AN ABSTRACT OF THE THESIS OF
Theresa Marquardt for the degree of Master of Science in Forest Resources presented
on February 22, 2010.
Title: Accuracy and Suitability of Several Stand Sampling Methods in Riparian Zones.
Abstract approved:
_____________________________________________________________________
Temesgen Hailemariam
Stand structure in headwater riparian areas of western Oregon is highly
variable making quantification of structural attributes within these stands difficult. A
reliable sampling design is important to characterize and monitor changes in stand
structure over time. This study examined sixteen alternatives for sampling live
conifers and six alternatives for sampling hardwoods and snags. Each sampling
alternative was replicated 500 times on a 72 meter square stem map at eight headwater
stream locations in western Oregon. A variety of plot sizes and shapes were
considered at two sampling intensities of 10 % and 20 % of the stand area. Applied to
live conifers, the performance of the alternatives were assessed with respect to
estimation of trees per hectare, basal area per hectare, and height to diameter ratio
using root mean square error, average mean deviation and percent bias as criteria.
Performance with respect to snags and hardwoods was similar. In general, rectangular
strip designs oriented perpendicular to the stream outperformed circular, radial plots,
and variable plots.
©Copyright by Theresa Marquardt
February 22, 2010
All Rights Reserved
Accuracy and Suitability of Several Stand Sampling Methods in Riparian Zones.
by
Theresa Marquardt
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Presented February 22, 2010
Commencement June 2010
Master of Science thesis of Theresa Marquardt presented on February 22, 2010.
APPROVED:
_____________________________________________________________________
Major Professor, representing Forest Resources
_____________________________________________________________________
Head of the Department of Forest Engineering, Resources, and Management
_____________________________________________________________________
Dean of the Graduate School
I understand that my thesis will become part of the permanent collection
of Oregon State University libraries. My signature below authorizes release of my
thesis to any reader upon request.
_____________________________________________________________________
Theresa Marquardt, Author
ACKNOWLEDGEMENTS
The author expresses sincere appreciation to the friends and family who offered their
support along the way. Thank you to Drs. Temesgen Hailemariam and Paul Anderson
for your guidance and direction throughout this project; to Bianca Eskelson for her
assistance with data analysis and comments; and the remaining members of my
committee for their support. Thank you to those who assisted in collection of field
data and Jim Kiser for training in the use of the total station survey equipment.
Special thanks to Jeremy Marquardt for supporting me in the continuance of my
education and assisting in data entry. The author would also like to thank her parents
who have always supported her goals in life.
CONTRIBUTION OF AUTHORS
Dr. Temesgen Hailemariam assisted in data analysis and the organizational structure
and writing of chapters two and three. Dr. Paul Anderson assisted in the
organizational structure and writing of chapters two and three. Dr. John Tappeiner II
assisted in the writing of chapters two and three.
TABLE OF CONTENTS
Page
Chapter 1 – General Introduction
1
Introduction
1
Literature Review
Riparian Areas
Headwater Streams
Forest Structure
Quantifying Riparian Areas
Evaluation of Sampling Designs
References
Chapter 2 - Examination of the accuracy and suitability of selected sampling
methods to quantify selected stand attributes within riparian zones
4
4
4
5
6
6
8
12
Abstract
12
Introduction
12
Methods
15
Results
24
Discussion
32
Conclusion
36
References
37
Chapter 3 - Examination of the suitability of selected sampling methods to
quantify stand attributes of hardwoods and snags within riparian zones
Abstract
40
40
TABLE OF CONTENTS (CONTINUED)
Page
Introduction
40
Methods
43
Results
51
Discussion
57
Conclusion
60
References
61
Chapter 4 – General Conclusion
64
Bibliography
66
Appendices
70
Appendix A: Sampling Alternatives Illustrated
70
Appendix B: List of Acronyms
75
LIST OF FIGURES
Figure
Page
2.1 Map of DMS site locations (Cissel et al. 2006)
15
2.2 Illustration of plot layout on the sampled stream
17
3.1 Map of DMS site locations (Cissel et al. 2006)
43
3.2 Illustration of plot layout on the sampled stream
46
LIST OF TABLES
Table
Page
2.1 Description of stream reaches sampled from the DMS site locations
16
2.2 List of tree parameters measured on each stem mapped tree and
the recording protocol for each measurement
19
2.3 Description of the sampling alternatives examined in this study
including the number of plots simulated at the 10 and 20% sampling
intensity
23
2.4 Summary of sampling alternatives evaluated using RMSE
26
2.5 Summary of performance of sampling alternatives evaluated using APB
27
2.6 Summary of performance of sampling alternatives evaluated using
MAD
28
2.7 Summary of alternatives that performed well when estimating trees per
hectare (TPH)
29
2.8 Summary of alternatives that performed well when estimating basal area per
hectare (BAPH)
30
2.9 Summary of alternatives that performed well when estimating the height to
diameter ratio (H/D)
31
3.1 Description of tree density, buffer width, slope and aspect for the eight
stream reaches sampled from the DMS site locations
45
3.2 Description of stream and tree characteristics of sampled reaches from
the DMS site locations
45
3.3 List of tree parameters measured on each stem mapped tree and the recording
protocol for each measurement
47
3.4 Description of the sampling alternatives examined in this study
including the number of plots simulated at the 10 and 20% sampling
intensity
50
LIST OF TABLES (CONTINUED)
Table
Page
3.5 Summary of performance of sampling alternatives evaluated using
RMSE
53
3.6 Summary of performance of sampling alternatives evaluated using
percent bias (PB)
54
3.7 Summary of the standard deviation for sampling alternatives evaluated
using root mean square error (RMSE)
55
3.8 Summary of the standard deviation for sampling alternatives evaluated
using percent bias (PB)
56
LIST OF APPENDIX FIGURES
Figure
Page
A.1 Illustration of the FAP5R sampling alternative at the 20% intensity at
the Bottomline 13 location
70
A.2 Illustration of the FAP9S sampling alternative at the 20% sampling
intensity at the O.M. Hubbard 36 location
71
A.3 Illustration of the ASTP3 sampling alternative at the Bottomline 13 location
and 10% sampling intensity
71
A.4 Illustration of the PAST3 sampling alternative at the 20% sampling intensity
at the O. M. Hubbard location
72
A.5 Illustration of the OSSP7 sampling alternative at the 10% sampling intensity
at the Bottomline 13 location
72
A.6 Illustration of the PEST7 sampling alternative at the Bottomline 13 location
and 20% sampling intensity
73
A.7 Illustration of the HLS08 at the Bottomline 13 location and 20%
sampling intensity
73
A.8 Illustration of the SEC11 sampling alternative at the O. M. Hubbard location
and 20% sampling intensity
74
1
Chapter 1 – General Introduction
Introduction
Sampling designs have been compared and contrasted for years. Designs have
been examined for a variety of forest types and across the globe. In the Pacific
Northwest, increased interest has been given to riparian forests and understanding the
ecology as well as monitoring changes in forest structure over time. However, few
studies have documented sampling designs that are best suited to riparian areas. There
is reason to suspect that certain sampling designs may not be well suited for the
amount of variation that occurs with these areas. In order to collect
data within riparian areas pertaining to forest measurements, a robust sampling design
is needed. This study aims to evaluate the performance of a variety of sampling
designs within riparian areas of western Oregon.
Forest vegetation and structure varies widely in riparian zones. The streams in
this study were either first or second order streams. In order to evaluate the selected
sampling designs, a 72 by 72 meter block was stem mapped at eight sites located in
the foothills of the Cascade and Coast range of western Oregon. This study seeks to
improve our knowledge of sampling designs that are suited for a forest that is highly
diverse in forest structure. The study looks at the sampling designs in two parts. First
the study evaluates the performance of the sampling designs to estimate trees per
hectare, basal area per hectare and the height to diameter ratio on the live conifers at
the eight sites. Second, the study evaluates the ability of the sampling designs to
estimate trees per hectare and basal area per hectare on dead trees and live hardwoods
at the eight sites.
This simulation of multiple sampling alternatives was based on data derived from
United States Bureau of Land Management (BLM) Density Management Study
(DMS) sites. A primary goal of the DMS is to study options for silvicultural
treatments of young stands to create and maintain late-successional forest
characteristics. Incepted in 1994, the DMS is a collaborative project between the
BLM; US Forest Service, Pacific Northwest Research Station (PNW); US Geological
2
Service (USGS); and Oregon State University (OSU). Stand age ranged among sites
from 40 to 70 years.
Sampling designs were first evaluated on live conifers, then on live hardwoods and
all dead trees. In this study, the selected stand attributes were trees per hectare (TPH),
basal area per hectare (BAPH), and height to diameter ratio (H/D) when sampling live
conifers and trees per hectare and basal are per hectare when sampling snags and
hardwoods. Some sampling methods mainly capture variation perpendicular to the
stream, while others focus on sampling attributes parallel to the stream. Sixteen
alternatives were compared for the first part of the study and six alternatives were
compared for the second part of the study. The overall goal was to recommend a
single sampling design that would capture forest structure within riparian areas of
headwater streams in western Oregon.
The performance of the individual sampling alternatives within this study provides
information about sources of spatial variation within riparian zones. The precision of
TPH, BAPH and H/D was compared for simple random sampling (SRS), systematic
random sampling (SYRS), stratified random sampling (STRS), two-stage sampling,
horizontal line sampling (HLS) and sector sampling designs for the first part of the
study. The wide range of sampling alternatives in the first part of the study was
narrowed down to the better performing designs for the second part of the study.
These included SYRS and STRS designs.
The second part of this study focused on sampling of hardwoods and snags. In a
conifer dominated forest, hardwoods and snags may not fall in very many sample
plots. These objects can be rare and therefore one could expect that a sampling
alternative that performs well to sample conifers may not perform as well in sampling
hardwoods and snags. It is difficult to discuss a sampling design that is adequate in
capturing forest structure without considering its performance in estimating attributes
of hardwoods and snags. From the stem maps at each site, the trees appeared to be
scattered with no discernable pattern. A second aim of this study was to find out
3
whether sampling alternatives that have performed well in sampling conifers in
riparian areas would also perform well when sampling hardwoods and snags.
The scope of inference for this study falls within headwater streams with 40-60
year old Douglas-fir forests of western Oregon with a buffer of approximately 220 feet
(one site potential tree height). Analysis and inference can be applied to forests of
similar species composition and stand density. The history of the study sites includes
no management prior to 1994 when thinning treatments began. In this study, an
unthinned control stand, high and moderate density retention were observed.
Slope
and aspect from the analysis were recorded and compared among study sites so that
differences between these variables could be deciphered. Two slope classes were
compared, mild and steep. Multiple aspects for each side of the stream were included
within the study. Structural heterogeneity among these sites may naturally vary and
caution should be used when applying these density treatments to larger areas of
different forest structure. This study may not be representative for areas with
narrower streamside buffers. In most cases, this study can be applied to 40-60 year
old Douglas-fir forests of western Oregon with an approximate 220 foot buffer.
4
Literature Review
Riparian Areas
Quantifying stand structure within riparian areas can be extremely difficult.
Riparian areas rank among the most complex, variable and dynamic terrestrial habitats
in the world (Naiman and Decamps 1997; Coroi et al. 2004). This is no different in
the Pacific Northwest (Acker et al. 2003). Stocking, horizontal and vertical structure
are all highly variable within riparian areas of headwater streams (Richardson &
Danehy 2007). For example, hardwood and conifer components vary in density,
diameter and total tree height. Because of the uniqueness of riparian areas, sampling
forest attributes and monitoring them over time can be difficult.
Riparian areas can be described as narrow ribbon like corridors (Barker et al.
2002) where the transition between aquatic and terrestrial ecosystems occurs (qtd. in
Anderson 1987). For this reason, riparian areas are diverse in both plant communities
and fish and wildlife (Cissel et al. 2006).
Studies that focus on structure of
unmanaged riparian areas of western Oregon include Pabst & Spies, (1999) and
Nierenberg & Hibbs, (2000). Conifer species that are common to these areas Douglasfir (Pseudotsuga menziesii (Mirb.)), western hemlock (Tsuga heterophylla (Raf.)),
western redcedar (Thuja plicata (Donn)), and Sitka Spruce (Picea sitchensis (Bong.))
(Pabst & Spies 1999, Barker et al. 2002).
The forests adjacent to streams are
important to stream temperature, bank stability, and provide a source of large woody
debris into stream systems (Barker et al. 2002). Due to their unique nature, many
states have adopted best management practices to protect water quality and sustain
aquatic habitat (Harrington & Schoenholtz, eds. 2005).
Headwater Streams
Headwater streams are unique from larger streams (Richardson & Danehy
2007) and serve as important habitat for amphibians and other non-fish vertebrates
(Olson & Weaver 2007). These streams make up a high percentage of total stream
length (Richardson & Danehy 2007) and drain much of the overall watershed area
(Anderson et al. 2007). For example, stream amphibians have strong associations with
5
physical habitat features (Olson & Weaver 2007). Headwater streams provide
sediment, nutrients, breeding ground for downstream organisms, and organic matter
(Gomi et al. 2002, Richardson & Danehy 2006). For this reason they are important to
fish production and overall stream health of downstream ecosystems (Richardson et al.
2005). Headwater streams can be characterized as being narrower than three meters in
bankfull width, a closed or nearly closed canopy, seasonal or intermittent, and having
a high amount of stream edge relative to the stream surface area (Richardson &
Danehy, 2006).
Forest Structure
There are many common tree measurements that could be used to quantify
forest structure in both the horizontal and vertical directions. Forest structure could be
described in a variety of ways such as using nearest neighbor indices (Kint et al. 2003,
Aguirre et al. 2003), measures of tree density (Coroi et al. 2004, Lhotka &
Loewenstein 2006), canopy characteristics (Nierenberg & Hibbs 2000, Harper &
Macdonald 2001, Pabst & Spies 1999, Lhotka & Loewenstein 2006), height
distribution (Chen & Bradshaw 1999), diameter distributions (Chen & Bradshaw
1999, Mason et al. 2007), snag density (Harper & Macdonald 2001, Pabst & Spies
1999), basal area per hectare (Pabst & Spies 1999), crown diameter (Chen &
Bradshaw 1999), spatial analysis (Kint et al. 2003, Aguirre et al. 2003, Mason et al.
2007, Chen & Bradshaw 1999), downed wood (Harper & McDonald 2001) and
species composition (Aguirre et al. 2003, Nierenberg & Hibbs 2000, Harper &
Macdonald 2001, Coroi et al. 2004, Pabst & Spies 1999). In a review of forest
structure McElhinny et al. (2005) listed foliage height diversity, canopy cover, tree
diameter, tree height, tree spacing, stand biomass, species richness, understory
richness and deadwood as attributes of forest structure previously used in published
literature.
As a component of forest structure, both hardwoods and snags are an important
part of forest diversity and habitat (Holden et al. 2006, Hagar 2007). Snags serve as
homes for cavity nesters (Holden et al. 2006), and may fall over and provide cover for
6
small mammals (Holden et al. 2006), or supply down wood for streams (Harmon, et
al. 1986). Hardwoods are important to a variety of songbird species, bats, and small
mammals (Hagar 2007). The stands where sampling took place during this study were
highly dominated by live conifer species. However, when quantifying forest structure,
a sampling design needs to be robust and capable of capturing snags and hardwoods
within a conifer dominated forest.
Quantifying Riparian Areas
Many techniques are used to quantify riparian areas.
These include strip
sampling, fixed area plots and line-intercept sampling (Welsh and Ollivier 1997; Coroi
2004; Pabst and Spies 1999; Acker et al. 2003; Welsh et al. 2005; Nierenberg and
Hibbs 2000; Uowolo et al. 2005; Lyon and Gross 2005). In Dignan and Bren (2003),
100m transects from stream to upslope were used to describe the effect of logging on
understory light environments in riparian buffer strips. Measurements were taken at 1,
3.4, 6.8 and 10m intervals (Dignan and Bren, 2003, p. 161). In another study of
vegetation by Uowolo et. al. (2005), 20 sites were investigated over time. One 1000
m2 plot was taken at each site. When sampling for amphibians within riparian zones,
Welsh and Ollivier (1997) established random “belts” which were strips running
perpendicular to the stream. These are just a few of the infinite sampling designs that
can be implemented within riparian areas.
Evaluation of Sampling Designs
A variety of methods have been used to compare sampling designs.
A
common metric for comparison has been efficiency, either by relative efficiency or a
measurement of time per plot (Coble & Grogan 2007, Dahl et al. 2008, Ducey et al.
2002, Johnson & Hixon 1952, Kenning et al., 2005, Lynch, 2003, Paulo et al. 2005,
Sparks et al. 2002, Stamatellos, 1995, Williams et al. 2005). Although time was not
considered in this study, generalizations about the timeliness of many of the
alternatives used in this study can be drawn from the literature. Other common
measures were standard error of the mean (Dahl et al., 2008, Ducey et al. 2002,
Lindsey et al. 1958, Paulo et al. 2005, Schreuder et al. 1987), and bias (Kenning et al.
7
2005, Nelson et al. 1998, Schreuder et al. 1992, Temesgen, 2003, Tokola & Shrestha,
1999). Williams et al. 2005 used root mean square error (RMSE). The methods
chosen to compare sampling designs within this study include RMSE, absolute percent
bias (APB), percent bias (PB), and mean absolute deviation (MAD).
