National Urban Forest Conference - PowerPoint PPT Presentation

1 / 39
About This Presentation
Title:

National Urban Forest Conference

Description:

Title: No Slide Title Author: DIST/ISD Last modified by: DIST/ISD Created Date: 8/31/2001 3:31:35 PM Document presentation format: On-screen Show Company – PowerPoint PPT presentation

Number of Views:131
Avg rating:3.0/5.0
Slides: 40
Provided by: DIST67
Category:

less

Transcript and Presenter's Notes

Title: National Urban Forest Conference


1
Wireless Forestry A Data Collection Model
  • Presented by
  • Chris Daniel
  • Montgomery County Government
  • Department of Information Systems and
    Telecommunications
  • 110 N. Washington St, Suite 501
  • Rockville, MD 20850
  • chris.daniel_at_co.mo.md.us
  • http//gis.co.mo.md.us

2
Acknowledgements
  • Montgomery County, MD Government
  • Lonnie Darr (Dept. of Environmental Protection)
  • Jim Himel (Forester)
  • Apollo Teng (GIS Manager, Dept. of Info. Systems
    and Telecommunications - Geographic Information
    Systems or DIST-GIS)
  • Andrew Greene (GIS Intern, DIST-GIS)
  • American Forests
  • Gary Moll (Vice President of Urban Forestry)
  • Ken Gorton (GIS Specialist)
  • United States Forest Service (Northeast Forest
    Exp. Station)
  • Dr. Wayne Zipperer
  • Dr. David Nowak

3
Purpose
  • Develop and test field survey methods for
    collecting and validating data using traditional
    data collection methods and wireless Geographic
    Information System (GIS) and Global Positioning
    System (GPS) technologies

4
Objectives
  • Develop a Study Plan
  • Obtain Approval
  • Implement Study Plan
  • Present or Publish Findings

5
Study Plan
  • Select the study site
  • Determine data collection methods
  • Determine method evaluation criteria
  • Identify data requirements
  • Evaluate and obtain survey equipment, hardware,
    and software
  • Hire contractor or train staff
  • Prepare for field survey
  • Conduct field survey
  • Perform post-processing and quality assurance
  • Analyze data
  • Present Findings

6
Study Site Bethesda Neighborhood
7
Data Collection Methods
  • Traditional
  • Hard-copy field inventory Sheets and Maps
  • ArcView GIS software
  • Hand-held GIS
  • Compaq IPAQ 3670 with Windows CE Version 3.0
  • ESRIs ArcPAD 5.0.1 and ArcView GIS software
  • Data downloaded locally and via the Web through
    Internet Map Services
  • GPS
  • Trimble GPS Pathfinder Pro XR Receiver with TSC1
    Data Collector
  • Pathfinder Office Software included
  • Pre-planning
  • Data dictionary

8
Traditional Method
9
Hand-held GISCompaq IPAQ with ArcPAD and
Merlin Wireless Modem
10
Trimble Pathfinder Pro XR GPS
Antenna
TSC 1 Data Collector
Pathfinder Office
11
Evaluation Criteria
  • Spatial Accuracy
  • Proximity Analysis conducted using ArcView GIS
  • Time
  • Field Preparation
  • Field Survey
  • Post-processing
  • Analysis
  • Costs
  • Staff
  • Training
  • Materials/Expenses
  • Desktop computer, Compaq IPAQ, ArcPAD ArcView
    software, printer, travel,etc..
  • GIS Data

12
Data Requirements
  • Montgomery County Department of Environmental
    Protection
  • Focus on Street Trees initially then progress to
    residential or privately owned trees
  • Tree ID , Species, DBH, Height, Height to Crown,
    Crown Diameter, Health, Comments
  • Spatial data to be in ArcView shape file
    compatible format
  • Projection Maryland State Plane
  • Datum NAD 83
  • Map units Feet
  • American Forests CITYgreen Model
  • Tree ID , Ownership (Public or Private), Species
    Code, DBH, Height to top of Tree, Reduced in
    Height, Health, Conflicts (overhead wires,
    structure, sidewalk, other trees), Ground Cover
    under canopy (shrubs, grass, pavement, exposed
    soil, mulch)
  • Height Class and Diameter Class were calculated
  • United States Forest Service Urban Forest Effects
    (UFORE) Model
  • Tree ID , Street Tree (Y/N), Species Code, DBH,
    Height to top of Tree, Height to Base of Crown,
    Health, Crown Width, Foliage Density, Percent
    Dieback, of Sides receiveing Sunlight (0
    below 5 above canopy)

