Transcript Document

Geographic
Information
Systems
Applications in Natural Resource Management
Chapter 10
Updating GIS databases
Michael G. Wing & Pete Bettinger
Chapter 10 Objectives
 Why
GIS databases need to be
periodically updated and maintained,
 What issues might be associated with
an update process, and
 What GIS processes could be used to
physically update a database.
GIS databases are rarely static

Particularly in forestry and natural resources






Vegetation change due to time, disturbances, and management
activities
Roads and trails are built and are sometimes obliterated
Stream conditions and structure changes
GIS databases should be updated periodically to reflect
changes that occur in resources over time
Either the spatial locations, the attributes of the
locations, or both the locations and attributes will require
updating
The question is how often and thoroughly a database
should be updated
Table 10.1 Reasons for updating GIS databases
Events
Stochastic disturbances
Transitions of forests
Management activities
Transactions
Regulations
Organizational policies
Improvements in
technology
Organizational initiatives
New data availability
Changing map projections
Collaborative projects
Periodic maintenance
Examples
Hurricanes, fires, insect outbreaks
Growth and yield
Harvesting, road construction,
installation / removal of culverts,
creation of trails, thinnings.
Land acquisitions, donations, sales
Riparian management areas, owl
habitat areas, woodpecker habitat areas
Special areas, personal reservations
Digital orthophotographs, GPS capture
of road data, ownership boundaries, etc.
Periodic / annual cruises,
photo interpretation of harvested areas
not normally recorded via normal processes
Databases developed by other organizations
New organizational standards
Watershed analyses, landscape planning
efforts
Cleaning up databases after spatial
operations, after digitizing or
attributing processes
Update
Spatial
Data
X
X
Update
Tabular
Data
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Process A
Process B
Process C
Digitize an
area
Digitize an
area
Collect GPS
data for an
area
Erase overlaps
using corporate
database
Erase overlaps
using corporate
database
Edit spatial
features (e.g.,
remove multipath)
Attribute the new
spatial features
Select and
copy newly
digitized features
Select and
delete features
to be updated
(if necessary)
Update
corporate
database
Paste new features
into corporate
database
Add GPS
features into
corporate database
Attribute the new
spatial features
Attribute the new
spatial features
Figure 10.1. Three example of update processes related to a land acquisition.
GIS database updates
Many organizations have developed
processes and protocols to guide
database updates
 The update interval is the period of time
between updates

 May
range from hours to years
 Will depend on organizational resources and
database analysis goals
Organizational resources and goals


Updates require investments of time and
personnel
If the goal of a forest management company is to
generate revenue for stockholders
 timber
resources should be updated more frequently
than other resources, such as hiking trails
 updating resources such as roads, streams, and
culverts may or may not be important

In addition to determining the update interval,
organizations will also need to determine how
best to meet update needs
Table 10.2. Inputs and process that can
be used to assist a GIS database update.
Input:
Hand-drawn maps
Tabular databases
A person's memory
GPS features
Field notes
GIS features developed by field personnel
GIS processes:
Digitize
Join
Link
Import
Querying & verification
Scan
Update
Copy / paste
Attributing
Overlay
Annual Update Example

Example: A forest management company in
Florida has decided to update its forest stand
inventory annually- why?
 This
GIS database is perhaps the most important
 Most corporations require an annual estimate of the
value and volume of corporate holdings for tax and
planning purposes
 Given the relatively short rotation that is typical of
industrial forests in the south, a longer update interval
may not be sufficient
 A shorter update interval will lead to increased costs
and may add confusion among field foresters if
databases change too often
Updating a forest stand database

Inventory changes can be recorded
through:
 Hardcopy
maps, cruise forms, or through
digital means (GPS)
Field foresters, timber procurement
managers, and others will note the changes
on hardcopy maps
 The hardcopy maps will be sent to a central
office

 Checked for errors
 Updated into the GIS database
 GIS database checked for errors
 GIS database delivered to field offices
Continual Update Example

In Washington state, all forest management
plans must be submitted to the Department of
Natural Resources (DNR) for review and
approval
 A map
must accompany the plan
 Must show the relation of the proposed activities
relative to the landscape and any surrounding stream
systems
 The DNR provides a statewide streams database for
mapping purposes
Updating a streams database



The DNR streams database is continuously updated as
new information becomes available
A land owner that surveys a stream and notes changes
from the DNR streams database can submit a change
request
The DNR has a review protocol for change requests



Requests are either denied or approved
Approved requests will be used to update the streams database
Although several months or more may elapse in the
reporting, review, and update process, this is considered
a continual update since an update may occur any time
during the year
Updating existing GIS databases
with new landscape features



Example 1: A land purchase results in the
addition of two stands into an existing GIS
database
Example 2: Incorporating a new trail into a GIS
database
For both examples assume
 The
new features were either digitized or measured
with a GPS or other equipment and are available in a
GIS database
 The new features are initially stored separately from
the existing resource database
Two new stands