In the comparison of sampling designs, plot shape has often been of interest
when sampling stand attributes (Bormann 1953, Freese 1961, Dale et al. 1991,
Johnson & Hixon 1952). When comparing circular fixed area plots to rectangular
plots, Johnson & Hixon (1952) found that rectangular plots outperformed circular
plots. In addition, Bormann 1953 also noted that rectangular plots resulted in smaller
variance when compared to square plots. Orientation of plots was also discussed in
Bormann 1953 and suggested that where gradients occurred, it was best to use a
rectangular plot running perpendicular to changes in forest composition.
In a
discussion of plot size and variability, Freese 1961 discussed orienting rectangular
plots across contours (gradients) as well as the resulting smaller variation in long,
narrow plots (p. 679).
8
References
Acker, S. A., Gregory, S. V., Lienkaemper, G., McKee, W. A., Swanson, F. J., &
Miller, S. D. 2003. Composition, complexity, and tree mortality in riparian
forests in the central Western Cascades of Oregon. For. Ecol. Manage., 173:
293-308.
Aguirre, O., Hui, G., von Gadow, K., & Jimenez, J. 2003. An analysis of spatial forest
structure using neighbourhood-based variables. For. Ecol. Manage., 183: 137145.
Anderson, P. D., Larson, D. J., & Chan, S. S. 2007. Riparian buffer and density
management influences on microclimate of young headwater forests of western
Oregon. For. Sci., 53(2): 254-269.
Anderson, W. E. 1987. Riparian Area Definition: A viewpoint. Rangelands, 9(2), 70.
Barker, J. R., Ringold, P. L., & and Bollman, M. 2002. Patterns of tree dominance in
coniferous riparian forests. For. Ecol. Manage., 166, 311-329.
Bormann, F. H. 1953. The statistical efficiency of sample plot size and shape in forest
ecology. Ecology, 34(3), 474-487.
Chen, J., & Bradshaw, G. A. 1999. Forest structure in space: a case study of an old
growth spruce-fir forest in Changbaishan Natural Reserve, PR China. For.
Ecol. Manage., 120, 219-233.
Cissel, J., Anderson, P., Berryman, S., Chan, S., Puettman, K., and Thompson, C.
2006. BLM Density Management and Riparian Buffer Study: Establishment
Report and Study Plan. U.S. Dept. Interior, U.S. Geological Survey. 161 p.
Coble, D. W., & Grogan, J. 2007. Comparison of systematic line-point and double
sampling designs for pine and hardwood forests in the western gulf. South. J.
Appl. For., 31(4): 199-206.
Coroi, M., Skeffington, M. S., Giller, P., Smith, C., Gormally, M., & O'Donovan, G.
2004. Vegetation diversity and stand structure in streamside forests in the
south of Ireland. For. Ecol. Manage., 202: 39-57.
Dahl, C. A., Harding, B. A., & Wiant, H. V. 2008. Comparing double-sampling
efficiency using various estimators with fixed-area and point sampling. North.
J. Appl. For., 25(2): 99-102.
Dale, V. H., Franklin, W. M., & Gardner, R. H. 1991. Sampling ecological
information: choice of sample size. Ecological Modeling, 57, 1-10.
Dignan, P., Bren, L., 2003. A study of the effect of logging on the understory light
environment in riparian buffer strips in a south-east Australian forest. For.
Ecol. Manage., 172: 161-172.
Ducey, M. J., Jordan, G. J., Gove, J. H., & Valentine, H. T. 2002. A practical
modification of horizontal line sampling for snag and cavity tree inventory.
Can. J. For. Res., 32: 1217-1224.
Freese, F. 1961. Relation of plot size to variability: An approximation. J. Forestry.,
59(9): 679.
9
Gomi, T., Sidle, R. C., & Richardson, J. S. 2002. Understanding processes and
downstream linkages of headwater systems. BioScience, 52(10): 905-916.
Hagar, J. C. 2007. Wildlife species associated with non-coniferous vegetation in
Pacific Northwest conifer forests: A review. For. Ecol. Manage., 246: 108-122.
Harmon, M. H., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J.
D., Anderson, N. H., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G.
W., Cromack Jr., K., & Cummins, K. W. 1986. Ecology of coarse woody
debris in temperate ecosystems. Adv. Ecol. Res., 15: 133-302.
Harper, K. A., & Macdonald, S. E. 2001. Structure and composition of riparian boreal
forest new methods for analyzing edge influence. Ecology. 82(3): 649-659.
Harrington, Constance A., Scchoenholtz, Stephen H., eds. 2005. Productivity of
Western forests: a forest products focus. Gen. Tech. Rep. PNW-GTR-642.
Portland, OR: U.S. Dept. of Agriculture, Forest Service, Pacific Northwest
Research Station. 176 p.
Holden, Z. A., Morgan, P., Rollins, M. G., & Wright, R. G. 2006. Ponderosa pine
snag densities following multiple fires in the Gila Wilderness, New Mexico.
For. Ecol. Manage., 221, 140-146.
Johnson, F. A., & Hixon, H. J. 1952. The most efficient size and shape of plot to use
for cruising in old-growth Douglas-fir timber. J. Forestry, 50(1): 17-20.
Kenning, R. S., Ducey, M. J., Brissette, J. C., & Gove, J. H. 2005. Field efficiency and
bias of snag inventory methods. Can. J. For. Res., 35: 2900-2910.
Kint, V., Meirvenne, M. V., Nachtergale, L., Geudens, G., & Lust, N. 2003. Spatial
methods for quantifying forest stand structure development: A comparison
between nearest-neighbor indices and variogram analysis. For. Sci. 49(1): 3649.
Lhotka, J. M., & Loewenstein, E. F. 2006. Indirect measures for characterizing light
along a gradient of mixed-hardwood riparian forest canopy structures. For.
Ecol. Manage., 226, 310-318.
Lyon, J., & Gross, N. M. 2005. Patterns of plant diversity and plant-environmental
relationships across three riparian corridors. For. Ecol. Manage., 204: 267-278.
Lynch, A. M. 2003. Comparison of fixed-area plot designs for estimating stand
characteristics and western spruce budworm damage in southwestern U.S.A.
forests. Can. J. For. Res., 33: 1245-1255.
Lindsey, A. A., Barton, J. D., & Miles, S. R. 1958. Field efficiencies of forest
sampling methods. Ecology, 39(3): 428-444.
Maclean, D. A., & Ostaff, D. P. 1983. Sample size - precision relationships for use in
estimating stand characteristics and spruce budworm caused tree mortality.
Can. J. For. Re., 13: 548-555.
Mason, W. L., Connolly, T., Pommerening, A., & Edwards, C. 2007. Spatial structure
of semi-natural and plantation stands of Scots pine (Pinus sylvestris L.) in
northern Scotland. Forestry, 80(5): 567-586.
10
McElhinny, C., Gibbons, P., Brack, C., & Bauhaus, J. 2005. Forest and woodland
stand structural complexity: Its definition and measurement. Forest Ecol.
Manage.: 218, 1-24.
Naiman, R. J., Decamps, H., & Pollock, M. 1993. The role of riparian corridors in
maintaining regional biodiversity. Ecol. Appl., 3(2): 209-212.
Nierenberg, T. R., & Hibbs, D. E. 2000. A characterization of unmanaged riparian
areas in the central Coast Range of western Oregon. For. Ecol. Manage., 129:
195-206.
Nelson, R. F., Gregoire, T. G., & Oderwald, R. G. 1998. The effects of fixed-area plot
width on forest canopy height simulation. For. Sci., 3(44): 438-444.
Olson, D. H., & Weaver, G. 2007. Vertebrate assemblages associated with headwater
hydrology in western Oregon managed forests. For. Sci., 52(2): 343-355.
Pabst, R. J., & Spies, T. A. 1999. Structure and composition of unmanaged riparian
forests in the coastal mountains of Oregon, U.S.A. Can. J. For. Res., 29: 15571573.
Paulo, M. J., Tomé, M., Otten, A., & Stein, A. 2005. Comparison of three sampling
methods in the characterization of cork oak stands for managementpurposes.
Can. J. For. Res., 35: 2295-2303.
Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany.
10(2): 290-300.
Richardson, J. S., & Danehy, R. J. 2007. A synthesis of the ecology of headwater
streams and their riparian zones in temperate forests. For. Sci., 53(2): 131-147.
SAS Institute Inc. 1990. SAS/STAT User’s Guide, Version 6, Fourth Edition.
Volume 2. SAS Institute Inc., Cary, NC 848 pp.
Schreuder, H. T., Banyard, S. G., & Brink, G. E. 1987. Comparison of three sampling
methods in estimating stand parameters for a tropical forest. For. Ecol.
Manage., 21: 119-127.
Schreuder, H. T., Rennie, J. C., & Williams, M. 1992. Comparison of three sampling
schemes for estimating frequency and D2H by diameter class - a simulation
study. For. Ecol. Manage., 50: 117-131.
Smith, C. Y., Moroni, M. T., & Warkentin, I. G. 2009. Snag dynamics in post-harvest
landscape of western Newfoundland balsam fir-dominated boreal forests. For.
Ecol. Manage., 258: 832-839.
Sparks, J. C., Masters, R. E., & Payton, M. E. 2002. Comparative evaluation of
accuracy and efficiency of six forest sampling methods. Proc. Okla. Acad. Sci.,
82, 49-56.
Stamatellos, G. S. 1995. Comparison of point and point-3P sampling for forest height
to diameter ratio estimation with cost analysis. For. Ecol. Manage., 74: 75-79.
Temesgen, H. 2003. Evaluation of sampling alternatives to quantify tree leaf area.
Can. J. For. Res., 33(1): 82-95.
Tokola, T., & Shrestha, S. M. 1999. Comparison of cluster-sampling techniques for
forest inventory in southern Nepal. For. Ecol. Manage. 116: 219-231.
11
Uowolo, Al. L., Binkley, D., Adair, E. C., 2005. Plant diversity in riparian forests in
northwest Colorado: Effects of time and river regulation. For. Ecol. Mngmnt.
218: 107-114.
Welsh, H. H. Jr., Ollivier, L. M., Hankin, D. G., 1997. A habitat based design for
sampling and monitoring stream amphibians with an illustration from
Redwood National Park. Northwestern Naturalist. 78: 1-16.
Williams, M. S., Ducey, M. J., & Gove, J. H. 2005. Assessing surface area of coarse
woody debris with line intersect and perpendicular distance sampling. Can. J.
For. Res., 35: 949-960.
Wisdom, M. J., & Bate, L. J. 2008. Snag density varies with intensity of timber
harvest and human access. For. Ecol. Manage., 255: 2085-2093.
12
Chapter 2 - Examination of the accuracy and suitability of selected sampling
methods to quantify selected stand attributes within riparian zones.
Abstract
Sixteen sampling alternatives were examined for their performance to quantify
selected attributes of overstory conifers in riparian areas of western Oregon. Each
alternative was examined at eight locations on a 72 meter square stem map. The
alternatives were evaluated based on how well selected stand attributes (trees per
hectare, basal area per hectare and height to diameter ratio) were estimated using root
mean square error, absolute percent bias, and mean absolute deviation. In general,
rectangular strip designs outperformed circular, radial plots and variable plots.
Sampling alternating strips perpendicular to the stream and 3.6 meters in width
outperformed all other alternatives.
Introduction
This study examines the accuracy and suitability of selected sampling methods
to quantify selected stand attributes in riparian forests of western Oregon. Forest
structure can be described in a variety of ways such as using nearest neighbor indices
(Kint et al. 2003, Aguirre et al. 2003), measures of tree density (Coroi et al. 2004,
Lhotka & Loewenstein 2006), canopy characteristics (Nierenberg & Hibbs 2000,
Harper & Macdonald 2001, Pabst & Spies 1999, Lhotka & Loewenstein 2006), height
distribution (Chen & Bradshaw 1999), diameter distributions (Chen & Bradshaw
1999, Mason et al. 2007), snag density (Harper & Macdonald 2001, Pabst & Spies
1999), basal area per hectare (Pabst & Spies 1999), crown diameter (Chen &
Bradshaw 1999), spatial analysis (Kint et al. 2003, Aguirre et al. 2003, Mason et al.
2007, Chen & Bradshaw 1999), downed wood (Harper & McDonald 2001) and
species composition (Aguirre et al. 2003, Nierenberg & Hibbs 2000, Harper &
Macdonald 2001, Coroi et al. 2004, Pabst & Spies 1999). The best sampling
alternative is the one which most accurately quantifies the chosen attributes of forest
structure in a highly diverse riparian forest.
13
Quantifying stand attributes within riparian areas can be extremely difficult.
Riparian areas rank among the most complex, variable and dynamic terrestrial habitats
in the world (Naiman and Decamps 1997; Coroi et al. 2004). This is no different in
the Pacific Northwest (Acker et al. 2003). Stocking, horizontal and vertical structure
are all highly variable within riparian areas of headwater streams (Richardson &
Danehy 2007). For example, hardwood and conifer components vary in density,
diameter and total tree height. Snags within riparian areas add to the variety of
vertical structure. In this study, the selected stand attributes were trees per hectare,
basal area per hectare, and height to diameter ratio. The aim of this study was to
recommend a sampling alternative which best accounts for the complexities of riparian
areas and therefore lends itself to monitoring forest structure.
The riparian areas of interest in this study were primarily headwater streams.
Headwater streams are different from larger streams (Richardson & Danehy 2007) and
serve as important habitat for amphibians and other non-fish vertebrates (Olson &
Weaver 2007). These streams are typically first or second order streams. They make
up a high percentage of total stream length (Richardson & Danehy 2007) and drain
much of the overall watershed area (Anderson et al. 2007). For example, stream
amphibians have strong associations with physical habitat features (Olson & Weaver
2007). An accurate sampling alternative is important to describe wildlife habitat and
species diversity within these unique stream systems.
A variety of methods have been used to compare sampling designs. A common
metric for comparison has been efficiency, either by relative efficiency or a
measurement of time per plot (Coble & Grogan 2007, Dahl et al. 2008, Ducey et al.
2002, Johnson & Hixon 1952, Kenning et al., 2005, Lynch, 2003, Paulo et al. 2005,
Sparks et al. 2002, Stamatellos, 1995, Williams et al. 2005). Although time was not
considered in this study, generalizations about the timeliness of many of the
alternatives used in this study can be drawn from the literature. Other common
measures are standard error of the mean (Dahl et al., 2008, Ducey et al. 2002, Lindsey
14
et al. 1958, Paulo et al. 2005, Schreuder et al. 1987), and bias (Kenning et al. 2005,
Nelson et al. 1998, Schreuder et al. 1992, Temesgen, 2003, Tokola & Shrestha, 1999).
Williams et al. 2005 used RMSE. This study evaluated the variation and accuracy in
stand structural parameter estimates for each sampling alternative compared to the
known parameter mean for each of eight headwater stream locations.
The performance of the individual sampling alternatives within this study provides
information about sources of spatial variation within riparian zones. This is an
observational study and when considering the selected sampling alternatives, one
should also be aware of the site characteristics on which this study was conducted.
Some sampling methods mainly capture variation perpendicular to the stream, while
others focus on sampling attributes parallel to the stream. The predictive
quantification of trees per hectare, basal area and height to diameter ratio was
compared for simple random sampling (SRS), systematic sampling with a random
starting point (SYRS), stratified random sampling (STRS), two-stage sampling,
horizontal line sampling (HLS) and sector sampling designs. The primary objective of
this study is to discern what types of sampling designs are best suited for riparian areas
of Western Oregon.
15
Methods
This study resided on United States Bureau of Land Management (BLM)
Density Management Study (DMS) sites. A primary goal of the DMS is to study
options for silvicultural treatments of young stands to create and maintain latesuccessional forest characteristics. Incepted in 1994, the DMS is a collaborative
project between the BLM; US Forest Service, Pacific Northwest Research Station
(PNW); US Geological Service (USGS); and Oregon State University (OSU). Sites
are located in the Coast Range and the western foothills of the Cascade Range in
Oregon. Stand age ranges among sites from 40 to 70 years. Streams classified as
headwater streams (generally first- or second-order streams) were the focus of this
study. Figure 2.1 shows the location of the DMS sites throughout western Oregon.
Figure 2. 1 Map of DMS site locations (Cissel et al. 2006).
16
The locations sampled in this study were chosen using a stratified random
sampling scheme. A list of all possible headwater reaches within the DMS was
generated from two sets of maps and information on stream size provided by the US
Forest Service. Stream reaches were stratified to sample two DMS treatments, three
DMS buffer treatments, and two slope classes. Twelve reaches from the various
locations were originally selected for sampling. However, due to time constraints,
sampling was completed on just eight reaches. Attributes of the sampled reaches are
summarized in Table 2.1. The two density management treatments were an unthinned
control with 200-350 trees per acres (TPA) and a moderate density retention where 6065 % of the stand had been thinned to 80 TPA. The three buffer widths were a two
site potential tree height buffer (2SPTH), a variable width buffer and an unthinned
control. The 2SPTH buffer is measured in slope distance from the stream center
(thalweg) and is based on the 50 year site index height of trees for each site, on
average 146 m. The variable radius buffer had a minimum width of 15.24 m, but
fluctuated based on sensitive areas, such as those prone to landslide or where
threatened species were present. The two slope classes were moderate, less than 30 %
or steep, greater than 30 % in slope.