13
Equipment, Hardware, and Software
  • All Methods
  • Tree measurement tools
  • Measuring chain or stick, clinometer, DBH tape,
    or equivalent
  • Clipboards and pencils
  • Camera
  • Olympus Digital Camera (3 mega-pixel)
  • Desktop PC and printer used to create and print
    data sheets and maps and to run analysis or
    planning software
  • Dell Precision 330 1.5ghz with 512mb RAM with 20
    gb hard-drive
  • HP 2000 C color printer
  • ArcView GIS Software

14
Equipment, Hardware, and Software
  • Hand-held Computer
  • Compaq IPAQ 206 mhz pentium
  • Windows CE operating system
  • 16 mb ROM (Flash) . Could expand up to at least
    128 mb
  • 64 mb RAM
  • Color display (240 x 320)
  • ArcPAD 5.0.1 and ArcView 3.2 software
  • Desktop PC (Dell Precision 330 1500 mhz with 500
    Kb RAM and 20 GBs hard drive) with Windows
    2000
  • Merlin Cellular Digital Packet Data (CDPD)
    Wireless Modem - 19.2 kbps (Optional)
  • Voice recognition software (Optional)
  • Naturally Speaking by Dragon Systems
  • Voice Insight (ArcPAD 6)

15
Equipment, Hardware, and Software
  • Trimble Pro XR/Beacon with TSC1 GPS Unit
  • Backpack unit with a hand-held TSC1 data
    collector
  • Capable of Real-Time Differential GPS Corrections
    (DGPS) through the US Coast Guard Beacon system
    or a private satellite subscription service
  • Spatial accuracy to 50 cm
  • Pathfinder Office Desktop Software
  • Desktop PC - (Dell Precision 330 1500 mhz with
    500 Kb RAM and 20 GBs hard drive) with
    Windows 2000

16
Hire or Train Staff
  • Hired a Qualified Professional Forester to assist
    with the tree survey
  • Obtained Trimble GPS training from Sam Walker
    Scinnovations Inc.

17
Field Survey Preparation Traditional
  • Developed an inventory sheet to satisfy data
    requirements
  • Generated hard-copy maps using ArcView GIS
    software
  • Plotted street trees on maps for validation
  • Maps were laminated for protection
  • Covered clipboard used to protect data sheets
  • Coordinated field visits with Contractor

18
Field Survey Preparation Hand-held GIS
  • Install and configure ArcPAD 5.0.1 and ArcView
    GIS software on Dell Precision 330 Desktop PC and
    Compaq IPAQ 3760
  • Created a new tree file and digital inventory
    sheet using ArcView 3.2 and ArcViews Dialog
    Designer Extension
  • Digital inventory sheets reduce error by
    enforcing field validation and incorporating
    pre-determined list boxes or choices specified in
    the data requirements
  • Downloaded data file, digital inventory sheet,
    GIS shape files (roads and street trees), and a
    color digital ortho-photo (one-foot pixel
    resolution) of the study area to Compaq IPAQ
    mobile mapping device with ArcPAD 5.0.1
  • Set up the ArcPAD project file on the IPAQ using
    the downloaded data and digital inventory sheet
  • Incorporate voice recognition (Optional)
  • Voice Insight for ArcPAD 6 soon available
  • Configured the wireless Merlin modem and the
    wireless Internet service for the PDA from
    Verizon Wireless, Inc