The owners
of the Daniel
Picket have
purchased
80 acres
(30.4 ha)
adjacent to
current land
holdings
Figure 10.3. Daniel Pickett forest
stands and land purchase area.
Original stands
Stands in land
purchase area
Vegetation Basal
Stand Acres
type
area
1
2
42.6
37.4
A
B
190
15
Age
MBF
55
7
21.3
0.8
Timber stands
The two new
polygons could
be manually
copied into the
existing
databasehowever,
attributes would
not follow
Stand
1
2
VegType
A
C
30
31
0
0
C
C
Basal Area Age
200
50
175
40
190
110
0
0
45
25
0
0
MBF
21.2
12.9
17.3
4.1
0
0
Figure 10.4. Daniel Pickett forest
stands and land purchase area after
copying and pasting features from
the land purchase GIS database to
the stands database.
Three options for updating

Digitize the shape of the two stands into the existing
stands database


Use a merge process to combine the newly digitized
stands into the existing stands database


Will require a manual entry of attribute data
Must ensure a perfect fit between the adjoining (common) edges
of the new stands and existing stands databases
Use an update command (available through ArcToolbox
and XTools in ArcGIS)

A good fit between databases in the update should also be
present but the update process typically will examine new link
and node intersections in the creation of an output database
Update
Results
Figure 10.5.
Daniel Pickett
forest stands
and land
purchase area
after updating
the stands
database using
the land
purchase GIS
database.
Timber stands
Stand
VegType
Basal Area
Age
MBF
1
A
200
50
21.2
2
C
175
40
12.9
30
C
190
45
17.3
31
C
110
25
4.1
0
A
190
55
21.3
0
B
15
7
0.8
Ensuring a perfect fit: options
 Use
snapping techniques while digitizing to
force the new features to fir the existing
 Use GIS tools
Digitize a shape that overlaps the existing
database
 Use the existing stands layer to erase the
overlap

Original stands
Stands in land
acquisition area
Overlap
Updating a
trails
database


The existing trail
system for the Brown
Tract was digitized
several years ago
from hardcopy maps
Over time, people
will tend to develop
new trails on their
own

These are know as
unauthorized trails
Trails
Roads
Forest
boundary
Figure 10.9. Trail system of the Brown Tract.
A new trail


An unauthorized
trails has emerged
as a candidate for
inclusion into the
existing trail system
The new trail’s
location is captured
by a GPS and
entered into a
proposed trails GIS
database
Figure 10.10.
Proposed new trail
on the Brown Tract.
Trails
Proposed
trail
Roads
Forest
boundary
Integrating
the new trail



Since the databases are
linear, and not polygon,
no need to worry about
area overlaps
A merge can be used to
bring the databases
together
Should ensure however,
that the lines meet at
intersections


Snapping tools may help
in this process
Some attribute work may
need to be done manually
Figure 10.11.
Proposed new
trail and its
relation to another
trail from
the original trail
GIS database.
Trails
Proposed
trail
Collect GPS
data for an
area
Edit spatial
features (e.g.,
remove multipath)
Figure 10.12.
Updated trail
system of the
Brown tract.
Trails
Forest
boundary
Attribute spatial
features
Merge new
features into
original database
Edit spatial
position of new
features
Verify / edit
attribute data in
updated database
Trail
Length
Condition
Source
1
410.5
Authorized
Trails
2
1183.2
Authorized
Trails
42
704.1
Unauthorized
Trails
43
1261.5
Unauthorized
Trails
1
3372.3
Authorized
Proposed_trails
Updating an existing GIS database by modifying
existing landscape features and attributes




This implies that we physically alter existing
databases rather than previous examples where
we created a separate database to capture
changes
Some danger in possibly damaging the existing
database- make a copy!
May be difficult to detect and correct any
mistakes unless processing steps have been
carefully documented
Two examples: using a digital orthophotograph
quadrangles (DOQs) and updating attributes
using a join process
Digital Orthophotograph
Quadrangles (DOQs)




Given that the projection of a DOQ exactly
matches that of your existing databases, you
can use DOQs to judge database accuracy in
representing resources
If the DOQ and is of sufficient quality, you can
use it to create update information
You can digitize point, line, and polygon features
directly from the DOQ and incorporate them into
existing databases or create new databases
The DOQ can be viewed simultaneously as
other vector layers
Brown Tract


DOQ being displayed
beneath the Brown
Tract boundary
Reveals that either the
boundary database
has an error or that a
harvesting operation
on an adjacent
landowner’s property
was incorrectly located
Figure 10.14. Boundary line issue on the Brown tract.
Uncertain boundary can be
adjusted


Use the DOQ as a guide to adjust the vertices
along the boundary database to match the
DOQ line
If precision and accuracy are an issue, you
may want to capture the boundary location
through a property boundary survey
 Total

station or survey GPS measurements
Updates to the boundary database may
impact other databases!
 Stands
and soils on the Brown Tract
 May want to also update other databases to reflect
the boundary update
Updating attributes using a join
process

In some cases, only attributes may require
updating
 If
stand units (boundaries) have not been altered, only
the growth and structural information need be
updated
 The changes might be estimated through a growth
and yield model



A structural changes could be summarized in the output and
saved in the form of a nonspatial database
The nonspatial database can be joined to the stands
database
The joined data can be used to replace existing structural
data with the updated data
Stand-level
forest
inventory data
Stands GIS
database
Forest growth
and yield model
Figure 10.15. A
process to
update attribute
data in the Daniel
Pickett stands
GIS database.
Summarize
stand-level
data
Stand-level
summaries
Stands GIS
database joined
with summaries
Replace old
summaries with
new summaries
Remove the join
Updated stands
GIS database