Table 2. 1 Description of stream reaches sampled from the DMS site locations.
Location
Bottom Line
Keel Mountain
Keel Mountain
Keel Mountain
Keel Mountain
O.M. Hubbard
Ten High
Ten High
Reach
BLM
Number District
13
Eugene
17
Salem
18
Salem
19
Salem
21
Salem
36
Roseburg
46
Eugene
75
Eugene
Density
Moderate
Control
Moderate
Moderate
Moderate
Moderate
Control
Moderate
Slope
Buffer
% Slope Class Aspect
Two Tree Height
S NE/SW
51
Control
M
N/S
18
Two Tree Height
M NW/SE
21.2
Two Tree Height
M NW/SE
14
Variable Width
S
N/S
38
Variable Width
S NW/SE
31
Control
M
N/S
19
Variable Width
S
N/S
33
Once the reaches were selected, a sample plot was randomly located along
each of the headwater streams. The plot ran 72 m parallel to the stream and 36 meters
17
upslope, on each side of the centerline of the plot (Figure 2.2). Using GIS stream
files, the length of the stream was measured and a random number to the nearest meter
was generated for the placement of the centerline at the lower end of the block. The
random number was between 0 and 72 meters less the entire length of the stream. For
instance, if the stream were 683 meters in length, a random number would be chosen
between 0 and 611 meters. A laser was used to locate the random starting point for
each of the blocks along the stream. The 72 m square block was oriented with its
centerline running along the general azimuth of the stream. An example of block
layout can be seen in Figure 2.2.
Figure 2. 2 Illustration of plot layout on the sampled stream. One can see the
centerline of the 72 m square block was oriented in the general direction of the
stream. Sampling of trees took place 36 m from the centerline to upslope.
Stem mapping was done using Total Station Survey Equipment. Coordinates
for a control point (station) were found using a Trimble survey-grade GPS.
Coordinates were found at two points and the angle between the two points was hand
calculated for greater precision (estimates of precision for coordinates). This
calculation was then checked with a hand compass. For reaches where only one GPS
18
point was recorded, a staff compass was used to calculate the correct bearing. Stems
were numbered sequentially as they were mapped. Biodegradable flagging was used
to mark each tree that was surveyed with the number recorded in the total station.
Each tree number and species was recorded within the survey software program. A
reflector was held against the outside of the tree or within one foot of the center of the
tree. The distance to each tree (estimate of precision for distance) was calculated by
internal software based on the bearing from the station to the tree, and the latitude and
longitude of the station. When all trees that could be measured from a control point
were recorded, the station was moved to a new control point. The tripod and total
station were then setup at the new point and numbering of trees continued from the
last tree measured at the previous control point.
Tree measurements including diameter at breast height (DBH), species,
condition, canopy classification, crown class, and decay class were made and recorded
during stem mapping. Data were recorded on paper and then entered into an excel
worksheet. Coordinate data for the trees was recorded in a handheld computer
connected to the total station. Table 2.3 shows the parameters being measured and
recording protocol.
Table 2. 2 List of tree parameters measured on each stem mapped tree and the
recording protocol for each measurement.
Tree Attribute
Recording Protocol
Diameter at breast height
Nearest 1/10th centimeter
Species
2 letter common name code
Canopy Classification
Visually determined, 1 letter code
Condition
Visually determined, Dead (D), Live (L)
Crown Class
2 Letter code, visually determined, Table N
Decay Class
2 Letter code, visually determined, Table N
DBH was measured to the nearest tenth of a centimeter at 1.37 m (4.5 ft) above the
ground as measured on the uphill side of the tree. Measurements were made using a
19
diameter tape. Trees forking below diameter at breast height were counted as two
trees. Diameter was measured above any abnormalities such as bulging. Pistol butt or
leaning trees were measured at 1.37 m along the stem. Diameter was measured on all
trees over 7.5 cm in diameter. Tree species was recorded using a two letter common
name code. Trees were defined as woody stems taller than 20 feet or larger than 7.62
cm DBH. The condition of the tree was determined visually by the surveyor. A dead
tree was one that had no green foliage: a live tree had at least some remaining green
foliage.
SAS (v. 9.1, SAS Institute 1990) was used to simulate sixteen sampling
alternatives. Illustrations of each of the alternatives can be found in the Appendix.
Dead trees and hardwoods were removed from the dataset and the sampling designs
were evaluated on live conifer trees only. The hardwood component of each reach
was minimal at most of the eight reaches. In order to avoid incomplete plots landing
along edges of the stem mapped area (edge effect), the data was wrapped edge to end
(Appendix). Each design was simulated for a 10 % (0.1*722 = 518.4 m2) and 20 %
(0.2 *722 = 1036.8 m2) intensity.
The sixteen sampling alternatives can be grouped into six sampling designs: SRS,
SYRS, STRS, two-stage sampling, HLS, and sector sampling. Every alternative was
simulated 500 times with a different random starting point for each simulation.
Circular plots with a 5.64 m and 9 m radius were the only plot size and shape that used
SRS. The plot center coordinates were randomly selected and the number of plots for
each of the plot radius sizes can be seen in Table 2.3.
SRS was used for circular fixed are plots. SYRS was used for circular as well as
several strip sampling alternatives. The alternating strip option, ASTP3 and ASTP7 in
Table 2.1 is unique. This design was simulated by breaking the 72 meter grid into
either 3.6 m or 7.2 m wide strips which were 36 meters in length. All strips were
oriented perpendicular to the centerline of the plot (Figure 2.2), approximately
perpendicular to the stream. Each strip consisted of a rectangular plot on one side of
the stream and the rectangular strip diagonal to it on the opposite side of the stream.
20
For example, one could think of this as running a transect perpendicular to the stream
and sampling a 3.6 m strip to the left of the transect line on one side of the stream, and
all trees within 3.6 m to the right of the transect line on the opposite side of the stream.
STRS was used to sample strips running parallel to the stream, but also with at least
one strip close to the stream, and another upslope. In the case of the 9 m strip width,
the 72 meter grid was first split into 9 m strips running parallel to the stream, strata 1
was the two outermost strips on either side of the stream and strata 2 was the four
innermost strips.
Two stage sampling was used to simulate the sampling of only one side of the
stream using the strip sampling technique. In the first stage, the side of the stream
being sampled was selected. In the second stage, the side of the stream selected was
broken into strips and then these strips were systematically sampled. Two strip
widths, 3.6 m and 7.2 m were used to simulate this alternative. Each stream side was
sampled the same number of times.
The HLS alternative was selected to include variable radius plots in the
comparison. Each transect for the HLS design began at the centerline of the plot and
were placed perpendicular to the stream. Lynch’s (2006) equations were used to
expand the data from each transect(s) to a per hectare basis. Two BAF were chosen
based on the approximate amount of area that was sampled. Unlike the other
alternatives that ran 36 m perpendicular to the stream, this alternative sampled at 22 m
and 27 m upslope from the stream for the 8 BAF and 10 BAF alternatives
respectively.
The sector sampling design was adapted from Iles and Smith, (2006). In the
alternatives simulated in this study, the sector was essentially a fixed area plot to
simplify area calculations. First, a 36 m radius fixed area circular plot was selected
randomly within the 72 meter grid. Next, two to four sectors of either 11.46 or 22.92
degrees were sampled depending on the alternative. Sectors were sampled based on a
bearing that was randomly generated as a starting point for the first plot. Remaining
plots were systematically sampled. In the 20 % intensity design for the 11.46 degree
21
sized sector, two 36 m fixed area plots were placed within the 72 m grid and four
sectors were sampled from each of the circular plots.
Three methods were used to evaluate each of the sampling designs compared to
the actual stem map. The designs were evaluated based on how well they predicted
trees per hectare (TPH), basal area per hectare (BAPH) and height to diameter ratio
(H/D) using root mean square error (RMSE), absolute percent bias (APB), and mean
absolute deviation (MAD). Height measurements were not taken and therefore a
height-diameter equation was used in calculating height to diameter ratio. The
Chapman-Richards function (Richards 1959) as applied by Garman et al. (1995) for
western Oregon species was used.
The RMSE was computed as:

^


Y
Y
−
k
∑


k =1 


500
2
500
RMSE =
[1]
^
where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
All the selected sampling alternatives are known to be unbiased, so the traditional
measure of bias was not computed. Instead, an absolute value of the cumulative bias
was used. It is the average of the absolute value of the percent bias for each
replication. The absolute value does not allow one to know if the alternative is over or
underestimating the mean, but only by how much. The equation used to estimate
absolute percent bias is as follows:
APB =
^
Y k −Y
∑
Y
k =1
500
500
^
where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
[2]
22
MAD was used as a measure of variation from the mean of the 72 meter square
subpopulations. The equation used to estimate MAD is as follows:
500 ^
∑Y
k
−Y
k =1
MAD =
500
^
where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
[3]
23
Table 2. 3 Description of the sampling alternatives examined in this study including
the number of plots simulated at the 10 and 20% sampling intensity. Under
“Shape”, R = Rectangular, C = Circular, and T =Transect, and Ra = Radial
plot.
Sampling
Alternative
FAP5R
FAP9R
FAP5S
FAP9S
ASTP3
ASTP7
Description
Random Fixed Area
Plots
Random Fixed Area
Plots
Systematic Fixed
Area Plots
Systematic Fixed
Area Plots
Alternate
Perpendicular Strips
Alternate
Perpendicular Strips
PEST3
Perpendicular Strips
PEST7
Perpendicular Strips
OSSP3
OSSP7
PAST3
PAST9
HLS08
HLS10
Perpendicular, One
Side Only
Perpendicular, One
Side Only
STRS Parallel Strip
Sampling
STRS Parallel Strip
Sampling
Horizontal Line
Sampling
Horizontal Line
Sampling
SEC11
Sector Sampling
SEC22
Sector Sampling
Plots
Plots
(10%) (20%)
Size
5.64 m
radius
Shape
C
5
10
9 m radius
C
2
4
5.64 m
radius
C
5
10
9 m radius
C
2
4
R
4
8
R
2
4
R
2
4
R
1
2
R
4
8
R
2
4
3.6m x 36 m
R
4
8
9 m x 28.8
m
R
2
4
BAF 8
22 m T
2
4
BAF 10
27 m T
2
4
11.46
Degrees
22.92
Degrees
36 m
Ra
36 m
Ra
4
8
2
4
3.6 m by 36
m
7.2 m by 36
m
3.6 m x 72
m
7.2 m x 72
m
3.6 m x 36
m
7.2 m x 36
m
24
Results
Tables 2.4 to 2.6 present the results from the analysis of the sixteen sampling
alternatives when evaluated using RMSE, APB and MAD. Sampling alternatives that
had among the smallest five values are highlighted in gray. From Tables 2.4 to 2.6,
one can see that ASTP3, ASTP7, PAST3, PEST3 and PEST7 frequently performed
better when compared to the other alternatives examined. Tables 2.7 to 2.9 show the
two sampling alternatives with the smallest RMSE, APB and MAD for trees per
hectare (TPH), basal area per hectare (BAPH) and height to diameter ratio (H/D). The
sampling alternatives that performed well at individual locations included ASTP3,
ASTP7, PAST3, PEST3, and SEC11. PEST7 and PAST9 also performed well, but not
as frequently as other sampling alternatives.
From Table 2.4 one can see that the RMSE values for TPH ranged from 64.3 to
261.8 at the 10% sampling intensity and from 40.4 to 261.6 for the 20% intensity. The
alternatives with the smallest values were ASTP3 and PEST3. The RMSE values for
BAPH (m2) ranged from 8.5 to 28.3 for the 10% sampling intensity and 5.6 to 28.4 for
the 20% sampling intensity. The sampling alternatives with the smallest values were
the ASTP3 and ASTP7. The RMSE values for the height to diameter ratio ranged
from 1.3 to 2.4 for the 10% sampling intensity and from 0.8 to 1.9 for the 20%
intensity. The sampling alternatives which most closely estimated the height to
diameter ratio was the ASTP3 and PEST3 alternatives. The standard deviation for the
mean RMSE values are also displayed in Table 2.4. One can see that the variation is
quite large for the HLS08, HLS10, and OSSP3 alternatives.
From the values in Table 2.5 one can see that ASTP3, ASTP7, PAST3, and
PEST3 alternatives had the smallest APB when estimating TPH, BAPH and the H/D
ratio. The standard deviation around the mean was fairly large among some of the
alternatives including the FAP5S and FAP9S alternatives. The values ranged from
15.8% to 61.4% and 9.7% to 61.3% when estimating TPH at the 10 and 20% intensity
respectively. When estimating BAPH, values ranged from 16.8% to 56.0% for the
10% intensity and 11.0% to 56.1% for the 20% intensity. The alternatives that
25
performed well when estimating the height to diameter ratio were the ASTP3, ASTP7,
PAST3, PEST3, SEC11, FAP5S and PEST7 alternatives. Values ranged from 1.7% to
3.0 % for the 10 % sampling intensity and from 1.0 % to 3.0 % for the 20 % intensity.
Sampling alternatives that performed well when evaluated using MAD were
ASTP3, ASTP7, PAST3, PEST3, PEST7, and SEC11. The values ranged from 64.3
to 261.8 when estimating TPH at the 10 % intensity and from 40.4 to 261.6 for the
20% intensity. Sampling alternatives that did not perform well included HLS08,
HLS10, and FAP9S. When evaluated for their ability to estimate BAPH, values
ranged from 8.5 to 28.3 at the 10% intensity and 5.6 to 28.4 at the 20% intensity.
When estimating the height to diameter ratio, values ranged from 1.3 to 2.3 and from
0.8 to 2.3 at the 20% intensity. The alternative that performed best overall was the
ASTP3.
From Table 2.7 to 2.9 one can see that although the sector sampling alternative
using the 11.46 degree sector (SEC11) did not appear to perform well overall, but
performed well several times at the different locations. The location codes 113, 317,
418, 519, 621, 736, 846, and 875 represent the Bottom Line 13, Keel Mountain 17,
Keel Mountain 18, Keel Mountain 19, Keel Mountain 21, OM Hubbard 36, Ten High
46, and Ten High 75 reaches respectively. The alternative that performed well at the
eight locations the greatest number of times was the ASTP3 alternative. Other
alternatives that performed well were SEC11 and ASTP7. Sampling alternatives that
performed well at multiple sites when estimating BAPH were the PEST3 and ASTP3
alternatives. Several alternatives performed well when estimating TPH. PAST3,
ASTP3, ASTP7, and PEST3 all performed well at multiple sites. The majority of the
sampling alternatives that performed well at more than one location also performed
well when evaluated using RMSE, APB and MAD. These included ASTP3, ASTP7,
PAST3, PEST3, and PEST7.
26
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
Intensity
Sampling
Alternative
Table 2. 4 Summary of sampling alternatives evaluated using RMSE. Values under
trees per hectare (TPH), basal area per hectare (BAPH, m2) and height to
diameter ratio (H/D) are the mean RMSE values from the eight locations.
Shaded values are the five smallest. “SD” is the standard deviation of the
mean RMSE.
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
TPH
64.3
71.0
83.2
105.1
94.3
115.2
261.8
260.0
211.2
91.0
66.6
94.0
64.5
70.3
90.4
90.1
40.4
41.5
60.2
51.1
87.3
101.9
261.6
260.5
214.0
55.0
45.9
49.7
42.3
46.1
53.2
71.2
BAPH
(m2)
8.5
8.5
10.5
12.6
11.3
13.7
28.3
27.9
25.4
12.0
9.0
11.6
9.1
9.4
11.1
10.5
5.6
5.6
7.5
6.2
9.4
10.5
28.4
28.0
25.6
6.7
6.1
6.7
6.2
6.0
7.0
8.0
H/D
Ratio SD TPH
1.3
11.5
1.7
17.0
1.7
15.5
1.7
29.1
1.9
16.6
2.3
21.7
2.4
80.8
2.2
82.7
2.3
53.9
1.9
45.1
1.5
20.6
2.0
27.0
1.4
11.3
1.7
16.5
1.7
17.3
1.8
12.7
0.8
14.2
1.0
7.6
1.2
11.1
0.9
23.5
2.3
22.5
2.2
31.3
1.9
81.1
1.8
82.5
1.8
53.9
1.5
21.8
1.1
14.9
1.2
15.7
0.9
18.7
1.0
9.1
1.2
12.9
1.3
14.2
SD
BAPH SD H/D
(m2)
Ratio
1.9
0.5
2.1
0.8
1.6
0.5
3.5
0.5
2.1
0.8
2.9
1.1
6.5
0.8
6.0
0.9
5.3
0.8
8.5
0.8
1.1
0.6
2.5
0.8
2.3
0.5
1.9
0.8
1.6
0.8
1.9
0.7
1.8
0.3
1.5
0.5
1.2
0.4
1.3
0.2
1.9
1.1
3.1
1.1
6.6
0.9
6.0
0.9
5.4
0.8
2.5
0.9
0.9
0.5
1.2
0.6
2.5
0.4
1.5
0.4
1.3
0.9
1.7
0.8
27
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
Intensity
Sampling
Alternative
Table 2. 5 Summary of performance of sampling alternatives evaluated using APB.
Values under trees per hectare (TPH, %), basal area per hectare (BAPH, %)
and height to diameter ratio (H/D, %) are the mean APB values from the eight
locations. Shaded values are the five smallest percentages. “SD” is the
standard deviation of the mean APB for TPH, BAPH, and H/D.
