Transcript Developing and testing spatial/knowledge based

NFA
Developing and testing
spatial/knowledge based models
for ecosite mapping and
rare species habitat prediction
Dr Bob Maher, Senior Research Scientist
David Colville, Research Scientist
Heather Stewart, Research Associate
Kim Wahl, Research Associate
Applied Geomatics Research Group (AGRG), Middleton, NS
Centre of Geographic Sciences
February 9, 2004
NFA
Outline
 1)
Project Overview
 2) Data Sets
 3) Ecosite Mapping Methodology
 4) Rare Species Habitat Prediction Methodology
 5) Deliverables
 6) New Proposal
NFA
Project Structure
NS DNR
ELC expertise
AC-CDC
Rare plant expertise
AGRG
GIS modeling expertise
ecosite mapping
Rare plant habitat prediction
NFA
Data Flow
NS DNR
FEC plot (points)
Geographic Database
AC-CDC
Rare plant occurrence (points)
AGRG
NFA
NS DNR Data
The NS DNR dataset represents the Forest Ecosystem Classification (FEC) plots
established by the Nova Scotia Department of Natural Resources (DNR) within
the NFA study area. FEC plot data was been collected since 2000 and represents
107 ground-based plots.
The FEC database contains the following six separate files:
1. FECBA
2. FECLC
3. FECPLOT
4. FECPRFIL
5. FECROCK
6. FECSROCK
Information relating to basal area, age and height
Information relating to lower canopy
Site information relating to slope, aspect, crown closure,
microtopography, and slope position
Soil profile information
Stoniness classes
Additional Stoniness descriptions
NFA
NS DNR continued
Note: This is a partial list of the 42 attributes that describe the FECPLOT dataset











CROWNCLOSE
COVERTYPE
SURFSTONE
MICROTOPO
SLOPEPOS
SLOPE
ASPECT
DRAINAGE
PARENTMAT
SOILCLASS
PCOMMENT
Crown closure as a percent
Three classes (hardwood, softwood and mixed wood)
Surficial stoniness class
Single letter class from (L) level to (E) extremely mounded
Letter code from 0 to 8 for slope classes
Slope in degrees
Aspect in degrees
Letter code for 5 classes from very poor to very rapid
Name of parent material
Letter code description of soil type
Helpful comments relating to cover assessment
NFA
AC-CDC Data
The AC-CDC dataset contains all rare plant species occurrences contained within the
Atlantic Canada Conservation Data Centre (AC-CDC) rare species database,
combined with additional historical rare plant records for the NFA study area,
observed as part of the Nova Forest Alliance (NFA) Rare Plant Project (Blaney et al.,
2000). These data consist of a total of 230 occurrences.
Attributes
Note: This is a partial list of the 31 attributes that describe the ACCDC dataset
GNAME
Global Scientific name of taxon
SRANKNS
Provincial Rank(NS) of taxon
PREC
Precision in meters (by power of 10)
SURVEYSITE
General locality of occurrence
SURVEYDATEDate survey occurred
EODATA
Size of population, phenology, population information
GENDESC
Site description, characteristics such as slope aspect
NFA
Geographic Data Sets
1.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Base Layers – hydrology, roads, etc.
Slope Gradient
Slope Aspect
Bedrock Geology
Surficial Geology
Soil Classification
Watersheds
Landscapes
Ecological Land Classification
Forest Inventory
NFA
Ecosite Mapping Methodology

Step 1:
Sept 2:
Step 3:
Step 4:

Step 5:

Step 6:

Step 7:

Step 8:



Model ecosites from DEM and other layers
Point in Polygon overlay
Compare each point’s attributes with each polygon’s attributes
Do FEC plots that represent different ecosites have different or same
attributes from polygon layer?
In particular, from DEM what is relationship between FEC plot and
slope position characteristics?
For those known FEC plots, find all polygons that have same
attributes
If two FEC plots are supposedly different and yet have same
attributes, what does this mean?
Iterate Process. Ultimate objective – map of unique ecosites that
are supportable by field data
NFA
Rare Species Habitat Prediction Methodology
There are two methods to define habitat requirements for a species:
1: Define habitat criteria from literature, studies in local area and expert
knowledge
2. Define habitat criteria from the characteristics from layers of the
geographic database
To do this:

Step 1: Point in Polygon analysis

Step 2: For each species analyze the polygon attributes and make
comparison with field data

Step 3: Define habitat characteristics for each species from all
information

Step 4: Produce predicted habitat map for each species
NFA
Deliverables




End of January 2004
1. Geographic Atlas for NFA model forest
2. NS DNR and AC-CDC point database descriptions
3. Preliminary analysis of ecosite characteristics
4. Preliminary analysis of habitat criteria for rare plant species
End of March 2004



5. Final Report
6. GIS models for Ecosite Mapping and Rare Species Habitat
Prediction
7. Spatial databases for future Research, where possible, with
permission from government agencies
NFA
New Proposal
1.
 2.
 3.
 4.

Field validation of sites predicted by models
GIS Analysis including new field data
Refinement of the model
Workshop