Your SlideShare is downloading. ×
Forest Structure & Spatial Restrictions: Interactions & How They Affect Harvest Goal Achievement
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×

Introducing the official SlideShare app

Stunning, full-screen experience for iPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Forest Structure & Spatial Restrictions: Interactions & How They Affect Harvest Goal Achievement

222
views

Published on

This presentation investigates the role of forest structure in limiting the achievement of harvest and/or revenue goals of forest management. Results show that some spatial arrangements of stands …

This presentation investigates the role of forest structure in limiting the achievement of harvest and/or revenue goals of forest management. Results show that some spatial arrangements of stands suffer much greater reductions in goal achievement than others. Some means of ameliorating the reductions is also given.


0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
222
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. Forest Structure & Spatial Restrictions: Interactions & How They Affect Harvest Goal Achievement Karl R. Walters Forest Planning Manager 1 © 2004 Forest Technology Group
  • 2. Spatially explicit harvest scheduling Stands are not spatially configured the way we would prefer them: Stands are smaller than harvest blocks Harvest blocks must exceed minimum size to be economical Stands are larger than harvest blocks Harvest blocks must not exceed some regulatory maximum opening size Forest structure is too dispersed Costs are minimized by harvesting large tracts in close proximity Forest structure is too concentrated Diversification of activities across a landscape is needed Social and political limitations Societal demands lead to legal remedies or self-regulation Spatial restriction rules on opening size, adjacency, green-up 2
  • 3. Stanley Based on a hierarchical approach to planning Solve strategic harvest schedule first Allocate subset of harvest schedule tactically Iterate as needed Area-restriction model approach Block configuration is not an input but part of solution Heuristic solution methodology 3
  • 4. Stanley input parameters Opening size limitations on harvest blocks Minimum feasible block size (economic considerations)=MINPARM Maximum allowable block size (BMP’s, SFI, state law)=MAXPARM Adjacency restrictions on harvest blocks Minimum buffer distance (proximity to contemporary blocks)=PROX Greenup (period to elapse before cutting adjacent block)=GREENUP Adjustments to strategic plan for spatial feasibility Deviations from strategic timing choices=TIMEDEV % Variation relative to optimal flow profile=FLOWPARM 4
  • 5. Stanley Input Parameters FlowParm Blue bars = strategic objective Yellow bars = achieved volume Maximum % variation is lowest achieved % minus highest achieved % 5
  • 6. Study Objectives To quantify how forest structures interact with spatial restrictions Why is there such variety in results in the literature? Is there a reproducible metric that will predict how much of an impact spatial restrictions will have in a given forest? To gain a better understanding of the process so as to improve implementation and/or affect policy How much difference would it make to increase minimum block size from 5ha to 10 ha? What is the impact of increasing the green-up interval by 1 year? Today, we will examine how GREENUP, PROX, TIMEDEV and MINPARM together affect harvest level achievement 6