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
TPH
15.9%
17.2%
20.1%
25.6%
22.8%
28.2%
61.4%
60.7%
49.7%
22.1%
16.0%
22.5%
15.8%
17.0%
22.3%
22.2%
9.7%
10.1%
14.5%
11.8%
21.2%
25.0%
61.3%
60.8%
50.4%
13.0%
11.0%
11.9%
9.9%
11.4%
13.3%
17.7%
BAPH
(m2)
16.8%
17.1%
21.0%
25.7%
22.7%
27.6%
56.0%
55.2%
50.1%
22.4%
18.4%
23.3%
17.8%
18.9%
22.7%
21.3%
11.0%
11.0%
15.0%
12.5%
18.6%
21.4%
56.1%
55.3%
50.5%
13.3%
12.5%
13.8%
12.1%
11.7%
14.1%
16.3%
H/D
Ratio SD TPH
1.7%
4.6%
2.2%
4.8%
2.2%
3.1%
2.2%
8.2%
2.5%
3.6%
3.0%
6.2%
3.1%
5.8%
2.9%
5.4%
3.0%
0.4%
2.5%
11.6%
1.9%
4.2%
2.5%
5.4%
1.8%
3.8%
2.2%
4.5%
2.2%
6.6%
2.3%
5.8%
1.0%
2.6%
1.2%
2.4%
1.6%
1.9%
1.2%
3.4%
3.0%
5.7%
2.9%
8.3%
2.4%
5.7%
2.3%
5.3%
2.3%
0.3%
2.0%
3.8%
1.4%
3.0%
1.5%
3.7%
1.2%
3.0%
1.3%
3.6%
1.5%
5.3%
1.7%
5.8%
SD
BAPH SD H/D
Ratio
(m2)
2.5%
0.6%
4.5%
1.1%
2.7%
0.7%
8.9%
0.7%
3.4%
1.1%
5.2%
1.5%
6.2%
1.1%
4.7%
1.1%
0.5%
1.1%
11.1%
1.1%
4.2%
0.7%
5.8%
1.1%
3.2%
0.7%
3.7%
1.1%
5.6%
1.0%
4.4%
0.9%
2.8%
0.4%
2.4%
0.6%
2.0%
0.5%
2.4%
0.3%
1.6%
1.4%
7.4%
1.4%
6.0%
1.1%
4.6%
1.1%
0.3%
1.1%
3.6%
1.2%
3.1%
0.6%
3.7%
0.8%
4.5%
0.6%
1.3%
0.5%
3.0%
1.1%
4.6%
1.0%
28
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
ASTP3
ASTP7
FAP5R
FAP5S
FAP9R
FAP9S
HLS08
HLS10
OSSP3
OSSP7
PAST3
PAST9
PEST3
PEST7
SEC11
SEC22
Intensity
Sampling
Alternative
Table 2. 6 Summary of performance of sampling alternatives evaluated using MAD.
Values under trees per hectare (TPH), basal area per hectare (BAPH m2) and
height to diameter ratio (H/D) are the MAD values from the eight locations.
Shaded values have the smallest MAD. “SD” is the standard deviation of the
mean APB for TPH, BAPH, and H/D.
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
20%
TPH
64.3
71.0
83.2
105.1
94.3
115.2
261.8
260.0
211.2
91.0
66.6
94.0
64.5
70.3
90.4
90.1
40.4
41.5
60.2
51.1
87.3
101.9
261.6
260.5
214.0
55.0
45.9
49.7
42.3
46.1
53.2
71.2
BAPH
(m2)
8.5
8.5
10.5
12.6
11.3
13.7
28.3
27.9
25.4
12.0
9.0
11.6
9.1
9.4
11.1
10.5
5.6
5.6
7.5
6.2
9.4
10.5
28.4
28.0
25.6
6.7
6.1
6.7
6.2
6.0
7.0
8.0
SD
BAPH SD H/D
H/D
Ratio SD TPH
Ratio
(m2)
1.3
11.5
1.9
0.5
1.7
17.0
2.1
0.8
1.7
15.5
1.6
0.5
1.7
29.1
3.5
0.5
1.9
16.6
2.1
0.8
2.3
21.7
2.9
1.1
2.4
80.8
6.5
0.8
2.2
82.7
6.0
0.9
2.3
53.9
5.3
0.8
1.9
45.1
8.5
0.8
1.5
20.6
1.1
0.6
2.0
27.0
2.5
0.8
1.4
11.3
2.3
0.5
1.7
16.5
1.9
0.8
1.7
17.3
1.6
0.8
1.8
12.7
1.9
0.7
0.8
14.2
1.8
0.3
1.0
7.6
1.5
0.5
1.2
11.1
1.2
0.4
0.9
23.5
1.3
0.2
2.3
22.5
1.9
1.1
2.2
31.3
3.1
1.1
1.9
81.1
6.6
0.9
1.8
82.5
6.0
0.9
1.8
53.9
5.4
0.8
1.5
21.8
2.5
0.9
1.1
14.9
0.9
0.5
1.2
15.7
1.2
0.6
0.9
18.7
2.5
0.4
1.0
9.1
1.5
0.4
1.2
12.9
1.3
0.9
1.3
14.2
1.7
0.8
Performed
Well
Performed
Poorly
Performed
Well
APB
Performed
Poorly
Performed
Well
RMSE
HLS10 HLS08 PEST3 ASTP3 HLS10 HLS08 FAP5R SEC11 HLS10 HLS08 PEST3 ASTP3
HLS10 HLS08 ASTP3 PAST3 HLS10 HLS08 ASTP3 PAST3 HLS10 HLS08 ASTP3 PAST3
HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3
HLS08 HLS10 PAST3 PAST9 HLS08 HLS10 PAST3 PAST9 HLS08 HLS10 PAST3 PAST9
HLS10 HLS08 ASTP7 ASTP3 HLS10 HLS08 ASTP7 ASTP3 HLS10 HLS08 ASTP7 ASTP3
HLS08 HLS10 SEC11 PAST9 HLS08 HLS10 SEC11 PAST9 HLS08 HLS10 SEC11 PAST9
HLS08 HLS10 FAP5S SEC11 HLS08 HLS10 FAP5S SEC11 HLS08 HLS10 FAP5S SEC11
HLS08 HLS10 FAP5S ASTP3 HLS08 HLS10 FAP5S ASTP3 HLS08 HLS10 FAP5S ASTP3
HLS10 HLS08 ASTP3 PAST3 HLS10 HLS08 ASTP3 PAST3 HLS10 HLS08 ASTP3 PAST3
HLS10 HLS08 ASTP7 PEST7 HLS10 HLS08 ASTP7 PEST7 HLS10 HLS08 ASTP7 PEST7
HLS10 HLS08 ASTP7 PEST3 HLS10 HLS08 ASTP7 PEST3 HLS10 HLS08 ASTP7 PEST3
HLS08 HLS10 PEST7 PAST3 HLS08 HLS10 PEST7 PAST3 HLS08 HLS10 PEST7 PAST3
HLS10 HLS08 PEST3 ASTP3 HLS10 HLS08 PEST3 ASTP3 HLS10 HLS08 PEST3 ASTP3
HLS08 HLS10 PAST3 PEST3 HLS08 HLS10 PAST3 PEST3 HLS08 HLS10 PAST3 PEST3
HLS08 HLS10 PEST3 PAST3 HLS08 HLS10 PEST3 PAST3 HLS08 HLS10 PEST3 PAST3
Location
Intensity
HLS08 HLS10 PAST3 PAST9 HLS08 HLS10 PAST3 PAST9 HLS08 HLS10 PAST3 PAST9
Performed
Poorly
MAD
29
Table 2. 7 Summary of alternatives that performed well when estimating trees per
hectare (TPH). The two alternatives that performed the best at each of the
locations as well as the two alternatives that performed the worst are listed
below when evaluated using root mean square error (RMSE), absolute percent
bias (APB) and mean absolute deviation (MAD). The location codes 113, 317,
418, 519, 621, 736, 846, and 875 represent the Bottom Line 13, Keel Mountain
17, Keel Mountain 18, Keel Mountain 19, Keel Mountain 21, OM Hubbard 36,
Ten High 46, and Ten High 75 reaches respectively.
113 317 418 519 621 736 846 875 113 317 418 519 621 736 846 875
10% 10% 10% 10% 10% 10% 10% 10% 20% 20% 20% 20% 20% 20% 20% 20%
Performed
Well
Performed
Poorly
Performed
Well
APB
Performed
Poorly
Performed
Well
RMSE
HLS10 OSSP3 ASTP3 PEST3 HLS10 OSSP3 ASTP3 PEST3 HLS10 OSSP3 ASTP3 PEST3
OSSP3 HLS10 ASTP7 PAST3 OSSP3 HLS10 ASTP7 PAST3 OSSP3 HLS10 ASTP7 PAST3
HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3
HLS08 OSSP3 PEST3 ASTP7 HLS08 OSSP3 PEST3 ASTP7 HLS08 OSSP3 PEST3 ASTP7
HLS10 OSSP3 OSSP7 ASTP3 HLS10 OSSP3 OSSP7 ASTP3 HLS10 OSSP3 OSSP7 ASTP3
HLS08 HLS10 ASTP7 PAST9 HLS08 HLS10 ASTP7 PAST9 HLS08 HLS10 ASTP7 PAST9
HLS08 HLS10 PEST3 SEC11 HLS08 HLS10 PEST3 SEC11 HLS08 HLS10 PEST3 SEC11
HLS08 HLS10 PEST7 ASTP3 HLS08 HLS10 PEST7 ASTP3 HLS08 HLS10 PEST7 ASTP3
HLS10 OSSP3 PEST3 ASTP3 HLS10 OSSP3 PEST3 ASTP3 HLS10 OSSP3 PEST3 ASTP3
OSSP3 HLS10 PAST3 ASTP7 OSSP3 HLS10 PAST3 ASTP7 OSSP3 HLS10 PAST3 ASTP7
HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3 HLS10 HLS08 ASTP3 PEST3
HLS08 OSSP3 ASTP7 PEST3 HLS08 OSSP3 ASTP7 PEST3 HLS08 OSSP3 ASTP7 PEST3
HLS10 OSSP3 ASTP3 OSSP7 HLS10 OSSP3 ASTP3 OSSP7 HLS10 OSSP3 ASTP3 OSSP7
HLS08 HLS10 ASTP3 ASTP7 HLS08 HLS10 ASTP3 ASTP7 HLS08 HLS10 ASTP3 ASTP7
HLS08 HLS10 ASTP7 PAST3 HLS08 HLS10 ASTP7 PAST3 HLS08 HLS10 ASTP7 PAST3
Location
Intensity
HLS08 HLS10 OSSP7 PAST9 HLS08 HLS10 OSSP7 PAST9 HLS08 HLS10 OSSP7 PAST9
Performed
Poorly
MAD
30
Table 2. 8 Summary of alternatives that performed well when estimating basal area
per hectare (BAPH). The two alternatives that performed the best at each of
the locations as well as the two alternatives that performed the worst are listed
below when evaluated using root mean square error (RMSE), absolute percent
bias (APB) and mean absolute deviation (MAD). The location codes 113, 317,
418, 519, 621, 736, 846, and 875 represent the Bottom Line 13, Keel Mountain
17, Keel Mountain 18, Keel Mountain 19, Keel Mountain 21, OM Hubbard 36,
Ten High 46, and Ten High 75 reaches respectively.
113 317 418 519 621 736 846 875 113 317 418 519 621 736 846 875
10% 10% 10% 10% 10% 10% 10% 10% 20% 20% 20% 20% 20% 20% 20% 20%
Performed
Well
Performed
Poorly
Performed
Well
APB
Performed
Poorly
Performed
Well
RMSE
FAP9S HLS08 PEST7 PEST3 FAP9S HLS08 PEST7 PEST3 FAP9S HLS08 PEST7 PEST3
OSSP3 OSSP7 PEST3 ASTP3 OSSP3 OSSP7 PEST3 ASTP3 OSSP3 OSSP7 PEST3 ASTP3
FAP9R FAP9S ASTP7 ASTP3 FAP9R FAP9S ASTP7 ASTP3 FAP9R FAP9S ASTP7 ASTP3
OSSP7 OSSP3 SEC11 PAST3 OSSP7 OSSP3 SEC11 PAST3 OSSP7 OSSP3 SEC11 PAST3
FAP9S FAP9R ASTP7 PEST3 FAP9S FAP9R ASTP7 PEST3 FAP9S FAP9R ASTP7 PEST3
FAP9R FAP9S SEC11 FAP5S FAP9R FAP9S SEC11 FAP5S FAP9R FAP9S SEC11 FAP5S
FAP9S FAP9R PAST3 ASTP7 FAP9S FAP9R PAST3 ASTP7 FAP9S FAP9R PAST3 ASTP7
FAP9S FAP9R FAP5S ASTP3 FAP9S FAP9R FAP5S ASTP3 FAP9S FAP9R FAP5S ASTP3
OSSP3 HLS08 PEST7 PEST3 OSSP3 HLS08 PEST7 PEST3 OSSP3 HLS08 PEST7 PEST3
OSSP3 HLS08 PEST3 ASTP3 OSSP3 HLS08 PEST3 ASTP3 OSSP3 HLS08 PEST3 ASTP3
HLS08 FAP5S ASTP7 ASTP3 HLS08 FAP5S ASTP7 ASTP3 HLS08 FAP5S ASTP7 ASTP3
FAP9S OSSP3 SEC11 PAST3 FAP9S OSSP3 SEC11 PAST3 FAP9S OSSP3 SEC11 PAST3
FAP9S HLS08 OSSP7 ASTP3 FAP9S HLS08 OSSP7 ASTP3 FAP9S HLS08 OSSP7 ASTP3
OSSP3 FAP9S ASTP3 PAST3 OSSP3 FAP9S ASTP3 PAST3 OSSP3 FAP9S ASTP3 PAST3
HLS08 OSSP7 ASTP3 ASTP7 HLS08 OSSP7 ASTP3 ASTP7 HLS08 OSSP7 ASTP3 ASTP7
SEC11 HLS10 FAP5S PAST9 SEC11 HLS10 FAP5S PAST9 SEC11 HLS10 FAP5S PAST9
Performed
Poorly
MAD
31
Table 2. 9 Summary of alternatives that performed well when estimating the height to
diameter ratio (H/D). The two alternatives that performed the best at each of
the locations as well as the two alternatives that performed the worst are listed
below when evaluated using root mean square error (RMSE), absolute percent
bias (APB) and mean absolute deviation (MAD). The location codes 113, 317,
418, 519, 621, 736, 846, and 875 represent the Bottom Line 13, Keel Mountain
17, Keel Mountain 18, Keel Mountain 19, Keel Mountain 21, OM Hubbard 36,
Ten High 46, and Ten High 75 reaches respectively.
Location
113 317 418 519 621 736 846 875 113 317 418 519 621 736 846 875
Intensity
10% 10% 10% 10% 10% 10% 10% 10% 20% 20% 20% 20% 20% 20% 20% 20%
32
Discussion
The sixteen sampling alternatives can be classified into eight different sampling
designs. These are circular fixed area plots (FAP5R, FAP9R, FAP5S, FAP5R), SYRS
rectangular plots (PEST3, PEST7, ASTP3, and ASTP7), stratified (PAST3, PAST9),
two-stage sampling (OSSP3, OSSP7), HLS (HLS08, HLS10) and sector sampling
(SEC11, SEC22). The designs that performed best were generally those that sampled
perpendicular to the stream. The best performing designs were the SYRS designs that
used rectangular shaped plots. However, the STRS designs (PAST3, PAST9) also
performed well. The alternatives that performed poorly in this simulation were the
OSSP3, FAP9R, FAP9S, HLS10, and HLS08 alternatives. Overall, the ASTP3
alternative performed the best among the alternatives. However, the PEST3, ASTP7,
and PAST3 alternatives also performed well.
In this comparison of alternatives, the circular fixed area plots did not perform as
well as the rectangular plots. The smaller 5.64m radius plots performed better than the
9 m radius plots. The FAP5S performed best among the circular fixed area plots at the
20% intensity, but the FAP5R alternative outperformed the other circular fixed area
plots at the 10% intensity. The designs performed similarly whether evaluated using
MAD, APB, or RMSE. The poor performance of the other designs could be attributed
to plots falling entirely within gaps in the canopy or under-representing conifers
upslope from the stream.
The STRS design performed better than all other designs except for the rectangular
SYRS designs. The PAST3 alternative performed better than the PAST9 alternative
overall. The alternative performed similarly in estimating TPH, BAPH or H/D.
Because the rectangular strips of PAST3 and PAST9 were oriented parallel to the
stream, one would have expected the alternatives to perform very poorly if there was
one prominent gradient within riparian areas that existed from the stream to upslope.
Even though strips were placed using stratified sampling, the performance of this
designs indicates that there are other sources of variation as one moves upstream in
addition to the strong gradient that exists as one moves from the stream to upslope.
33
The SYRS designs performed the best among the eight designs. The alternatives
within the SYRS design category often performed well when estimating TPH, BAPH
and height diameter ratio. Although ASTP3 alternative performed the best at the most
locations and had the lowest amount of error, PEST3, PEST7 and ASTP7 also
performed very well. These alternatives performed similarly when evaluated using
RMSE, APB, and MAD. The alternatives performed well at multiple locations which
can be seen in Tables 2.7 to 2.9. Overall, it appears that these designs were able to
capture the riparian vegetation better than the other alternatives.