19
ArcPAD Digital Inventory Sheet
20
Field Survey Preparation GPS
  • Installed the Trimble Pathfinder software onto a
    Desktop PC and prepared the GPS unit for field
    work (charge batteries and check cables)
  • Created a Data Dictionary in accordance with the
    data requirements using the Pathfinder Office
    software and transferred it to theTSC1 data
    collection unit
  • The data dictionary reduces error by enforcing
    field validation and incorporating pre-determined
    list boxes or choices specified in the data
    requirements
  • Downloaded the latest Almanac Information from
    the local GPS base station (within 300 miles for
    sub-meter) using the Pathfinder software
  • Created a Mission Plan to determine the number of
    available GPS satellites as well as the
    anticipated Position Dilution of Precision (PDOP)
    using the Pathfinder software
  • PDOP is an index used by the GPS device that
    measures satellite geometry and predicts
    positional accuracy. If the PDOP is low, then the
    GPS satellite constellation is spread out, which
    yields a more accurate position.
  • The PDOP mask was set to 6, the number of
    positions mask was set to 1 position per second
    for 30 seconds, and the elevation mask was set to
    15 degrees
  • A new point file was created using the GPS TSC1
    Data Collector. The data dictionary was
    incorporated, so that each time a tree was
    located the surveyor could enter the attribute
    data required

21
Field Survey
  • All Methods
  • Street trees were inventoried initially
  • Consent letters, created to notify local
    residents for permission to inventory trees on
    private property, had not been delivered the
    survey was started
  • The tree location (point) and attribute
    information were recorded in the field
  • A team of at least two collected data
  • One person recorded the tree location, attribute
    information, and measured tree height while the
    other person conducted the other tree
    measurements
  • Photos were taken with digital cameras for
    presentation and web publication
  • Traditional Method
  • Tree location and identification number were
    recorded on a hard-copy map
  • Tree attribute information, including tree
    identification number, were recorded on a data
    sheet
  • A crew of two were able to complete about 30 to
    40 trees per hour during a typical humid
    sub-tropical summer day (90 degrees F with 95
    humidity)

22
Field Survey Traditional Method
23
Field Survey Hand-held GIS
  • Tree location and attribute information,
    including tree identification number, were
    recorded using the Compaq IPAQ/ArcPAD 5.0.1
  • The tree location was eye-balled using
    landmarks identified in the field that could be
    distinguished on the downloaded digital
    ortho-photo
  • The IPAQ color display screen (240 x 320) was
    clearly visible in the field
  • As a tree was added to the database, the digital
    inventory sheet popped up and enabled the
    operator to enter the attribute information
    quickly and easily
  • A crew of two were able to complete about 20
    trees per hour during a typical humid
    sub-tropical summer day (90 degrees F with 95
    humidity)
  • One person recorded the tree location, attribute
    information, while the other person conducted the
    other tree measurements
  • The IPAQs LiPolymer battery-power lasted about 2
    hours and 15 minutes. You can recharge the
    battery using an automobile A/C adapter. However,
    be careful not to plug it in when you start-up
    the car
  • The wireless data connection (19.2 kbps) from the
    County GIS Internet Map Server (IMS) did not work
    in the Study Area
  • When a connection was established (outside of the
    study area), it took about 1 minute to refresh
    the IMS photo (19.2kpbs) while it took lt 5
    seconds to bring up the photo stored on the
    IPAQs Flash ROM
  • The downloaded photos were used in the field to
    map the trees

24
Field Survey Hand-held GIS
25
Look for Wireless Redwoods near you!
26
(No Transcript)
27
Field Survey GPS
  • Tree location and attribute information,
    including tree identification number, were
    recorded using the Trimble Pathfinder Pro XR GPS
    receiver and TSC1 Data Collection unit
  • The tree position was recorded in the GPS data
    receiver/logger for at least 30 seconds. This
    would yield sub-meter positional accuracy
    required
  • The Trimble LCD screen (240 x 200) was clearly
    visible in the field
  • As a tree position was added to the GPS database,
    the data dictionary popped up and enabled the
    operator to enter the attribute information
    quickly and easily
  • A crew of two were able to complete about 20
    trees per hour during a typical humid
    sub-tropical summer day (90 degrees F with 95
    humidity)
  • One person recorded the tree location, attribute
    information, while the other person conducted the
    other tree measurements
  • Tree positions were recorded when PDOPs lt 6.
    When a structure or heavy tree canopy/biomass
    blocked the satellite feed, then the surveyor
    conducted an off-set. This occurred on lt 10 trees
    measured
  • The GPS batteries lasted all day (at least 8
    hours), provided that they are charged the night
    before conducting the field survey