  • 7. Experimental Design Forest structure 32,000 ha, uniform site quality, single species Uniform age-class distribution (1-40), 800 ha each Volume objective, non-declining flow constraint Minimum harvest age = 19 Planning horizon Strategic = 80 yrs, tactical = 15 yrs Varied spatial distribution of stands Square or hexagon grids 20 ha cells (1600 cells in a 40x40 grid of squares & hexagons) 5 ha cells (6400 cells in a 80x80 grid of squares) 7
  • 8. HX40R – 20 ha, random 8
  • 9. SQ40R – 20ha, random 9
  • 10. SQ80R – 5ha, random 10
  • 11. SQ40S – 20 ha, systematic-random 11
  • 12. SQ40C – 20ha, clustered 12
  • 13. SQ40CS – 20 ha, systematic-cluster 13
  • 14. Experimental Design Stanley blocking and scheduling software, 15 year planning horizon PROX – 3 levels of proximity (adjacents, 2nd ring, 3rd ring) GREENUP – 3 levels of greenup interval (3, 4 & 5 years) MINPARM – 2 levels (minimum 20 and 40 ha blocks) MAXPARM – 3 levels (maximum 120, 180 and 240 ha blocks) TIMEDEV – 3 levels (±5, ±10, ±15 yr deviations from timing) FLOWPARM – 2 levels (5% & 10% variation about flow) Factorial experiment 3888 obs. (6 factors x 54 levels x 2 replicates x 6 forests) 14
  • 15. Ranking of Solutions Highest average harvest volume AvgVol MinVol MaxVol BlockAvg HX40R SQ80R 0.684180 0.658666 0.711808 56.71373 SQ40CS 0.539730 0.511570 0.575562 117.53030 Lowest periodic harvest volume SQ40C 0.692941 0.672847 0.718571 97.61975 SQ40S 0.550097 0.521164 0.575552 99.33364 SQ40CS HX40R 0.703730 0.676697 0.730613 55.82006 Highest periodic harvest volume SQ40R 0.656411 0.629567 0.683752 64.03164 HX40R AvgVol MinVol MaxVol BlockAvg SQ80R 3 3 3 5 Largest average blocks SQ40CS 6 6 5 1 SQ40CS SQ40C 2 2 2 3 SQ40S 5 5 6 2 Smallest average blocks HX40R 1 1 1 6 HX40R SQ40R 4 4 4 4 15
  • 16. ANOVA – 20 ha cell configurations FACTOR F Value Pr(F) Null hypothesis: no differences in PROX 2261.404 0 GREENUP 1410.437 0 average harvest due to spatial TIMEDEV 540.237 0 MINPARM 241.224 0 configuration MPFD 240.700 0 GREENUP:PROX 185.640 0 Rejected, alpha = 1% FLOWPARM 135.775 0 MAXPARM 117.744 0 All parameters significant at 1% MINPARM:MPFD 112.221 0 PROX:MPFD 71.509 0 Significant interaction MINPARM:TIMEDEV 52.115 0 GREENUP:MPFD 40.881 0 PROX:MAXPARM 27.184 0.0000002 GREENUP & PROX PROX:FLOWPARM 26.202 0.0000003 PROX:TIMEDEV 17.775 0.0000256 MPFD & MINPARM MINPARM:TIMEDEV:MPFD 17.355 0.0000318 GREENUP:PROX:TIMEDEV 14.720 0.0001272 MINPARM & TIMEDEV FLOWPARM:MPFD 11.714 0.0006285 GREENUP:FLOWPARM 9.747 0.0018124 MINPARM:FLOWPARM 8.721 0.0031699 FLOWPARM:TIMEDEV 8.521 0.0035367 PROX:MINPARM 8.259 0.0040816 MAXPARM:MPFD 4.919 0.0266416 PROX:MAXPARM:TIMEDEV 4.333 0.0374717 GREENUP:MINPARM:MPFD 4.123 0.0423895 16
  • 17. Response of AvgVol to Greenup:Prox 17
  • 18. Response conditioned by MPFD MPFD MPFD MPFD MPFD M PFD 18
  • 19. SQ40C Clustered age classes 11 Sensitivity to proximity 0.617 distance increases with greenup 9 interval PROX 0.6 7 48 0.6 78 Relatively insensitive to 0.7 09 proximity at greenup = 3 5 0.7 39 3.0 3.5 4.0 4.5 Relatively insensitive to timing GREENUP choice deviations for smaller minimum blocks 0.67 8 Deviations initially help by 13 allowing more area to be 11 TIMEDEV harvested 9 Blocks become more 0.678 7 heterogeneous in composition 5 20 25 30 35 MINPARM 19
  • 20. SQ40CS Clustered, systematic age classes 11 0.458 Sensitivity to proximity distance increases with greenup interval 9 0. 39 7 PROX Sensitivity to proximity distance is 0.45 7 8 higher at short greenup intervals 0.51 8 0.57 9 5 than SQ40C 0.63 9 0.700 3.0 3.5 4.0 4.5 More sensitive to timing choice GREENUP deviations for smaller minimum blocks than SQ40C 0.4 58 Deviations initially help by 0.51 13 8 allowing more area to be harvested 11 TIMEDEV Blocks become more 9 0.518 heterogeneous in composition 7 5 20 25 30 35 MINPARM 20