The two-stage designs performed inconsistently throughout this analysis. The
OSSP7 alternative (7.2 m wide strips) performed well at several locations, but the
OSSP3 alternative (3.6 m wide strips) did not perform well at any of the locations.
One would expect that stand attributes would differ markedly for opposing north and
south facing slopes bounding a steeply incised stream. About half of the locations
sampled appeared in the field to have fairly symmetrical selected stand attributes each
side of the stream. The poor performance of the OSSP3 alternative shows that when a
source of variation within a riparian area is ignored, sampling designs perform poorly.
Overall, this design is not recommended for use in sampling selected stand attributes.
Even though the HLS alternatives did perform well when estimating height to
diameter ratio, the rectangular strips were more consistent. Unlike other alternatives,
the HLS alternatives did not sample the full 36 m upslope from the stream on either
side of the stream. Schreuder et al. (1987) suggested that using longer lines with a
larger angle gauge will lead to more accurate samples. If one was going to evaluate
this design again, increasing the transect length to 36 m would be recommended.
Even though the HLS10 had the smallest error at the 20 % intensity, overall it was not
among the best performing designs. Kenning (2005) found that a modified horizontal
sampling design was more efficient in sampling basal area compared to fixed area
sampling. Several research studies have found this design to be timely and perform
well (Kenning et al., 2005, Schreuder et al., 1987, Schreuder et al., 1992), especially in
34
forests with larger diameter trees. Had a longer transect line been used in this study,
one would suspect it would have been able to perform among the top designs.
The modified sector sampling design was suggested for use when sampling small
clusters of objects by Iles & Smith (2006). The sector sampling alternative using a
11.46 wide angle performed better than the alternative using a 22.92 wide angle. One
reason this alternative may not have performed as well as the strip sampling
alternatives could have been the random selection of the azimuth. There was potential
for only the trees closest to the stream to be sampled rather than sampling from the
stream to upslope. It may have been better to orient the sectors opposite each other
with the center point in the stream. From Tables 2.7 to 2.9 one can see that this
alternative did perform well at several of the locations when evaluated using RMSE,
APB, and MAD. This alternative was not among the worst alternatives, but was
outperformed by ASTP3, ASTP7, PEST3 and PEST 7.
Although this study did not specifically focus on plot size, there were some trends
that emerged. The fixed area circular shaped plots performing the best were 5.6 m
radius in size. The larger 9 m plot size may not have been as efficient in capturing the
variation occurring parallel to the stream. In general, PAST3 performed better than
PAST9. Because this was a stratified design, there were more 3 m wide strips to
choose from and strips landing on or in the stream may have been sampled less
frequently. In general, PEST3 and ASTP3 performed better than the ASTP7 and
PEST7 alternative. One would expect the larger strip width to perform better. In a
prior study, Lynch (2003) found that as plot size increased from 0.01 to 0.05 ha,
precision increased, but did not increase enough to justify the larger plot size which
was consistent with the findings of Maclean & Ostaff (1983). The OSSP7 alternative
performed much better than the OSSP3 alternative. Nelson et al. (1998) found that
widths of rectangular plots best for measuring forest canopy height were at least 6 m
to 8 m in width. It should be noted that there were a limited number of plot sizes
evaluated for each of the sampling alternatives and that the stream size and location
likely played a role in the results of each alternative.
35
Throughout this study, rectangular plots performed better than other shapes.
When comparing circular fixed area plots to rectangular plots, Johnson & Hixon
(1952) found that rectangular plots outperformed circular plots. The ASTP3, ASTP7,
PEST3, and PEST7 generally had the smallest difference from the actual mean at each
site. The only exception was the performance of HLS for estimating height to
diameter ratio. Husch et al., (2003, p. 329) note that “it is desirable to orient strips at
right angles to the drainage pattern in order to increase the likelihood of having the
strip intersect all stand conditions.” The alternatives using circular fixed area plots
may not have done a good job of intersecting the different stand conditions when
moving from stream to upslope. Despite the circular fixed area alternatives
performing well in some cases, these alternatives were much less consistent than the
rectangular alternatives in accurately estimating TPH, BAPH, and H/D.
The results of this study indicate that the source of spatial variation most
pivotal to determining the effectiveness of a sampling design occurred along slopes
running upslope from the stream. It is difficult for a sampling alternative to provide
reliable estimates of TPH, BAPH, or H/D without capturing variation gradients
present within riparian areas. In comparing PEST3, PEST7, PAST3 and PAST9, one
can see that PEST3 and PEST7 did better at estimating TPH, BAPH and H/D. PAST3
alternative did perform well when estimating TPH, but was inconsistent in estimating
BAPH and H/D at the individual locations (Tables 2.7-2.9). In addition, alternatives
that did not sample both sides of the stream were outperformed by those that sampled
from both sides of the stream. This can be seen in the performance of OSSP3 and
OSSP7 compared to PEST3 and PEST7. ASTP3 and ASTP7 performed similarly to
PEST3 and PEST7. However, PEST3 and ASTP3 performed well at a variety of
locations which can be seen in Tables 2.7 to 2.9. Variation running parallel to the
stream impacted the performance of the sampling alternatives and alternating the strips
across the stream led to more accurate estimates. In addition, the smaller strip widths
may have captured the patchiness of the trees within this location. Overall, the ASTP3
alternative performed better than the other alternatives examined.
36
Conclusion
From this analysis, one can see that there is a high amount of variation within
riparian areas, even those located relatively close to each other. The four Keel
Mountain sites were very similar, but from Tables 2.7 to 2.9 one can see that different
alternatives performed well at the 317, 418, 519 and 621 locations. Although some
alternatives performed quite a lot better than others, the standard deviation around the
mean was quite large for some of the alternatives. Overall, rectangular strip plots
outperformed all other plot shapes. Plot sizes varied from 3.6 m in width to 9 m in
width. In general the narrower strip widths performed better than the wider strip
widths. Although the wedge shaped sector sampling alternative did perform well at
some of the locations, it did not perform among the most accurate alternatives overall.
The circular shaped plots performed better than the alternatives that used horizontal
line sampling. In addition, one would have expected a greater decrease in error as one
moved from the 10% to 20% sampling intensity.
This analysis was conducted using small scale stem maps from eight headwater
streams in western Oregon. Under these conditions, the ASTP3 alternative was found
to be the most accurate and was able to perform well at the eight locations. This
alternative consisted of rectangular plots systematically sampled. The rectangular
plots were offset 3.6m wide strips each side of the stream. The other alternatives that
performed well were the ASTP7, PEST3, PEST7, and PAST3 alternatives. The
alternatives that did not perform well included the circular fixed area plots that were
9m in radius. In general, strips oriented perpendicular to the stream outperformed
those oriented parallel to the stream.
37
References
Acker, S. A., Gregory, S. V., Lienkaemper, G., McKee, W. A., Swanson, F. J., &
Miller, S. D. 2003. Composition, complexity, and tree mortality in riparian
forests in the central Western Cascades of Oregon. For. Ecol. Manage., 173:
293-308.
Aguirre, O., Hui, G., von Gadow, K., & Jimenez, J. 2003. An analysis of spatial forest
structure using neighbourhood-based variables. For. Ecol. Manage., 183: 137145.
Anderson, P. D., Larson, D. J., & Chan, S. S. 2007. Riparian buffer and density
management influences on microclimate of young headwater forests of western
Oregon. For. Sci., 53(2): 254-269.
Chen, J., & Bradshaw, G. A. 1999. Forest structure in space: a case study of an old
growth spruce-fir forest in Changbaishan Natural Reserve, PR China. For.
Ecol. Manage., 120, 219-233.
Cissel, J., Anderson, P., Berryman, S., Chan, S., Puettman, K., and Thompson, C.
2006. BLM Density Management and Riparian Buffer Study: Establishment
Report and Study Plan. U.S. Dept. Interior, U.S. Geological Survey. 161 p.
Coble, D. W., & Grogan, J. 2007. Comparison of systematic line-point and double
sampling designs for pine and hardwood forests in the western gulf. South. J.
Appl. For., 31(4): 199-206.
Coroi, M., Skeffington, M. S., Giller, P., Smith, C., Gormally, M., & O'Donovan, G.
2004. Vegetation diversity and stand structure in streamside forests in the
south of Ireland. For. Ecol. Manage., 202: 39-57.
Dahl, C. A., Harding, B. A., & Wiant, H. V. 2008. Comparing double-sampling
efficiency using various estimators with fixed-area and point sampling. North.
J. Appl. For., 25(2): 99-102.
Ducey, M. J., Jordan, G. J., Gove, J. H., & Valentine, H. T. 2002. A practical
modification of horizontal line sampling for snag and cavity tree inventory.
Can. J. For. Res., 32: 1217-1224.
Garman, S. L., Acker, S. A., Ohmann, J. L., and Spies, T. A. 1995. Asymptotic
height-diameter equations for twenty-four tree species in western Oregon.
Forest Research Laboratory, Oregon State University, Corvallis. Research
Contribution 10. 22p.
Harper, K. A., & Macdonald, S. E. 2001. Structure and composition of riparian boreal
forest new methods for analyzing edge influence. Ecology. 82(3): 649-659.
Husch, B., Beers, T. W., & Kershaw, J. A. Jr. 2003. Forest Mensuration. 4th ed. John
Wiley & Sons, Inc., New Jersey. 443 p.
Iles, K., & Smith, N. J. 2006. A new type of sample plot that is particularly useful for
sampling small clusters of objects. For. Sci., 52(2): 148-154.
Johnson, F. A., & Hixon, H. J. 1952. The most efficient size and shape of plot to use
for cruising in old-growth Douglas-fir timber. J. Forestry, 50(1): 17-20.
38
Kenning, R. S., Ducey, M. J., Brissette, J. C., & Gove, J. H. 2005. Field efficiency and
bias of snag inventory methods. Can. J. For. Res., 35: 2900-2910.
Kint, V., Meirvenne, M. V., Nachtergale, L., Geudens, G., & Lust, N. 2003. Spatial
methods for quantifying forest stand structure development: A comparison
between nearest-neighbor indices and variogram analysis. For. Sci. 49(1): 3649.
Lhotka, J. M., & Loewenstein, E. F. 2006. Indirect measures for characterizing light
along a gradient of mixed-hardwood riparian forest canopy structures. For.
Ecol. Manage., 226, 310-318.
Lyon, J., & Gross, N. M. 2005. Patterns of plant diversity and plant-environmental
relationships across three riparian corridors. For. Ecol. Manage., 204: 267-278.
Lynch, A. M. 2003. Comparison of fixed-area plot designs for estimating stand
characteristics and western spruce budworm damage in southwestern U.S.A.
forests. Can. J. For. Res., 33: 1245-1255.
Lynch, T. B. 2006. Horizontal line sampling for riparian forests without land area
estimation. For. Sci. 52(2): 119-129.
Lindsey, A. A., Barton, J. D., & Miles, S. R. 1958. Field efficiencies of forest
sampling methods. Ecology, 39(3): 428-444.
Maclean, D. A., & Ostaff, D. P. 1983. Sample size - precision relationships for use in
estimating stand characteristics and spruce budworm caused tree mortality.
Can. J. For. Re., 13: 548-555.
Mason, W. L., Connolly, T., Pommerening, A., & Edwards, C. 2007. Spatial structure
of semi-natural and plantation stands of Scots pine (Pinus sylvestris L.) in
northern Scotland. Forestry, 80(5): 567-586.
Naiman, R. J., Decamps, H., & Pollock, M. 1993. The role of riparian corridors in
maintaining regional biodiversity. Ecol. Appl., 3(2): 209-212.
Nierenberg, T. R., & Hibbs, D. E. 2000. A characterization of unmanaged riparian
areas in the central Coast Range of western Oregon. For. Ecol. Manage., 129:
195-206.
Nelson, R. F., Gregoire, T. G., & Oderwald, R. G. 1998. The effects of fixed-area plot
width on forest canopy height simulation. For. Sci., 3(44): 438-444.
Olson, D. H., & Weaver, G. 2007. Vertebrate assemblages associated with headwater
hydrology in western Oregon managed forests. For. Sci., 52(2): 343-355.
Pabst, R. J., & Spies, T. A. 1999. Structure and composition of unmanaged riparian
forests in the coastal mountains of Oregon, U.S.A. Can. J. For. Res., 29: 15571573.
Paulo, M. J., Tomé, M., Otten, A., & Stein, A. 2005. Comparison of three sampling
methods in the characterization of cork oak stands for managementpurposes.
Can. J. For. Res., 35: 2295-2303.
Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany.
10(2): 290-300.
Richardson, J. S., & Danehy, R. J. 2007. A synthesis of the ecology of headwater
streams and their riparian zones in temperate forests. For. Sci., 53(2): 131-147.
39
SAS Institute Inc. 1990. SAS/STAT User’s Guide, Version 6, Fourth Edition.
Volume 2. SAS Institute Inc., Cary, NC 848 pp.
Schreuder, H. T., Banyard, S. G., & Brink, G. E. 1987. Comparison of three sampling
methods in estimating stand parameters for a tropical forest. For. Ecol.
Manage., 21: 119-127.
Schreuder, H. T., Rennie, J. C., & Williams, M. 1992. Comparison of three sampling
schemes for estimating frequency and D2H by diameter class - a simulation
study. For. Ecol. Manage., 50: 117-131.
Sparks, J. C., Masters, R. E., & Payton, M. E. 2002. Comparative evaluation of
accuracy and efficiency of six forest sampling methods. Proc. Okla. Acad. Sci.,
82, 49-56.
Stamatellos, G. S. 1995. Comparison of point and point-3P sampling for forest height
to diameter ratio estimation with cost analysis. For. Ecol. Manage., 74: 75-79.
Temesgen, H. 2003. Evaluation of sampling alternatives to quantify tree leaf area.
Can. J. For. Res., 33(1): 82-95.
Tokola, T., & Shrestha, S. M. 1999. Comparison of cluster-sampling techniques for
forest inventory in southern Nepal. For. Ecol. Manage. 116: 219-231.
Williams, M. S., Ducey, M. J., & Gove, J. H. 2005. Assessing surface area of coarse
woody debris with line intersect and perpendicular distance sampling. Can. J.
For. Res., 35: 949-960.
40
Chapter 3 - Examination of the suitability of selected sampling methods to quantify
stand attributes of hardwoods and snags within riparian zones.
Abstract
Six sampling alternatives were examined for their ability to quantify selected
attributes of snags and hardwoods in riparian areas of western Oregon. Each
alternative was simulated 500 times at eight locations on a 72 meter square stem map.
The alternatives were evaluated based on how well they estimated trees per hectare
and basal area per hectare using root mean square error and percent bias. In this
simulation, a 3.6 m wide strip oriented perpendicular to the stream outperformd th
other alternatives. However, the variance of all six sampling alternatives was quite
high and further research should be done to determine an optimal sampling method for
hardwoods and snags in a forest dominated by live conifers.
Introduction
This study examines the accuracy and suitability of selected sampling methods
to quantify selected stand attributes in riparian forests of western Oregon. Forest
structure could be described in a variety of ways such as using nearest neighbor
indices (Kint et al. 2003, Aguirre et al. 2003), measures of tree density (Coroi et al.
2004, Lhotka & Loewenstein 2006), canopy characteristics (Nierenberg & Hibbs
2000, Harper & Macdonald 2001, Pabst & Spies 1999, Lhotka & Loewenstein 2006),
height distribution (Chen & Bradshaw 1999), diameter distributions (Chen &
Bradshaw 1999, Mason et al. 2007), snag density (Harper & Macdonald 2001, Pabst &
Spies 1999), basal area per hectare (Pabst & Spies 1999), crown diameter (Chen &
Bradshaw 1999), spatial analysis (Kint et al. 2003, Aguirre et al. 2003, Mason et al.
2007, Chen & Bradshaw 1999), downed wood (Harper & McDonald 2001) and
species composition (Aguirre et al. 2003, Nierenberg & Hibbs 2000, Harper &
Macdonald 2001, Coroi et al. 2004, Pabst & Spies 1999). The best sampling
alternative is the one which most accurately quantifies the chosen attributes of forest
structure in a highly diverse riparian forest.
41
Quantifying stand structure within riparian areas can be extremely difficult.
Riparian areas rank among the most complex, variable and dynamic terrestrial habitats
in the world (Naiman and Decamps 1997; Coroi et al. 2004). This is no different in
the Pacific Northwest (Acker et al. 2003). Stocking, horizontal and vertical structure
are all highly variable within riparian areas of headwater streams (Richardson &
Danehy 2007). For example, hardwood and conifer components vary in density,
diameter and total tree height. Snags within riparian areas add to the variety of
vertical structure. In this study, measures of selected stand attributes were trees per
hectare and basal area per hectare for hardwoods and snags.
