Clone Hybrid 11
- Unique opportunity to explore growth curves through time and spacing
effects
Regression Analysis Results for Tree Height
GreenWood Resources 1984 Study
AvgHT = 17.1 + 4.48 StandAge - 0.234 SqftperTree + 0.0632 AgeXSqft-tree (R-squared = 85%)
Regression Analysis Results of Stand Basal Area
GreenWood Resources 1984 Study
StandBA = - 7.94 + 27.4 StandAge - 0.333 SqftperTree - 0.0692 AgeXSqft-tree (R-squared = 84%)
GreenWood 1984 MAI Biomass Production
Although not shown, age 8 BA and volume
decreases in dense treatments and increases
at wider spacings
Stands appear to hit a maximum of 190
square feet per acre and then begin slow
decline (max MAI has been reached)
regardless of spacing
Similar behavior in the North
Spacing
9
12
16
18
24
32
36
48
64
72
96
144
Age3
Age4
4.7
4.2
2.9
3.9
3.0
2.4
1.9
1.5
1.7
1.1
0.9
0.6
Age5
6.2
5.9
4.7
6.3
5.2
4.3
3.8
3.2
3.6
2.6
2.2
1.7
Age6
6.3
5.9
4.6
6.9
6.0
5.5
4.9
4.6
5.0
4.0
3.4
2.9
Age7
6.6
6.8
5.4
7.3
6.8
6.4
6.1
5.7
6.5
5.4
4.9
4.3
6.2
6.5
5.3
7.2
6.8
6.8
6.4
5.9
6.7
5.8
5.6
5.2
Krinard Nelder Plot Study
(USFS - SO 322)
- Study reported on data through age 19 but used through age 9
- Reported average diameter and height in plots, we translated to MAI
- Nelder Spoke Design – from 6 X 6 equivalent (1135 tpa) to 26 X 26 feet (63 tpa)
Trees/Acre
Tree Spacing
(sq.ft./tree)
63
691
113
385
201
217
356
122
637
68
1135
38
63
691
113
385
201
217
356
122
637
68
1135
38
63
691
113
385
201
217
356
122
637
68
1135
38
63
691
113
385
201
217
356
122
637
68
1135
38
Height
(ft)
StandAge
2
2
2
2
2
2
4
4
4
4
4
4
6
6
6
6
6
6
9
9
9
9
9
9
26
27
27
28
28
28
49
51
52
51
49
45
67
67
67
64
60
55
77
76
75
70
63
58
AvgDBH StandBA
(in)
(sq.ft./acre)
4.1
5.8
4
9.9
3.8
15.8
3.6
25.2
3.1
33.4
2.6
41.8
8.6
25.4
7.9
38.5
7
53.7
6
69.9
4.8
80.0
4
99.0
10.8
40.1
9.76
58.7
8.4
77.4
7
95.1
5.88
120.1
4.8
142.6
13
58.1
11.4
80.1
9.8
105.3
8.1
127.4
6.6
151.3
5.8
208.2
Estimated MAI
(dry tons/acre/year)
0.4
0.7
1.2
1.9
2.5
3.1
1.7
2.6
3.8
4.8
5.3
6.0
1.8
2.7
3.5
4.1
4.8
5.2
2.7
3.7
4.7
5.4
5.7
7.3
Regression of Tree Height on Age and Spacing
Krinard Nelder Plot
Height = 21.1 + 5.32 StandAge + 0.00237 SqftXAge (R-squared = 0.86)
Regression of Stand Basal Area on Age and Spacing
Krinard Nelder Plot Data
StandBA = 10.2 + 17.8 StandAge - 0.0385 Sqft-Tree - 0.0140 SqftXAge (R-squared = 0.85)
Yield Estimates for Southeast Uplands
- Were used the ratio of mean height of the top 10 clones in clone tests to that
estimated from the mean of the two models (Krinard-Nelder and GWR84)
- Calculate the % reduction in MAI based on same models estimating MAI using
age and stand spacing
- Used expected versus actual height to adjust MAI accounting for spacing
- Need more data and will continue to collect over time
- Assumed a rotation age of 7 except for slow-growing sites such as Moultry (8)
Stand Data
Spacing
(sqft/tree)
Age
Pred Ht (ft)
Height
(ft)
Pred BA (sq.ft/ac)
GWR 84 Krinard Average Actual/Pred GWR 84 Krinard
Average EstMAI
Floyd, GA
9
96
77
89.6
66.6
78.1
0.99
105
109
107.5
5.7
Eastover, SC
5
40
42.1
42.8
43.4
43.1
0.98
151
137
144.2
5.6
Moultry, SC
4
84
27.9
36.6
42.4
39.5
0.71
115
115
115.4
3
Randolf, AL
4
84
22.1
36.6
42.4
39.5
0.56
115
115
115.4
1.9
Randolf, AL
4
48
21.9
35.9
39.4
37.7
0.58
144
133
139.0
1.9
Greenville, NC
3
40
28.3
28.8
34.1
31.4
0.90
151
137
144.2
4.7
Bellville, GA
3
40
31.4
28.8
34.1
31.4
1.00
151
137
144.2
6
Comparison of Krinard (SO 347) and 0.4085XBAXHt Bole Volume Estimation