  • 21. SQ40S Systematic, random age classes 11 0.456 Sensitivity to proximity distance increases with greenup interval 9 PROX 0.45 6 Sensitivity to proximity distance is 7 0.519 higher at short greenup intervals 5 0.58 1 Relatively insensitive to timing 0.7 07 0.6 44 3.0 3.5 4.0 4.5 choice deviations for smaller GREENUP minimum blocks Deviations initially help by 0.4 56 0.51 allowing more area to be harvested 9 13 0.58 1 Blocks become more 11 TIMEDEV heterogeneous in composition; 9 0.519 more pronounced than in 7 clustered age classes 56 0.4 5 20 25 30 35 MINPARM 21
  • 22. SQ40R Random age classes 11 Much lower sensitivity to 9 proximity at all greenup 0.6 2 PROX 9 intervals 7 Relatively more sensitive to 5 0.699 timing choice deviations than 3.0 3.5 4.0 4.5 GREENUP clustered age classes Blocks become more heterogeneous in composition; 0.6 29 13 more pronounced than in 11 TIMEDEV clustered age classes 9 0.629 99 0.6 7 9 0.55 5 20 25 30 35 MINPARM 22
  • 23. HX40R Random age classes 8 Insensitive to proximity at 0.6 7 5 9 short greenup intervals 6 PROX More sensitive to proximity at 0.6 87 5 higher greenups than SQ40R 4 0.71 5 3 More sensitive to timing choice 3.0 3.5 4.0 4.5 GREENUP deviations than clustered age classes and SQ40R 13 0.68 7 11 TIMEDEV 9 0.687 0.659 5 71 7 0. 5 20 25 30 35 MINPARM 23
  • 24. SQ80R Random age classes Relatively insensitive to 0.6 17 30 proximity at short greenup 13 PROX intervals 9 Less sensitive to proximity at 0. higher greenup intervals than 5 70 0 3.0 3.5 4.0 4.5 GREENUP SQ40R Decreasing sensitivity to timing choice deviations at higher 13 minimum block sizes 11 TIMEDEV 9 0.630 7 0.700 5 20 25 30 35 MINPARM 24
  • 25. Summary Identical forests from a strategic standpoint Yielded very different outcomes under spatial restrictions Forest structure was significant determinant Although contrived, forests have analogs in the real world SQ40C – similar to disturbance dominated natural forests SQ40CS – similar to plantation management of the US southeast SQ40R – similar to forests of the Northeast (small, heterogeneous stands) 25
  • 26. Summary Forest fragmentation Highly fragmented forests are less sensitive to greenup interval Neighboring stands are not likely to be harvested in same period anyway In non-fragmented forests, sensitivity to proximity distance increases with greenup interval Neighboring stands are of similar age, and therefore likely candidates for harvesting in same period; proximity distance determines how much of this area is made unavailable during greenup Allocation units Substands (SQ80R) yielded better solutions than stands (SQ40R) Smaller allocation units present more alternative block configurations 26
  • 27. Summary Block size Minimum block size can be much more limiting on volume achievement than maximum block size Maximum block size is limiting only in non-fragmented forests Mean block size is much smaller than maximum allowed Mean approaches maximum only in clustered forests (SQ40CS, SQ40C) Timing choice deviations Mitigate shortfalls by allowing for the creation of larger harvest blocks (sensitive to minimum block size) Significant deviations from original timing can nullify gains arising from increased harvest area 27
  • 28. Forest Technology Group Knowledge for Growth 28 © 2004 Forest Technology Group