28
Field Survey GPS Method
29
Post-processing and Quality Assurance
  • Traditional Method
  • Attribute information from the field data sheets
    were entered into a Microsoft Access database
  • Tree points were interpolated from the maps and
    digitized into a GIS shape file using ArcView
  • Property lines, building foot-prints, and roads
    were used to interpolate the position of the
    trees
  • Reviewed attribute data for data entry mistakes
  • Spelling mistakes
  • Incorrect Species Codes
  • Examine the Comments field
  • Placed tree points over the digital ortho-photo
    using ArcView and visually reviewed the spatial
    accuracy to ensure that the data points were in
    the correct map projection

30
Post-processing and Quality Assurance
  • Hand-Held GIS Method
  • Download GIS shape file or data file to Desktop
    PC
  • Reviewed attribute table for data entry mistakes
  • Overlayed GIS data on digital ortho-photo and
    visually reviewed spatial accuracy
  • Wrote ArcView script to Grow Trees based on
    crown diameter for CITYgreen model
  • Trimble Pro XR GPS Method
  • Transferred data to Desktop PC
  • Downloaded base-station differential correction
    files
  • Performed differential correction using
    Pathfinder Office software
  • Converted to GIS shape files with the correct map
    projection parameters
  • Reviewed attribute table data entry mistakes
  • Overlayed GIS data on digital ortho-photo and
    visually review spatial accuracy
  • Check the off-sets

31
Analysis
  • Sample Size 31 trees
  • Used GPS points as the Control
  • Conducted Proximity Analysis using ArcView 3.2 to
    estimate spatial accuracy of the Hand-held GIS
    and Traditional methods
  • Wrote a script to compute distances from GPS
    points to Hand-held GIS and Traditional method
    points
  • Calculated average, minimum, and maximum distance
    to each corresponding tree based on unique
    identifier field (Tree_id)
  • Recorded time and costs for each method

32
Proximity Analysis
Blue Traditional Green Hand-held GIS
Red GPS
33
Findings Traditional
  • Spatial Accuracy (Hand drawn on map and
    digitized)
  • Average 13 feet off
  • Minimum 2.5 feet off
  • Maximum 50 feet off
  • Time
  • Field Preparation 2 hours
  • Field Survey 30 to 40 trees per hour for 2
    hours
  • Post-processing 3 hours
  • Data Conversion (entry) is necessary
  • Analysis 1 hour
  • Total 8 hours
  • Costs (Prices are subject to change)
  • Staff 240 x 2
  • Training 0 per hour
  • Materials/Expenses 4,850
  • Desktop computer (3,500), ArcView (900),
    printer (400), travelfood (30), other (20)
  • GIS Data 425
  • Total Cost 5,755

34
Findings Hand-held GIS
  • Spatial Accuracy
  • Average 7 feet off
  • Minimum 2 feet off
  • Maximum 22 feet off
  • Time
  • Field Preparation 3 hours (On-the-Job-Training)
  • Field Survey 20 trees per hour for 2 hours
  • Post-processing 1 hour
  • Analysis 1 hour
  • Total 7 hours
  • Costs (Prices are subject to change)
  • Staff 210 x 2
  • Training 0 per hour
  • Materials/Expenses 5,845
  • Desktop computer (3,500), IPAQ (500), ArcPAD
    (495) ArcView (900) software, printer (400),
    travel food (30), other (20)
  • Wireless service (300) modem (150) or IMS
    Server (22,000) not included in the price
  • GIS Data 425
  • Total Cost 6,690

