modeling mortality for mixed species stands in coastal british columbia l.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Modeling mortality for mixed species stands in coastal British Columbia PowerPoint Presentation
Download Presentation
Modeling mortality for mixed species stands in coastal British Columbia

Loading in 2 Seconds...

play fullscreen
1 / 26

Modeling mortality for mixed species stands in coastal British Columbia - PowerPoint PPT Presentation


  • 89 Views
  • Uploaded on

Modeling mortality for mixed species stands in coastal British Columbia. Leah Rathbun PhD Candidate, University of British Columbia Tuesday, June 23 rd 2009 Co-advisors: Dr. Valerie LeMay Dr. Nick Smith. Background.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

Modeling mortality for mixed species stands in coastal British Columbia


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
modeling mortality for mixed species stands in coastal british columbia

Modeling mortality for mixed species stands in coastal British Columbia

Leah Rathbun

PhD Candidate,

University of British Columbia

Tuesday, June 23rd

2009

Co-advisors: Dr. Valerie LeMay

Dr. Nick Smith

background
Background
  • Vancouver Island, the Queen Charlotte Islands and along the coastal mainland
  • Coastal Western Hemlock (CWH) BEC zone
  • Multiple subzones:
    • very dry maritime subzone in the east
    • moist maritime subzone in the central area
    • very wet maritime and hypermaritime subzone in the west
background3
Background
  • Second growth multi-species stands
  • regenerated both naturally and from plantings
objectives and hypotheses
Objectives and Hypotheses
  • Develop survival models for western hemlock, Douglas-fir, and western redcedar
  • Hypotheses:

1) A model developed solely for coastal BC would outperform existing models developed for similar species in other areas

2) The inclusion of inter-tree competition variables such as basal area of larger trees, would be highly related to the probability of mortality for trees located in mixed- species stands with structural diversity.

slide5
Data
  • 2,265 untreated plots
  • 0.008 to 0.806 ha
  • 1932 to 2002
  • 1 to 17 year measurement intervals
  • 160 to 11,750 stems ha-1
  • Site index 6.2 to 52.8 m
  • Average annual plot mortality 0 to 12% with an average of 1.13%.
model development
Model Development
  • Logistic survival model
  • Variety of measurement periods, q
  • Where f(X) is a linear function of the predictor variables at the beginning of the projection period, X, and q is the number of years within a projection period.
variable selection
Variable Selection
  • Tree size and stage of development
    • Dbh (cm)
    • Relative Diameter (RDBH)
    • Height (m)
  • Site productivity
    • Site index (m)
    • Growth Effective Age (GEA)
  • Inter-tree competition
    • Curtis’ Relative Density
    • G (m2/ha)
    • SPH (stems ha-1)
    • Basal area of larger trees (BAL, m2/ha)
    • Crown competition factor of larger trees (CCFL)
selected model
Selected Model
  • Important variables were:
    • Dbh
    • Height
    • BAL
    • Curtis’ relative density
    • Site index
    • Growth effective age
    • Basal area per hectare
selected model9
Selected Model
  • BAL performed the best for western hemlock and western redcedar but not for Douglas-fir
  • BAL, a competition index, was replaced by relative diameter, a measure of size, for Douglas fir
  • Inclusion of competition variables within a mortality model is species dependent for mixed-species stands
model validation
Model Validation
  • Temesgen (2002)
    • Created for cedar-hemlock and Douglas-fir found within interior British Columbia
  • Hamilton (1986)
    • Created for mixed conifer stands in Idaho
model validation12
Model Validation

Western Hemlock

Large Trees (≥7.5cm)

Small Trees (<7.5cm)

Sensitivity- % correct for live trees

Specitivity- % correct for dead trees

objectives and hypotheses13
Objectives and Hypotheses
  • Evaluate three alternative implementation methods.
  • Hypotheses:

1) Little differences between approaches would be seen when results are aggregated to the stand level. However, differences would become more evident when analyzed within diameter classes

2) Differences would be more apparent for plots with higher mortality rates

3) Of the threshold probabilities, the optimum threshold value would return the best results and should be determined individually for each species within a mixed-species stand

model implementation
Model Implementation

Using the observations at the beginning of the first (establishment) measurement period to predict the probability of survival at the end of the first measurement period using three different methods, two deterministic and one stochastic.

probability multiplier
Probability Multiplier
  • The probability of survival is used to estimate the surviving stems ha-1 at the end of the period

SPHq = p(s)* SPH0

threshold values
Threshold Values
  • The probability of survival and a specified threshold value are used to estimate the surviving stems ha-1 at the end of the period
threshold values17
Threshold Values
  • Two subjective threshold values, 0.50 and 0.75
  • One optimum threshold value, Cohen’s (1960)

optimal threshold, 0.83

Where: po is the proportion of units in which the predicted and actual outcomes agree and pc is the proportion of units for which agreement is expected by chance

stochastic
Stochastic
  • For a single trial, the probability of

survival and a random number are

used to estimate the surviving stems

ha-1 at the end of the period

The number of trials used was varied as 1, 5, 20, and 1,000 times.

dbh class mean bias values
DBH ClassMean Bias values

Western hemlock

Douglas-fir

Western redcedar

dbh class
DBH Class
  • Trees less than 7.5cm in dbh
  • Generally located within the understorey in mixed-species stands
  • Mortality rates dependent upon differences in shade tolerance
  • Western hemlock and western redcedar vs. Douglas-fir
  • The development of separate mortality models may improve predictions for these smaller trees.
mortality class mean bias values
Mortality ClassMean Bias values

Western hemlock

Douglas-fir

Western redcedar

mortality class
Mortality Class
  • 122 plots, greater than 4% average annual mortality
  • Average tree size was generally larger, both dbh and height
  • Larger values for stems per hectare and basal area per hectare
  • Similar range in Site Index values
  • On average comprised of 58.9% western hemlock, 27.6% Douglas-fir, and 13.5% western redcedar.
  • 70.5 % of plots established in the 1950’s and 1960’s
  • In 1966 through 1968, coastal BC experienced lower than average rainfall for the summer months.
where do we go from here
Where do we go from here?
  • Currently working on the addition of treatment regimes to the mortality models (variations in thinning events, addition of fertilization)
  • Also currently working on diameter and height increment models
acknowledgements
Acknowledgements
  • Ken Epps and Island Timberland for providing the data for this project
  • Forest Science Program of British Columbia for funding this project
  • Dr. Peter Marshall, Dr. Lori Daniels, and Dr. David Hann
slide26
I like trees because they seem more resigned to the way they have to live than other things do.- Willa Cather