(outside-bark volume total bole)
Arborgen Poplar Specific Gravity Study
Blue bars: Hybrid
Red bars: P. deltoides
Source – ArborGen, Bijay Tamang, SFTIC Conference
•
•
•
Average SG of 0.35 for poplar in South, Mid-South and Midwest = 22 lbs/cubic foot density
Generalized biomass equation: OD tons (all aboveground components – top, limb, bole, bark) =
BA X Ht X 0.4085 X 22 X 1.2
Consistent with Krinard and northern volume estimates (actually universal) with density
remarkably similar across regions
Final Dataset for Calibration
Site Name
Moultry 2009 normals
Kniesel 1999 normals
Sebeka 1996 normals
Woelfel 2001 normals
Schultz 2007 normals
Hansen 2006 normals
DOE 2005 Puyallup 2005 normals
Joppru 1996 normals
Island 3 2005 normals
Hemming 2005 normals
Wolf Island 2005 normals
Wooten Farm 2010 normals
LEA Grand Rapids 2007 normals
Peck 2005 normals
Eastover 2008 normals
Floyd 2003 normals
Bellville 2010 normals
Wise 2003 normals
Angelo 1999 normals
Site Score
56
62
62
62
61
61
47
57
80
62
80
62
MAI (dry tons/ac/year)
3
3.4
3.4
3.5
3.6
3.6
4.2
4.5
4.6
4.7
4.7
4.7
66
79
62
70
65
80
80
4.9
5.5
5.6
5.7
6
6.2
6.8
- After meeting with OSU staff in Corvallis, collapsed dataset to 19 sites
Final model calibration from OSU site scores
(Mike Halbleib/Chris Daly)
South Uplands - Bellville, GA, Floyd, GA, Eastover, SC
Minnesota – Improved Clones
Alluvial Mid-South
Moultry, SC – Extremely Droughty
U.S. Map of Poplar Production Potential
- Note: red stars indicate site locations in the calibration dataset – not the case for PNW
Observations
- Analysis of available yield data from Alluvial South (USFS-Krinard) and Pacific
Northwest (GWR-Stanton/Shuren) shows that height growth is significantly higher
in the PNW and South compared to Midwest
- However, stand basal area production across all regions is much more consistent
through time with final age stand BA of 160 to 180 ft2 acre-1 across regions
expected (mean annual production higher in South and PNW due to tree height and
slightly higher BA growth)
- Growth curves in height and stand basal area developed from GWR dataset may
have utility in guiding research in high-density plantings in other locations
- Due to heavily leached soils in the Southern upland region, site selection (pH) and
fertilization will likely play a greater role in that region
- Unique opportunity to compare loblolly pine growth rates to cottonwood in paired
sites in ArborGen’s set of trials (pellet market may be very attractive for
cottonwood and hybrids)
- Continued measurement and expansion of tests in all regions and, particularly
Southeast Uplands, will help better define yield potential and site effects
Final Thoughts
- Dataset fraught with various issues that leads to imperfect, but reasonable
estimates of yield potential based on actual field data
- Spacing and coppice effects on yield in all regions unknown and need work
- SE US yield – continued measurement of existing clone trials and next generation
of large-block yield tests needed using best clones
- SE US – effects of drought periods on uplands unknown but poplar surviving well
on some very drought-prone sites (Moultry)
- Clone selection critical and extreme variation is evident in clone tests in all regions
- Opportunities for significant yield improvement through genetics at all locations
Acknowledgments:
Sun Grant Program – Tim Rials
Chris Daly, Mike Hableib – OSU
Laurence Eaton – ORNL
Bryce Stokes – CNJV
Don Kaczmarek - perspective on poplar yield in the SE