As a component of forest structure, both hardwoods and snags are an important
part of forest diversity and habitat (Holden et al. 2006, Hagar 2007). Snags serve as
homes for cavity nesters (Holden et al. 2006), and may fall over and provide cover for
small mammals (Holden et al. 2006), or supply down wood for streams (Harmon, et
al. 1986). Hardwoods are important to a variety of songbird species, bats, and small
mammals (Hagar 2007). In a conifer dominated forest, hardwoods and snags may not
fall in sample plots. These objects can be rare and therefore one could expect that a
sampling alternative that performs well to sample conifers may not perform as well in
sampling hardwoods and snags. It is difficult to discuss a sampling design that is
adequate in capturing forest structure without considering its performance in
estimating attributes of hardwoods and snags. The aim of this study was to find out
whether sampling alternatives that have performed well in sampling conifers in
riparian areas would also perform well when sampling hardwoods and snags.
The riparian areas of interest in this study were primarily headwater streams.
Headwater streams are unique from larger streams (Richardson & Danehy 2007) and
serve as important habitat for amphibians and other non-fish vertebrates (Olson &
Weaver 2007). These streams make up a high percentage of total stream length
(Richardson & Danehy 2007) and drain much of the overall watershed area (Anderson
et al. 2007). For example, stream amphibians have strong associations with physical
habitat features (Olson & Weaver 2007). An accurate sampling alternative is
42
important to describe wildlife habitat and species diversity within these unique stream
systems.
Sampling designs have been compared in a variety of ways. A common metric for
comparison has been efficiency, either by relative efficiency or a measurement of time
per plot (Coble & Grogan 2007, Dahl et al. 2008, Ducey et al. 2002, Johnson & Hixon
1952, Kenning et al., 2005, Lynch, 2003, Paulo et al. 2005, Sparks et al. 2002,
Stamatellos, 1995, Williams et al. 2005). Although time was not considered in this
study, generalizations about the timeliness of many of the alternatives used in this
study can be drawn from the literature. Other common measures were standard error
of the mean (Dahl et al., 2008, Ducey et al. 2002, Lindsey et al. 1958, Paulo et al.
2005, Schreuder et al. 1987), and bias or percent bias (Kenning et al. 2005, Nelson et
al. 1998, Schreuder et al. 1992, Temesgen, 2003, Tokola & Shrestha, 1999). Williams
et al. 2005 used RMSE. This study evaluated the variation and accuracy in stand
structural parameter estimates for each sampling alternative compared to the known
parameter mean for each of eight headwater stream locations.
43
Methods
This study resided on United States Bureau of Land Management (BLM)
Density Management Study (DMS) sites. A primary goal of the DMS is to study
options for silvicultural treatments of young stands to create and maintain latesuccessional forest characteristics. Incepted in 1994, the DMS is a collaborative
project between the BLM; US Forest Service, Pacific Northwest Research Station
(PNW); US Geological Service (USGS); and Oregon State University (OSU). Sites
are located in the Coast Range and the western foothills of the Cascade Range in
Oregon. Stand age ranges among sites from 40 to 70 years. Streams classified as
headwater streams (generally first- or second-order streams) were the focus of this
study. Figure 3.1 shows the location of the DMS sites throughout western Oregon.
Figure 3. 1 Map of DMS site locations (Cissel et al. 2006)
44
The locations sampled in this study were chosen using a stratified random
sampling scheme. A list of all possible headwater reaches within the DMS was
generated from two sets of maps and information on stream size provided by the US
Forest Service. Stream reaches were stratified to sample two DMS treatments, three
DMS buffer treatments, and two slope classes. Twelve reaches from the various
locations were originally selected for sampling. However, due to time constraints,
sampling was completed on just eight reaches. Attributes of the sampled reaches are
summarized in Table 3.1 and 3.2. The two density management treatments were an
unthinned control, 200-350 trees per acres (TPA) and a moderate density retention
where 60-65 % of the stand had been thinned to 80 TPA. The three buffer widths
were a two site potential tree height buffer (2SPTH), a variable width buffer and an
unthinned control. The 2SPTH buffer is measured in slope distance from the stream
center (thalweg) and is based on the 50 year site index height of trees for each site, on
average 146 m. The variable radius buffer had a minimum width of 15.24 m, but
fluctuated based on sensitive areas, such as those prone to landslide or where
threatened species were present. The two slope classes were moderate, less than 30 %
or steep, greater than 30 % in slope. The randomly sampled locations of interest for
this study were Bottomline, Keel Mountain, O.M. Hubbard, and Ten High.
45
Table 3. 1 Description of tree density, buffer width, slope and aspect for the eight
stream reaches sampled from the DMS site locations. Slopes greater than 30%
were classified as an “S” for steep, slopes less than 30% were classified as “M”
for moderate.
Location
Bottom Line
Keel Mountain
Keel Mountain
Keel Mountain
Keel Mountain
O.M. Hubbard
Ten High
Ten High
Reach
BLM
Number District
13
Eugene
17
Salem
18
Salem
19
Salem
21
Salem
36
Roseburg
46
Eugene
75
Eugene
Density
Moderate
Control
Moderate
Moderate
Moderate
Moderate
Control
Moderate
Slope
Buffer
% Slope Class
Two Tree Height
S
51
Control
M
18
Two Tree Height
M
21.2
Two Tree Height
M
14
Variable Width
S
38
Variable Width
S
31
Control
M
19
Variable Width
S
33
Aspect
NE/SW
N/S
NW/SE
NW/SE
N/S
NW/SE
N/S
N/S
Table 3. 2 Description of stream and tree characteristics of sampled reaches from the
DMS site locations. “PSME” is the abbreviation for Pseudotsuga menziesii (Mirb.) and
TSHE is the abbreviation for Tsuga heterophylla (Raf.).
Mean % Dead
Avg. Stream Dominant
%
DBH Trees Hardwood
Reach Flow Width Conifer
Number
Location
(m)
Species TPH (cm) (TPH)
(TPH)
Bottom Line
13
0.2
PSME 405
34
20%
7%
Keel Mountain
17
1.1
TSHE 461
42
7%
8%
Keel Mountain
18
2.2
TSHE 740
30
13%
5%
Keel Mountain
19
0.8
PSME 658
28
17%
2%
Keel Mountain
21
3.0
TSHE 434
36
13%
10%
O.M. Hubbard
36
0.5
PSME 467
31
13%
8%
Ten High
46
0.0
PSME 461
41
10%
0%
Ten High
75
4.3
PSME 716
31
8%
26%
Once the reaches were selected, a sampling plot was randomly located along
each of the headwater streams. The plot ran 72 m parallel to the stream and 36 meters
upslope, each side of the centerline of the plot. Using GIS stream files, the length of
the stream was measured and a random number to the nearest meter was generated for
the placement of the centerline at the lower end of the plot. The random number was
between 0 and 72 meters less the entire length of the stream. For instance, if the
46
stream were 683 meters in length, a random number would be chosen between 0 and
611 meters. A laser was used to locate the random starting point for each of the
blocks along the stream. The 72 m square block was oriented with its centerline
running along the general azimuth of the stream. An example of block layout can be
seen in Figure 3.2.
Figure 3. 2 Illustration of plot layout on the sampled stream. One can see the
centerline of the 72 m square block was oriented in the general direction of the
stream. Sampling of trees took place 36 m from the centerline to upslope.
Stem mapping was done using Total Station Survey Equipment. Coordinates
for a control point (station) were found using a Trimble survey-grade GPS.
Coordinates were found at two points and the angle between the two points was hand
calculated for greater precision (estimates of precision for coordinates). This
calculation was then checked with a hand compass. For reaches where only one GPS
point was recorded, a staff compass was used to calculate the correct bearing. Stems
were numbered sequentially as they were mapped. Biodegradable flagging was used
to mark each tree that was surveyed with the number recorded in the total station.
47
Each tree number and species was recorded within the survey software program. A
reflector was held against the outside of the tree or within one foot of the center of the
tree. The distance to each tree (estimate of precision for distance) was calculated by
internal software based on the bearing from the station to the tree, and the latitude and
longitude of the station. When all trees that could be measured from a control point
were recorded, the station was moved to a new control point. The tripod and total
station were then setup at the new point and numbering of trees continued from the
last tree measured at the previous control point.
Tree measurements including diameter at breast height, species, condition,
canopy classification, crown class, and decay class were recorded during stem
mapping. Data was recorded on paper and then entered into an excel worksheet.
Coordinate data for the trees was recorded in a handheld from the total station. Table
3.3 shows the parameters being measured and recording protocol.
Table 3. 3 List of tree parameters measured on each stem mapped tree and the
recording protocol for each measurement.
Tree Attribute
Recording Protocol
Diameter at breast height
Nearest 1/10th centimeter
Species
4 Letter code, appendix
Canopy Classification
Visually determined, 1 letter code
Condition
Visually determined, Dead (D), Live (L)
Crown Class
2 Letter code, visually determined, Table N
Decay Class
2 Letter code, visually determined, Table N
Diameter at breast height (DBH), was measured to the nearest tenth of a
centimeter at 1.37 m (4.5 ft) above the ground as measured on the uphill side of the
tree. Measurements were made using a diameter tape. Trees forking below diameter
at breast height were counted as two trees. Diameter was measured above any
abnormalities such as bulging. Pistol butt or leaning trees were measured at 1.37 m
along the stem. Diameter was taken on all trees over 7.5 cm in diameter. Tree species
48
was recorded using a two letter common name code. Trees were defined as woody
stems taller than 20 feet or larger than 7.5 centimeters in diameter. The condition of
the tree was determined visually by the surveyor. A dead tree was one that had no
green foliage: a live tree had at least some remaining green foliage.
SAS (v. 9.1, SAS Institute 1990) was used to simulate six sampling alternatives.
Illustrations of each of the alternatives can be found in Appendix II. In order to avoid
incomplete plots landing along edges of the stem map, the data was wrapped edge to
end (Appendix I). The simulation was conducted on each of the eight locations. The
live conifers were removed from the dataset and only the dead trees and hardwoods
were used in simulating the six alternatives. Each design was simulated at a 10 %
(0.1*722 = 518.4 m2) and 20 % (0.2 *722 = 1036.8 m2) intensity of the area.
The six sampling alternatives can be grouped into two sampling designs:
systematic random sampling (SYRS) and stratified random sampling (STRS). Every
alternative was simulated 500 times with a different random starting point for each
replication. The plot center coordinates were randomly selected and the number of
plots for each of the plot radius sizes can be seen in Table 3.3.
SYRS was used for four of the strip sampling alternatives. The PEST3 and PEST7
alternative were 3.6 m and 7.2 m wide strips which were oriented perpendicular to the
centerline of the plot (Figure 3.2). The alternating strip option, ASTP3 and ASTP7 in
Table 2.1 is unique. This design was simulated by breaking the 72 meter grid into
either 3.6 m or 7.2 m strips which were 36 meters in length. All strips were oriented
perpendicular to the centerline of the plot (Figure 3.2), approximately perpendicular to
the stream. Each plot consisted of a rectangular plot on one side of the stream and the
rectangular strip diagonal to it on the opposite side of the stream. For example, one
could think of this as running a transect perpendicular to the stream and sampling a 3.6
m strip to the left of the transect line on one side of the stream, and all trees within 3.6
m to the right of the transect line on the opposite side of the stream.
STRS was used to sample strips running parallel to the stream, but also with at
least one strip close to the stream, and another upslope. Strips were either 9 m or 3.6
49
m wide. There were two strata, one strata was close to the stream and the other
upslope. The PAST3 alternative had a strip width of 3.6 m and the PAST9 alternative
had a strip width of 9 m. In the case of the 9 m strip width, the 72 meter grid was first
split into 9 m strips running parallel to the stream, strata 1 was the two outermost
strips on either side of the stream and strata 2 was the four most inner strips.
Two methods were used to evaluate each of the designs compared to the actual
stem map. The designs were evaluated based on how well they predicted trees per
hectare (TPH) and basal area per hectare (BAPH) using root mean square error
(RMSE), percent bias (PB).
The RMSE was computed as:

^


Y
Y
−
k
∑


k =1 


500
2
200
RMSE =
[1]
^
Where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
The equation used to estimate percent bias is as follows:
^
Y k −Y
∑
Y
PB = k =1
500
200
^
where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
[2]
50
Table 3. 4 Description of the sampling alternatives examined in this study including
the number of plots simulated at the 10 and 20% sampling intensity.
Sampling
Alternative
ASTP3
ASTP7
PEST3
PEST7
PAST3
PAST9
Description
Alternate
Perpendicular
Strips
Alternate
Perpendicular
Strips
Perpendicular
Strips
Perpendicular
Strips
STRS Parallel
Strip Sampling
STRS Parallel
Strip Sampling
Plots
Plots
(10%) (20%)
Size
Shape
3.6 m by 36
m
Rectangle
4
8
7.2 m by 36
m
Rectangle
2
4
Rectangle
2
4
Rectangle
1
2
3.6m x 36 m
Rectangle
4
8
9 m x 28.8
m
Rectangle
2
4
3.6 m x 72
m
7.2 m x 72
m
51
Results
From Table 3.4, one can see the resulting RMSE values for the six alternatives.
The RMSE values for TPH ranged from approximately 44 to 55 for the 10 % intensity
and approximately 29 to 39 for the 20 % sampling intensity. For the 10 % sampling
intensity, the values ranged from 4.3 to 4.7 and 3.0 to 3.5 for the 20 % sampling
intensity. ASTP7, PAST3 and PEST3 all performed well. The “Consistency”
columns indicate the number of sites where the sampling alternative among the
smallest two RMSE values. In this case, PEST3 was among the smallest RMSE
values at five of the eight sites when predicting TPH at the 20 % sampling intensity.
Table 3.5 shows the PB values for each of the alternatives at the 10 and 20%
sampling intensities. PB values ranged from 1.26% to 5.33% and from -0.93% to
0.78% for the 10% and 20% sampling intensities when evaluated for their ability to
estimate TPH. The PB was slightly larger for BAPH, values ranged from 0.74% to
6.49% and from -0.73% to 0.78% for the 10 and 20% sampling intensities
respectively. The alternatives that performed well were PEST3, ASTP7, PEST3 and
PAST3. However, the alternative that performed well at both intensities as well as at
at least four of the eight sites was PEST3.
In Table 3.7 and 3.8 one can see the standard deviation calculated for the mean
PB and RMSE values across the eight sites. The two tables show there was high
variation between sites in the performance of the alternatives. The standard deviation
was quite large for all sampling alternatives, however PEST3 had among the smallest
variances at both intensities when evaluated using PB and RMSE. Other alternatives
that performed well were ASTP3, PAST3 and PEST7. From Table 3.7, one can see
that in general when estimating TPH at the 10% sampling intensity, there was less
variation in the smaller strip widths. When evaluated using RMSE other intensities
for TPH and BAPH, there was no obvious trend in strip width and variation. From
Table 3.8, one can see for the 10% intensity, when estimating TPH, ASTP3, PAST3
and PEST3 outperformed ASTP7, PAST9 and PEST7 respectively. There were no
apparent trends for the ASTP3 and ASTP7 alternatives when estimating BAPH at
52
either intensity. However for BAPH at both intensities, PAST3 and PEST3 performed
better than PAST9 and PEST7. This trend was similar for the 20% intensity when
estimating TPH at the 20% intensity.
53
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
BAPH
4.47
4.34
4.38
4.69
4.29
4.41
3.30
3.24
3.06
3.54
3.19
3.20
Consistency
BAPH
Intensity
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
TPH
45.6
46.8
43.9
55.2
43.3
51.3
32.2
32.6
32.3
38.7
29.9
29.0
Rank Rank
TPH BAPH
3
5
4
2
2
3
6
6
1
1
5
4
3
5
5
4
4
1
6
6
2
2
1
3
Consistency
TPH
Sampling
Alternative
Table 3. 5 Summary of performance of sampling alternatives evaluated using
RMSE. Values under trees per hectare (TPH) and basal area per hectare
(BAPH m2) are the RMSE values from the eight locations. Shaded values have
the two smallest RMSE values. The values under the “Consistency” columns
are the number of times the alternative was among the two most accurate
alternatives for each location. Alternatives were shaded if they performed well
at four or more of the eight locations.
5
3
1
1
3
3
2
2
2
1
5
4
3
3
3
0
3
4
3
2
5
2
2
2
54
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
BAPH
3.48%
3.74%
1.77%
4.18%
0.74%
6.49%
0.25%
0.78%
-0.34%
-0.73%
-0.54%
-0.33%
Rank Rank
TPH BAPH
3
3
4
4
2
2
5
5
1
1
6
6
4
4
2
2
5
5
3
3
6
6
1
1
Consistency
BAPH
Intensity
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
TPH
2.95%
4.00%
2.71%
4.55%
1.26%
5.33%
-0.40%
0.24%
0.78%
-0.27%
-0.93%
0.21%
Consistency
TPH
Sampling
Alternative
Table 3. 6 Summary of performance of sampling alternatives evaluated using
percent bias (PB). Values under trees per hectare (TPH) and basal area per
hectare (BAPH m2) are the PB values from the eight locations. Shaded values
have the two smallest PB values. The values under the “Consistency” columns
are the number of times the alternative was among the two most accurate
alternatives for each location. Alternatives were shaded if they performed well
at four or more of the eight locations.
3
3
4
1
4
1
5
2
1
2
3
3
1
4
2
3
5
1
4
3
2
3
3
1
55
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
Intensity
Sampling
Alternative
Table 3. 7 Summary of the standard deviation for sampling alternatives evaluated
using root mean square error (RMSE). Values under standard deviation (SD) trees per
hectare (TPH) and SD basal area per hectare (BAPH m2) are the standard deviation for
the mean RMSE values from the eight locations. Shaded values have the two smallest
standard deviation values.