35
Findings GPS
  • Spatial Accuracy (Used as the Control)
  • lt 3 feet off or sub-meter accuracy after
    Differential Correction
  • Time
  • Field Preparation 1 hour
  • Field Survey 20 trees per hour for 2 hours
  • Post-processing 1 hour
  • Analysis 1 hour
  • Total 5 hours
  • Costs (Prices are subject to change)
  • Staff 150 x 2
  • Training 495
  • Materials/Expenses 15,850
  • GPS Unit (11,000), Desktop computer (3,500),
    ArcView 3.2 (900), printer (400), travel
    food(30), other (20)
  • GIS Data 425
  • Total Cost 17,070

36
Summary
  • Traditional Method
  • Low cost solution
  • Spatial accuracy is generally unacceptable for
    County base-mapping
  • Need large scale field maps with planimetric data
  • Accuracy could be improved if tree locations were
    measured from landmarks in the field, however,
    this would slow field survey speed
  • Able to survey 30 to 40 trees in the field per
    hour
  • Post-processing takes time
  • Data entry is required
  • Chances of data error are higher
  • Requires little computer training
  • Most popular method

37
Summary
  • Hand-held GIS (Most Scalable Solution for Urban
    Areas)
  • Affordable solution for County Government
  • Easy to use in the Field
  • Portable (fits in your shirt pocket), screen
    resolution is acceptable
  • Voice recognition will be available in ArcPAD 6
    through Voice Insight, Inc.
  • Spatial accuracy is border-line acceptable for
    County base-mapping
  • Landmarks in the field can be interpreted on the
    ortho-photos to enable users to place the tree
    point locations within 7 feet of its exact
    location without measuring. Accuracy could be
    improved if tree locations were measured from
    landmarks in the field, however, this would slow
    field survey speed
  • Able to survey a good average rate of trees per
    hour (20)
  • Very little post-processing required
  • Attribute data stored on IPAQ in the field
  • Reduces post-processing error
  • Enables access GIS data sets and ortho-photos
  • Data downloaded onto IPAQ Flash ROM is currently
    the best solution
  • ArcIMS Internet Map Services provide data to
    wireless IPAQ units and ArcPAD enables users to
    use the map services. Existing cellular modem
    technology is very slow (19.2 kbps) and you may
    have difficulties establishing a connection with
    the service provider, especially if you are in a
    low area. Modem technology is improving though
    (i.e. Ricochet, Merlin GT100).
  • Provides connectivity to a GPS device for
    navigation and mapping
  • Good for rural areas
  • Requires hand-held computer and GIS training

38
Summary
  • GPS
  • Somewhat expensive solution
  • If you need engineering or survey quality
    accuracy, this solution works great!
  • Spatial accuracy is excellent for County
    base-mapping
  • Minimal field preparation time
  • Charge batteries, check cables, create data
    dictionary, download Almanacs from base station,
    quick plan, download dictionary to TSC1 data
    collector and go!
  • Able to survey a good average rate of trees per
    hour (20)
  • Occasionally had to wait for the PDOP to drop
    below 6
  • If youre in a deep swale or next to large
    structures, you may have difficulty obtaining a
    good PDOP value (lt6). Heavy tree canopy/biomass
    may also some connection difficulties
  • Backpack could be cumbersome for a smaller framed
    person
  • Very little post-processing required
  • Attribute data stored on data collector in the
    field
  • Reduces data post-processing errors
  • Recommend GPS and GIS Training

39
Useful Web Sites
  • Montgomery County GIS http//gis.co.mo.md.us
  • Montgomery County DEP http//www.co.mo.md.us/servi
    ces/dep/
  • USDA Forest Service, Urban and Community Forest
    Ecosystem Research http//www.fs.fed.us/ne/syracu
    se/
  • American Forests http//www.americanforests.org
  • Trimble GPS http//www.trimble.com/mappinggis.htm
    l
  • ESRIs ArcPAD http//www.esri.com/software/arcpad
  • Compaq IPAQnet Mobility Solutions
    http//www.compaq.com/products/wireless/wwan/index
    .shtml
  • Novatel Wireless http//www.novatelwireless.com/p
    cproducts/merlin.html
  • Verizon Wireless Service http//www.app.airtouch.
    com/mobile_ip/internet_access/pocketpc.html
  • Dragon Systems http//www.dragonsys.com (Voice
    Recognition)
Write a Comment
User Comments (0)
About PowerShow.com