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
SD
SD
TPH BAPH
31.56
4.39
34.97
3.92
36.86
3.75
42.35
4.29
31.59
3.83
33.43
4.09
23.20
2.30
23.17
2.23
23.44
2.16
24.91
2.27
18.81
2.22
20.09
2.38
Rank Rank
TPH BAPH
1
6
4
3
5
1
6
5
2
2
3
4
4
5
3
3
5
1
6
4
1
2
2
6
56
Sampling
Alternative
Intensity
Table 3. 8 Summary of the standard deviation for sampling alternatives evaluated
using percent bias (PB). Values under standard deviation (SD) trees per hectare
(TPH) and SD basal area per hectare (BAPH m2) are the standard deviation for the
mean PB values from the eight locations. Shaded values have the two smallest
standard deviation values.
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
ASTP3
ASTP7
PAST3
PAST9
PEST3
PEST7
10%
10%
10%
10%
10%
10%
20%
20%
20%
20%
20%
20%
SD
TPH
48.05%
51.13%
54.85%
63.38%
47.58%
53.31%
34.87%
35.78%
36.01%
42.08%
31.63%
31.31%
SD
BAPH
89.53%
85.30%
81.93%
93.86%
83.86%
88.92%
57.44%
57.02%
53.24%
62.22%
55.74%
58.30%
Rank Rank
TPH BAPH
2
5
3
3
5
1
6
6
1
2
4
4
3
4
4
3
5
1
6
6
2
2
1
5
57
Discussion
ASTP3 and ASTP7 typically performed better than the PAST9 alternative, but
not as well as PEST3 and PEST7. However, there was little difference in the
performance of ASTP3 and ASTP7. In addition, there didn’t seem to be a trend for a
single alternative to be better at estimating TPH compared to BAPH. Although
ASTP3 and ASTP7 were very similar to PEST3 and PEST7, they performed slightly
better than PEST7 and slightly worse than PEST3 and PAST3. One would think that
if the trees were located completely random along the stream that the ASTP3
alternative would have performed as well as PEST3.
PAST3 outperformed PAST9 and was second to the performance of the PEST3
alternative. When evaluated using RMSE, PAST3 outperformed PAST9 when
estimating TPH and BAPH at both intensities. However, when evaluated using PB,
PAST9 performed better at the 20% intensity when estimating TPH. The standard
deviation for the mean RMSE and PB was smaller for PAST3 than PAST9 at both
intensities. PAST9 was almost always ranked at the bottom of the six alternatives in
terms of variation. Possible reasons for the performance of this alternative could be
the arrangement of snags close to the stream compared to upslope. Had all snags been
located very close to the stream, this alternative may not have performed as well.
PEST3 overall performed the best in estimating TPH and BAPH of snags and
hardwoods in the stem mapped stands. The alternative performed the worst at the
20% sampling intensity when evaluated using PB. However, in general the sampling
alternative ranked either first or second in estimating TPH and BAPH at the two
intensities. PEST7 performed best at the 20% intensity. The inconsistency in the
performance of these alternatives could be attributed to the high variation reflected in
Tables 3.7 and 3.8. PEST3 was most consistent when evaluated using PB at the 10%
intensity. The performance of the smaller strip widths may be attributed large spaces
between snags and the smaller number of wide plots may have had more empty plots
compared to the plots with smaller strip width.
58
The variation for the six alternatives simulated within this study was quite
large and if confidence intervals were calculated for the mean TPH or BAPH, they
would be very wide. Reasons for the high variation may be attributed to the small
number of snags and hardwoods within the stem mapped plots. The variation of the
RMSE for TPH was smaller for PAST3 than PAST9 and smaller for PEST3 than
PEST7 at both intensities. However, ASTP7 had a smaller standard deviation than the
ASTP3 alternative. When evaluated using PB, PAST3 outperformed PAST9. There
were no clear trends in strip width for either intensity for the ASTP3, ASTP7, PEST3,
and PEST7 alternatives. This suggests that the alternatives simulated may not be
optimal for sampling trees that make up a small fraction of the population.
Rectangular strip plots have been used in the past to sample snags for the
purpose of estimating snag density (Dahlström, Jönsson & Nilsson 2005, Holden et al.
2006, Wisdom & Bate 2008). However, the results from this study indicate that in
forests where snags or hardwoods are rare, the use of strips may result in high
variation among samples. Kenning et al. (2005) compared n-tree sampling to fixed
area and modified horizontal line sampling. Fixed area sampling with small plots was
recommended for cases where sample size could be fairly large, when only one crew
member was available and basal area per hectare was of primary importance. Ducey
et al. (2002) compared horizontal point sampling and modified horizontal line
sampling (MHLS) for snag inventory in the northeast United States and found
relatively small standard error when using MHLS to estimate snags and basal area per
hectare. However, this study was implemented on stands where snags were known to
be somewhat common. Although the use of transects for strip sampling is quite
prevalent, based on this simulation, in conifer dominated stands, one can expect high
variation with this sampling method and other sampling methods may be considered.
Several recommendations can be made based on this study. Due to the high
variation in the sampling alternatives and similarities in mean TPH and BAPH values,
it would be good to see if a simulation that was run 1000 times differed from on that
was run 500 times. In addition, one could simulate narrower strip widths to see if
59
variation could be further decreased. Further simulation should be conducted which
includes variable radius plots and modified horizontal line sampling. Other sampling
methods which specialize in sampling of rare objects should also be considered.
Although PEST3 performed the best of the six simulated, the high variation leads one
to believe that it is not an optimal design when sampling snags and hardwoods in
riparian areas.
60
Conclusion
PEST3 is recommended, among the simulated alternative but due to the high
variation, a larger sampling intensity is recommended. If a larger sampling intensity is
not desired, one could consider another design which specializes in capturing a small
component of the overall forest composition. Other sampling designs such as MHLS
has performed well in other studies to capture snags and had a relatively small
variation around the mean (Ducey et al. 2002, Kenning et al. 2005).
Although there
appeared to be a trend between smaller strip width and smaller variance overall,
further simulation is needed to verify these trends. Sampling using a 3.6 m strip
perpendicular to the stream should be applied with knowledge of high variance when
sampling a minor component of the overall forest composition within a stand.
There were several reasons for the high variation that occurred within this
study. First, there were eight sites on which these alternatives were simulated. From
Tables 3.4 and 3.5 one can see that none of the alternatives performed well at every
site. Live conifers made up 66 to 90% within the study site, making the number of
hardwoods and snags within each stem mapped plot quite small. In these cases, if
sampling of snags and hardwoods are a primary objective, it may be wise to consider
sampling alternatives that are more successful at estimating rare objects. It’s possible
that with the high variation within riparian areas, a higher sampling intensity is needed
for sampling snags and hardwoods. Further research is needed to find an optimal
design for stands with the attributes described in this study if snags and hardwoods are
the primary objective.
61
References
Acker, S. A., Gregory, S. V., Lienkaemper, G., McKee, W. A., Swanson, F. J., &
Miller, S. D. 2003. Composition, complexity, and tree mortality in riparian
forests in the central Western Cascades of Oregon. For. Ecol. Manage., 173:
293-308.
Aguirre, O., Hui, G., von Gadow, K., & Jimenez, J. 2003. An analysis of spatial forest
structure using neighbourhood-based variables. For. Ecol. Manage., 183: 137145.
Anderson, P. D., Larson, D. J., & Chan, S. S. 2007. Riparian buffer and density
management influences on microclimate of young headwater forests of western
Oregon. For. Sci., 53(2): 254-269.
Anderson, W. E. 1987. Riparian Area Definition: A viewpoint. Rangelands, 9(2), 70.
Chen, J., & Bradshaw, G. A. 1999. Forest structure in space: a case study of an old
growth spruce-fir forest in Changbaishan Natural Reserve, PR China. For.
Ecol. Manage., 120, 219-233.
Cissel, J., Anderson, P., Berryman, S., Chan, S., Puettman, K., and Thompson, C.
2006. BLM Density Management and Riparian Buffer Study: Establishment
Report and Study Plan. U.S. Dept. Interior, U.S. Geological Survey. 161 p.
Coble, D. W., & Grogan, J. 2007. Comparison of systematic line-point and double
sampling designs for pine and hardwood forests in the western gulf. South. J.
Appl. For., 31(4): 199-206.
Coroi, M., Skeffington, M. S., Giller, P., Smith, C., Gormally, M., & O'Donovan, G.
2004. Vegetation diversity and stand structure in streamside forests in the
south of Ireland. For. Ecol. Manage., 202: 39-57.
Dahl, C. A., Harding, B. A., & Wiant, H. V. 2008. Comparing double-sampling
efficiency using various estimators with fixed-area and point sampling. North.
J. Appl. For., 25(2): 99-102.
Ducey, M. J., Jordan, G. J., Gove, J. H., & Valentine, H. T. 2002. A practical
modification of horizontal line sampling for snag and cavity tree inventory.
Can. J. For. Res., 32: 1217-1224.
Garman, S. L., Acker, S. A., Ohmann, J. L., and Spies, T. A. 1995. Asymptotic
height-diameter equations for twenty-four tree species in western Oregon.
Forest Research Laboratory, Oregon State University, Corvallis. Research
Contribution 10. 22p.
Hagar, J. C. 2007. Wildlife species associated with non-coniferous vegetation in
Pacific Northwest conifer forests: A review. For. Ecol. Manage., 246: 108-122.
Harmon, M. H., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J.
D., Anderson, N. H., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G.
W., Cromack Jr., K., & Cummins, K. W. 1986. Ecology of coarse woody
debris in temperate ecosystems. Adv. Ecol. Res., 15: 133-302.
Harper, K. A., & Macdonald, S. E. 2001. Structure and composition of riparian boreal
forest new methods for analyzing edge influence. Ecology. 82(3): 649-659.
62
Holden, Z. A., Morgan, P., Rollins, M. G., & Wright, R. G. 2006. Ponderosa pine
snag densities following multiple fires in the Gila Wilderness, New Mexico.
For. Ecol. Manage., 221, 140-146.
Husch, B., Beers, T. W., & Kershaw, J. A. Jr. 2003. Forest Mensuration. 4th ed. John
Wiley & Sons, Inc., New Jersey. 443 p.
Iles, K., & Smith, N. J. 2006. A new type of sample plot that is particularly useful for
sampling small clusters of objects. For. Sci., 52(2): 148-154.
Johnson, F. A., & Hixon, H. J. 1952. The most efficient size and shape of plot to use
for cruising in old-growth Douglas-fir timber. J. Forestry, 50(1): 17-20.
Kenning, R. S., Ducey, M. J., Brissette, J. C., & Gove, J. H. 2005. Field efficiency and
bias of snag inventory methods. Can. J. For. Res., 35: 2900-2910.
Kint, V., Meirvenne, M. V., Nachtergale, L., Geudens, G., & Lust, N. 2003. Spatial
methods for quantifying forest stand structure development: A comparison
between nearest-neighbor indices and variogram analysis. For. Sci. 49(1): 3649.
Lhotka, J. M., & Loewenstein, E. F. 2006. Indirect measures for characterizing light
along a gradient of mixed-hardwood riparian forest canopy structures. For.
Ecol. Manage., 226, 310-318.
Lyon, J., & Gross, N. M. 2005. Patterns of plant diversity and plant-environmental
relationships across three riparian corridors. For. Ecol. Manage., 204: 267-278.
Lynch, A. M. 2003. Comparison of fixed-area plot designs for estimating stand
characteristics and western spruce budworm damage in southwestern U.S.A.
forests. Can. J. For. Res., 33: 1245-1255.
Lynch, T. B. 2006. Horizontal line sampling for riparian forests without land area
estimation. For. Sci. 52(2): 119-129.
Lindsey, A. A., Barton, J. D., & Miles, S. R. 1958. Field efficiencies of forest
sampling methods. Ecology, 39(3): 428-444.
Maclean, D. A., & Ostaff, D. P. 1983. Sample size - precision relationships for use in
estimating stand characteristics and spruce budworm caused tree mortality.
Can. J. For. Re., 13: 548-555.
Mason, W. L., Connolly, T., Pommerening, A., & Edwards, C. 2007. Spatial structure
of semi-natural and plantation stands of Scots pine (Pinus sylvestris L.) in
northern Scotland. Forestry, 80(5): 567-586.
Naiman, R. J., Decamps, H., & Pollock, M. 1993. The role of riparian corridors in
maintaining regional biodiversity. Ecol. Appl., 3(2): 209-212.
Nierenberg, T. R., & Hibbs, D. E. 2000. A characterization of unmanaged riparian
areas in the central Coast Range of western Oregon. For. Ecol. Manage., 129:
195-206.
Nelson, R. F., Gregoire, T. G., & Oderwald, R. G. 1998. The effects of fixed-area plot
width on forest canopy height simulation. For. Sci., 3(44): 438-444.
Olson, D. H., & Weaver, G. 2007. Vertebrate assemblages associated with headwater
hydrology in western Oregon managed forests. For. Sci., 52(2): 343-355.
63
Pabst, R. J., & Spies, T. A. 1999. Structure and composition of unmanaged riparian
forests in the coastal mountains of Oregon, U.S.A. Can. J. For. Res., 29: 15571573.
Paulo, M. J., Tomé, M., Otten, A., & Stein, A. 2005. Comparison of three sampling
methods in the characterization of cork oak stands for managementpurposes.
Can. J. For. Res., 35: 2295-2303.
Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany.
10(2): 290-300.
Richardson, J. S., & Danehy, R. J. 2007. A synthesis of the ecology of headwater
streams and their riparian zones in temperate forests. For. Sci., 53(2): 131-147.
SAS Institute Inc. 1990. SAS/STAT User’s Guide, Version 6, Fourth Edition.
Volume 2. SAS Institute Inc., Cary, NC 848 pp.
Schreuder, H. T., Banyard, S. G., & Brink, G. E. 1987. Comparison of three sampling
methods in estimating stand parameters for a tropical forest. For. Ecol.
Manage., 21: 119-127.
Schreuder, H. T., Rennie, J. C., & Williams, M. 1992. Comparison of three sampling
schemes for estimating frequency and D2H by diameter class - a simulation
study. For. Ecol. Manage., 50: 117-131.
Smith, C. Y., Moroni, M. T., & Warkentin, I. G. 2009. Snag dynamics in post-harvest
landscape of western Newfoundland balsam fir-dominated boreal forests. For.
Ecol. Manage., 258: 832-839.
Sparks, J. C., Masters, R. E., & Payton, M. E. 2002. Comparative evaluation of
accuracy and efficiency of six forest sampling methods. Proc. Okla. Acad. Sci.,
82, 49-56.
Stamatellos, G. S. 1995. Comparison of point and point-3P sampling for forest height
to diameter ratio estimation with cost analysis. For. Ecol. Manage., 74: 75-79.
Temesgen, H. 2003. Evaluation of sampling alternatives to quantify tree leaf area.
Can. J. For. Res., 33(1): 82-95.
Tokola, T., & Shrestha, S. M. 1999. Comparison of cluster-sampling techniques for
forest inventory in southern Nepal. For. Ecol. Manage. 116: 219-231.
Williams, M. S., Ducey, M. J., & Gove, J. H. 2005. Assessing surface area of coarse
woody debris with line intersect and perpendicular distance sampling. Can. J.
For. Res., 35: 949-960.
Wisdom, M. J., & Bate, L. J. 2008. Snag density varies with intensity of timber
harvest and human access. For. Ecol. Manage., 255: 2085-2093.
64
CHAPTER 4 – GENERAL CONCLUSION
This analysis evaluated sixteen sampling alternatives for their ability to
estimate selected stand attributes of live conifers and six sampling alternatives for
their ability to estimate selected stand attributes of hardwoods and snags. Snags
included all dead hardwoods and conifer at each location. Monte Carlo simulation
was used to evaluate the performance of these alternatives for their ability to quantify
the selected stand attributes. All sampling alternatives were replicated 500 times at
each of the eight stem mapped locations. The selected stand attributes were trees per
hectare (TPH), basal area per hectare (BAPH), and height to diameter ratio (H/D) for
live conifers and TPH and BAPH for hardwoods and snags. Sampling alternatives
that performed well were ASTP3, ASTP7, PEST3, and PEST7. The ASTP3
alternative performed best when estimating selected stand attributes of live conifers,
however PEST3 performed best when estimating selected stand attributes of
hardwoods and snags. Overall, PEST3 is recommended as the alternative that could
be used to sample live conifers, hardwoods and snags.
The scope of inference for this study falls within headwater streams with 40-60
year old Douglas-fir forests of western Oregon with a buffer of approximately 220 feet
(one site potential tree height). Analysis and inference can be applied to forests of
similar species composition and stand density. The history of the study sites includes
no management prior to 1994 when thinning treatments began. Multiple aspects for
each side of the stream were included within the study. Structural heterogeneity
among these sites may naturally vary and caution should be used when applying these
density treatments to larger areas of different forest structure. This study may not be
representative for areas with narrower streamside buffers.
The first part of this study examined sampling alternatives for live conifers
within headwater stream riparian areas. Several trends emerged during this study, but
more replication is needed to see if these hold true for all of western Oregon forest.
Alternatives which sampled perpendicular to the stream outperformed those that
sampled parallel to the stream. Variation in stand structure appeared to be strongest
65
when moving from the stream to upslope, but variation running parallel to the stream
also impacted the performance of the sampling alternatives. Alternatives that sampled
both sides of the stream did better than those that only sampled one side of the stream.
Alternatives based on rectangular shaped plots performed better in this study than
circular or wedge shaped plots. Strips that were 3.6m in width performed better than
those that were 7.2 m or 9 m in width. Where SYRS and SRS were used, SYRS
performed better than SRS. The most dominant gradient in riparian areas appears to
occur from the stream to upslope and therefore the alternatives sampling perpendicular
to this gradient were the most successful in estimating TPH, BAPH and H/D.
The second part of this study examined six sampling alternatives for
hardwoods and snags within headwater stream riparian areas. PEST3 is recommended
for sampling snags and hardwoods; however there was high variation around the mean
values for all alternatives simulated. There did not appear to be any obvious gradients
when estimating snags and hardwoods as both alternatives sampling parallel to the
stream and perpendicular to the stream (PEST3 and PAST3) performed quite well.
Live conifers made up 66 to 90% within the study site, making the number of
hardwoods and snags within each stem mapped plot quite small. In these cases, if
sampling of snags and hardwoods are a primary objective, it may be wise to consider
sampling alternatives that are more successful at estimating rare objects. It’s possible
that with the high variation within riparian areas, a higher sampling intensity is needed
for sampling snags and hardwoods.
This study confirms that there is high variation within headwater riparian
areas, even those that are located relatively close to one another and have similar
species composition. This was particularly prevalent in the simulation of sampling
alternatives for snags and hardwoods. None of the sampling alternatives that were
examined performed with the highest accuracy at every location. This was especially
true when estimating selected stand attributes of snags and hardwoods. Further
research should be conducted to find an optimal design for estimating attributes of
snags and hardwoods in conifer dominated riparian forest.
66
Bibliography
Acker, S. A., Gregory, S. V., Lienkaemper, G., McKee, W. A., Swanson, F. J., &
Miller, S. D. 2003. Composition, complexity, and tree mortality in riparian
forests in the central Western Cascades of Oregon. For. Ecol. Manage., 173:
293-308.
Aguirre, O., Hui, G., von Gadow, K., & Jimenez, J. 2003. An analysis of spatial forest
structure using neighbourhood-based variables. For. Ecol. Manage., 183: 137145.
Anderson, P. D., Larson, D. J., & Chan, S. S. 2007. Riparian buffer and density
management influences on microclimate of young headwater forests of western
Oregon. For. Sci., 53(2): 254-269.
Barker, J. R., Ringold, P. L., & and Bollman, M. 2002. Patterns of tree dominance in
coniferous riparian forests. For. Ecol. Manage., 166, 311-329.
Bormann, F. H. 1953. The statistical efficiency of sample plot size and shape in forest
ecology. Ecology, 34(3), 474-487.
Chen, J., & Bradshaw, G. A. 1999. Forest structure in space: a case study of an old
growth spruce-fir forest in Changbaishan Natural Reserve, PR China. For.
Ecol. Manage., 120, 219-233.
Cissel, J., Anderson, P., Berryman, S., Chan, S., Puettman, K., and Thompson, C.
2006. BLM Density Management and Riparian Buffer Study: Establishment
Report and Study Plan. U.S. Dept. Interior, U.S. Geological Survey. 161 p.
Coble, D. W., & Grogan, J. 2007. Comparison of systematic line-point and double
sampling designs for pine and hardwood forests in the western gulf. South. J.
Appl. For., 31(4): 199-206.
Coroi, M., Skeffington, M. S., Giller, P., Smith, C., Gormally, M., & O'Donovan, G.
2004. Vegetation diversity and stand structure in streamside forests in the
south of Ireland. For. Ecol. Manage., 202: 39-57.
Dahl, C. A., Harding, B. A., & Wiant, H. V. 2008. Comparing double-sampling
efficiency using various estimators with fixed-area and point sampling. North.
J. Appl. For., 25(2): 99-102.
Dale, V. H., Franklin, W. M., & Gardner, R. H. 1991. Sampling ecological
information: choice of sample size. Ecological Modeling, 57, 1-10.
Dignan, P., Bren, L., 2003. A study of the effect of logging on the understory light
environment in riparian buffer strips in a south-east Australian forest. For.
Ecol. Manage. 172: 161-172.
Ducey, M. J., Jordan, G. J., Gove, J. H., & Valentine, H. T. 2002. A practical
modification of horizontal line sampling for snag and cavity tree inventory.
Can. J. For. Res., 32: 1217-1224.
Freese, F. 1961. Relation of plot size to variability: An approximation. J. Forestry.,
59(9): 679.
Garman, S. L., Acker, S. A., Ohmann, J. L., and Spies, T. A. 1995. Asymptotic
height-diameter equations for twenty-four tree species in western Oregon.
67
Forest Research Laboratory, Oregon State University, Corvallis. Research
Contribution 10. 22p.
Gomi, T., Sidle, R. C., & Richardson, J. S. 2002. Understanding processes and
downstream linkages of headwater systems. BioScience, 52(10): 905-916.
Hagar, J. C. 2007. Wildlife species associated with non-coniferous vegetation in
Pacific Northwest conifer forests: A review. For. Ecol. Manage., 246: 108-122.
Harmon, M. H., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J.
D., Anderson, N. H., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G.
W., Cromack Jr., K., & Cummins, K. W. 1986. Ecology of coarse woody
debris in temperate ecosystems. Adv. Ecol. Res., 15: 133-302.
Harper, K. A., & Macdonald, S. E. 2001. Structure and composition of riparian boreal
forest new methods for analyzing edge influence. Ecology. 82(3): 649-659.
Harrington, Constance A., Scchoenholtz, Stephen H., eds. 2005. Productivity of
Western forests: a forest products focus. Gen. Tech. Rep. PNW-GTR-642.
Portland, OR: U.S. Dept. of Agriculture, Forest Service, Pacific Northwest
Research Station. 176 p.
Holden, Z. A., Morgan, P., Rollins, M. G., & Wright, R. G. 2006. Ponderosa pine
snag densities following multiple fires in the Gila Wilderness, New Mexico.
For. Ecol. Manage., 221, 140-146.
Husch, B., Beers, T. W., & Kershaw, J. A. Jr. 2003. Forest Mensuration. 4th ed. John
Wiley & Sons, Inc., New Jersey. 443 p.
Iles, K., & Smith, N. J. 2006. A new type of sample plot that is particularly useful for
sampling small clusters of objects. For. Sci., 52(2): 148-154.
Johnson, F. A., & Hixon, H. J. 1952. The most efficient size and shape of plot to use
for cruising in old-growth Douglas-fir timber. J. Forestry, 50(1): 17-20.
Kenning, R. S., Ducey, M. J., Brissette, J. C., & Gove, J. H. 2005. Field efficiency and
bias of snag inventory methods. Can. J. For. Res., 35: 2900-2910.
Kint, V., Meirvenne, M. V., Nachtergale, L., Geudens, G., & Lust, N. 2003. Spatial
methods for quantifying forest stand structure development: A comparison
between nearest-neighbor indices and variogram analysis. For. Sci. 49(1): 3649.
Lhotka, J. M., & Loewenstein, E. F. 2006. Indirect measures for characterizing light
along a gradient of mixed-hardwood riparian forest canopy structures. For.
Ecol. Manage., 226, 310-318.
Lyon, J., & Gross, N. M. 2005. Patterns of plant diversity and plant-environmental
relationships across three riparian corridors. For. Ecol. Manage., 204: 267-278.
Lynch, A. M. 2003. Comparison of fixed-area plot designs for estimating stand
characteristics and western spruce budworm damage in southwestern U.S.A.
forests. Can. J. For. Res., 33: 1245-1255.
Lynch, T. B. 2006. Horizontal line sampling for riparian forests without land area
estimation. For. Sci. 52(2): 119-129.
Lindsey, A. A., Barton, J. D., & Miles, S. R. 1958. Field efficiencies of forest
sampling methods. Ecology, 39(3): 428-444.
68
Maclean, D. A., & Ostaff, D. P. 1983. Sample size - precision relationships for use in
estimating stand characteristics and spruce budworm caused tree mortality.
Can. J. For. Re., 13: 548-555.
Mason, W. L., Connolly, T., Pommerening, A., & Edwards, C. 2007. Spatial structure
of semi-natural and plantation stands of Scots pine (Pinus sylvestris L.) in
northern Scotland. Forestry, 80(5): 567-586.
McElhinny, C., Gibbons, P., Brack, C., & Bauhaus, J. 2005. Forest and woodland
stand structural complexity: Its definition and measurement. Forest Ecol.
Manage.: 218, 1-24.
Naiman, R. J., Decamps, H., & Pollock, M. 1993. The role of riparian corridors in
maintaining regional biodiversity. Ecol. Appl., 3(2): 209-212.
Nierenberg, T. R., & Hibbs, D. E. 2000. A characterization of unmanaged riparian
areas in the central Coast Range of western Oregon. For. Ecol. Manage., 129:
195-206.
Nelson, R. F., Gregoire, T. G., & Oderwald, R. G. 1998. The effects of fixed-area plot
width on forest canopy height simulation. For. Sci., 3(44): 438-444.
Olson, D. H., & Weaver, G. 2007. Vertebrate assemblages associated with headwater
hydrology in western Oregon managed forests. For. Sci., 52(2): 343-355.
Pabst, R. J., & Spies, T. A. 1999. Structure and composition of unmanaged riparian
forests in the coastal mountains of Oregon, U.S.A. Can. J. For. Res., 29: 15571573.
Paulo, M. J., Tomé, M., Otten, A., & Stein, A. 2005. Comparison of three sampling
methods in the characterization of cork oak stands for managementpurposes.
Can. J. For. Res., 35: 2295-2303.
Richards, F. J. 1959. A flexible growth function for empirical use. J. Exp. Botany.
10(2): 290-300.
Richardson, J. S., & Danehy, R. J. 2007. A synthesis of the ecology of headwater
streams and their riparian zones in temperate forests. For. Sci., 53(2): 131-147.
SAS Institute Inc. 1990. SAS/STAT User’s Guide, Version 6, Fourth Edition.
Volume 2. SAS Institute Inc., Cary, NC 848 pp.
Schreuder, H. T., Banyard, S. G., & Brink, G. E. 1987. Comparison of three sampling
methods in estimating stand parameters for a tropical forest. For. Ecol.
Manage., 21: 119-127.
Schreuder, H. T., Rennie, J. C., & Williams, M. 1992. Comparison of three sampling
schemes for estimating frequency and D2H by diameter class - a simulation
study. For. Ecol. Manage., 50: 117-131.
Smith, C. Y., Moroni, M. T., & Warkentin, I. G. 2009. Snag dynamics in post-harvest
landscape of western Newfoundland balsam fir-dominated boreal forests. For.
Ecol. Manage., 258: 832-839.
Sparks, J. C., Masters, R. E., & Payton, M. E. 2002. Comparative evaluation of
accuracy and efficiency of six forest sampling methods. Proc. Okla. Acad. Sci.,
82, 49-56.
69
Stamatellos, G. S. 1995. Comparison of point and point-3P sampling for forest height
to diameter ratio estimation with cost analysis. For. Ecol. Manage., 74: 75-79.
Temesgen, H. 2003. Evaluation of sampling alternatives to quantify tree leaf area.
Can. J. For. Res., 33(1): 82-95.
Tokola, T., & Shrestha, S. M. 1999. Comparison of cluster-sampling techniques for
forest inventory in southern Nepal. For. Ecol. Manage. 116: 219-231.
Uowolo, Al. L., Binkley, D., Adair, E. C., 2005. Plant diversity in riparian forests in
northwest Colorado: Effects of time and river regulation. For. Ecol. Mngmnt.
218: 107-114.
Welsh, H. H. Jr., Ollivier, L. M., Hankin, D. G., 1997. A habitat based design for
sampling and monitoring stream amphibians with an illustration from
Redwood National Park. Northwestern Naturalist. 78: 1-16.
Williams, M. S., Ducey, M. J., & Gove, J. H. 2005. Assessing surface area of coarse
woody debris with line intersect and perpendicular distance sampling. Can. J.
For. Res., 35: 949-960.
Wisdom, M. J., & Bate, L. J. 2008. Snag density varies with intensity of timber
harvest and human access. For. Ecol. Manage., 255: 2085-2093.
70
APPENDICES
Appendix A: Sampling Alternatives Illustrated
72
66
60
Y Coordinate (m)
54
48
42
36
30
24
18
12
6
0
0
6
12
18
24
30
36
42
48
54
60
66
72
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A.1 Illustration of the FAP5R sampling alternative at the 20% intensity at the
Bottomline 13 location. The locations of the circular plots drawn around the
sampled trees are approximate.
71
72
63
54
Y Coordinate (m)
45
36
27
18
9
0
-9
-9
0
9
18
27
36
45
54
63
72
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 2 Illustration of the FAP9S sampling alternative at the 20% sampling
intensity at the O.M. Hubbard 36 location. The locations of the circular plots
drawn around the sampled trees are approximate.
72
Y Coordinate (m)
60
48
36
24
12
0
0
6
12
18
24
30
36
42
48
54
60
66
72
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 3 Illustration of the ASTP3 sampling alternative at the Bottomline 13
location and 10% sampling intensity. The rectangular shaped plots were visually
estimated and approximately to scale.
72
72
Y Coordinate (m)
60
48
36
24
12
0
0
12
24
36
48
60
72
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 4 Illustration of the PAST3 sampling alternative at the 20% sampling
intensity at the O. M. Hubbard location. Rectangles show the approximate
location of each plot.
72
Y Coordinate (m)
60
48
36
24
12
0
0
7
14
21
28
35
42
49
56
63
70
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 5 Illustration of the OSSP7 sampling alternative at the 10% sampling
intensity at the Bottomline 13 location. Rectangles show the approximate
location of the 7.2 m by 36 m plot.
73
72
60
48
36
24
12
0
0
6
12
18
24
30
All Trees
36
Stream
42
48
54
60
66
72
Sampled Trees
Figure A. 6 Illustration of the PEST7 sampling alternative at the Bottomline 13
location and 20% sampling intensity.
72
66
60
Y Coordinate (m)
54
48
42
36
30
24
18
12
6
0
0
12
24
36
48
60
72
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 7 Illustration of the HLS08 at the Bottomline 13 location and 20% sampling
intensity. Lines with arrows are the approximate location of the transect line.
74
72
Y Coordinate (m)
60
48
36
24
12
0
0
12
24
36
48
60
72
84
96
X Coordinate (m)
All Trees
Stream
Sampled Trees
Figure A. 8 Illustration of the SEC11 sampling alternative at the O. M. Hubbard
location and 20% sampling intensity.
75
Appendix B: List of Acronyms
APB
APB =
^
Y k −Y
∑
Y
k =1
500
500
^
Where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
ASTP3
The sampling alternative with strips oriented approximately
perpendicular to the stream. Plots were 3.6m by 36m and were
staggered as one moved from one side of the stream to the other.
ASTP7
The sampling alternative with strips oriented approximately
perpendicular to the stream. Plots were 7.2m by 36m and were
staggered each side of the stream.
BAF
Basal area factor.
BAPH
Basal area per hectare.
FAP5R
Randomly sampled circular fixed area plots of 5.64 meters radius.
FAP9R
Randomly sampled circular fixed area plots 9 meters in radius.
FAP5S
Systematically sampled circular fixed area plots 5.64 meters in radius.
FAP9S
Systematically sampled circular fixed area plots 9 meters in radius.
H/D
Height to diameter ratio. The height of a tree divided by the diameter
of a tree in a common unit of measure.
HLS08
Horizontal line sampling using a BAF of 8 (metric) to sample a transect
22 m in length oriented approximately perpendicular to the stream.
HLS10
Horizontal line sampling using a BAF of 8 (metric) to sample a transect
22 m in length oriented approximately perpendicular to the stream.
76
500 ^
∑Y
k
−Y
k =1
MAD
MAD =
500
^
Where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
PAST3
Stratified random sampling was used to sample strips oriented
approximately parallel to the stream. Strips were 3.6m in width and
36m in length.
PAST9
Stratified random sampling was used to sample strips oriented
approximately parallel to the stream. Strips were 9m in width and
28.8m in length.
PB
^
Y k −Y
∑
Y
PB = k =1
500
200
^
Where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
PEST3
Systematic random sampling used to sample strips oriented
approximately perpendicular to the stream, 3.6m wide by 72 meters in
length.
PEST7
Systematic random sampling used to sample strips oriented
approximately perpendicular to the stream, 3.6m wide by 72 meters in
length.

^


Y
Y
−
k
∑


k =1 


500
2
500
RMSE
RMSE =
^
Where Yk is the estimated attribute (TPA, H/D, or BAPH) for the kth
replication and Y is the known attribute value.
77
SEC11
A radial method of sampling where all trees are sampled within two
azimiths. The plots were systematically sampled with a random
starting point and were 11.46 degrees wide.
SEC22
A radial method of sampling where all trees are sampled within two
azimiths. The plots were systematically sampled with a random
starting point and were 22.92 degrees wide.
SRS
Simple random sampling.
STRS
Stratified sampling with a random starting point.
SYRS
Systematic sampling with a random starting point.
TPH
Trees per hectare.