A GPS handheld receiver and a digital camera are seemingly simple tools, but chosen carefully and used properly they can acquire useful and accurate data for a GIS. The objective of my research was to use these tools in a survey of valley oak trees and to assess data accuracy. Another context for the research was created by linking GIS output in the form of a map and tabular attribute data to an educational resource for studying oak trees within the Cosumnes River Preserve.
Within the Cosumnes River Preserve (CRP) in South Sacramento County there are standing many significant and large valley oak trees, a tree species that once dominated the Central Valley landscape. The CRP through education outreach provides curriculum for use in the primary grades to study and explore the rich habitat within the Preserve, though not specifically the largest of the valley oaks. Influenced by the preservation and education efforts of the California Oaks Foundation, the author believes that new curriculum development focusing on locating and studying the Preserve's valley oaks would be a valued educational resource.
Within the CRP the Cosumnes River Walk trail features many large valley oaks viewable in the oak savanna and in the riparian forest along the Cosumnes River. But where exactly are they and which ones are most significant in attributes such as age, diameter, height, and canopy cover? Which ones can be easily hiked to and what is there distance from the trailhead? These are important considerations when conducting an education field trip with students in the primary grades.
This paper describes the use of a handheld GPS receiver, a digital camera, and ArcGIS tools to survey the Cosumnes River Walk trail and catalogue select valley oak trees of interest. A thematic map is then created and is intended to be used in conjunction with select curriculum materials.
Essentially, the use of a handheld GPS receiver and a digital camera to produce data sources for a GIS is well documented and discussed in numerous sources including lecture materials for the class this paper is written for. Capturing track logs and marking waypoints using a GPS handheld is the common work of, for instance, forestry technicians assisting in the preparation of a timber plan. In fact many of the internet literature resources on this subject are from web sites featuring forest management content. Perhaps some years ago when GPS receiver technology first became packaged in a handheld device one may have found substantive technical literature to review and cite, but not today — its use as a GIS data source is common place.
With respect to the related subject of this study on valley oak trees, the book Investigating the Oak Community (Mayolo, 2000) is an informative and an excellent educational resource to accompany the planned thematic map. The book is also the source of field collection methods for oak tree data.
The data sources used in this project included a handheld GPS receiver, a digital camera, field collected data and aerial imagery. Specific collection methods for the GPS receiver and digital camera were found in Quick Start Guide Using Any Digital Camera, Garmin and GPS PhotoLink (Cusick, 2008). Tree measurement methods were found in the book Investigating the Oak Community (Mayolo, 2000). The aerial imagery was acquired from the California Spatial Information Library (CaSIL), specifically the NAIP 2009 Sacramento County extent in MrSID file format.
GPS track and waypoint data was collected using the Garmin eTrex H high sensitivity handheld receiver. To improve overall accuracy the receiver was always set to WAAS mode. Other general setup details were: Position Format to UTM/UPS; Map Datum to WGS 84; Distance Units to Metric.
A single track log was collected for this project for the purpose of measuring the route along the Cosumnes River Walk trail to selected valley oak trees. The configuration of the eTrex Track Setup was: Record Interval to Time; Value to 00:00:05; Wrap When Full to No. After the track was collected, the MN DNR Garmin program was used to upload the active track data which consisted of 625 data points for the 4.6 kilometer route. A sample of the data is shown below. The MN DNR Garmin program was also used to save the track data to an ArcMap unprojected point shapefile.
Waypoint data was collected for each oak tree of interest and this data was later used during the photo georeferencing process. The MN DNR Garmin program was used to upload the waypoint data. Six waypoints were collected and a sample of the uploaded data is shown below. The MN DNR Garmin program was also used to save the waypoint data to an ArcMap unprojected point shapefile.
The camera used was a Canon digital PowerShot A560 with 7.1 megapixels and 4x optical zoom. Since the photos were taken for reference only and not subject to analysis, the only collection method worth noting was the setting of camera time to GPS time and photographing the GPS screen showing time. The GPS time shot was used later during photo georeferencing for determining the time difference between the camera and GPS unit.
Photo record of GPS receiver time
Tree circumference, geographic location and photographs were recorded for six of the largest valley oak trees observed in the oak savanna along a looping portion of the Cosumnes River Walk trail. Tree circumference was measured and recorded at a standard 1.4 meters (4.5 feet) from the ground. Geographic location was recorded using the GPS unit's waypoint record feature.
Data analysis consisted primarily of evaluating the accuracy of the collected GPS data. The GIS was a standard desktop PC running ESRI ArcMap and ArcCatalog 9.3 programs. First, the NAIP Sacramento County 2009 MrSID image was imported into ArcMap, and then a smaller extent showing just the area of interest within the Preserve was saved as an image file. The saved image file was then imported and became the base layer image. The track and waypoint shapefiles were then imported, projected and analyzed for accuracy. The resulting overlays and accuracy assessment are presented in the sections below.
GPS data samples, field collected data and the thematic map are shown below.
| type | ident | lat | long | y_proj | x_proj | new_seg | display | color | altitude | depth | temp | time | model | filename | ltime |
| TRACK | ACTIVE LOG | 38.26606750 | -121.44048929 | 28713.9632819849 | -125872.655514035 | true | False | 255 | 7.827880859375 | 0 | 0 | 2009/12/06-21:38:19 | eTrex H | 2009/12/06 13:38:19 |
| type | ident | lat | long | y_proj | x_proj | comment | display | symbol | unused1 | dist | prox_index | color | altitude | depth | temp | time | wpt_class | sub_class | attrib | link | state | country | city | address | facility | crossroad | unused2 | ete | dtype | model | filename | ltime |
| WAYPOINT | 001 | 38.26276612 | -121.43835258 | 28344.1751568954 | -125691.525966917 | 0 | 178 | 0 | 0 | 0 | 255 | 13 | 0 | 0 | 0 | 96 | 0 | 0 | 0 | eTrex H |
| Tree Name | Geographic Coordinates | Circumference |
| Valley Oak 1 | lat. 38.26276612, long. -121.43835258 | 387 cm |
| Valley Oak 2 | lat. 38.26171185, long. -121.43369601 | 353 cm |
| Valley Oak 3 | lat. 38.26248902, long. -121.43280543 | 370 cm |
| Valley Oak 4 | lat. 38.26107600, long. -121.43100928 | 380 cm |
| Valley Oak 5 | lat. 38.25953012, long. -121.42952853 | 331 cm |
| Valley Oak 6 | lat. 38.25911438, long. -121.43181545 | 290 cm |
The GPS data, both track and waypoint, appears with a high degree of accuracy when overlaid onto the NAIP imagery except for waypoint number 1 (Valley Oak 1). The method used for collecting waypoints was changed after waypoint 3 (Valley Oak 3) when it was discovered that the coordinate fix was locked immediately after entering the waypoint Mark menu but before actually marking the point. Up until then I had been setting up for waypoint collection by entering the Mark menu while located on the trail adjacent to the tree of interest and then walking up to the tree to mark the point -- a lateral change in distance of between 5 to 10 meters. This can be seen in the location of waypoints 2 & 3 which accurately mark me standing on the trail adjacent to the tree. The remaining waypoints 4 through 6 were collected by standing next to the tree and after observing a stable position fix, then entering the Mark menu and collecting the point. This, however, does not explain the accuracy error for waypoint 1.
A notable failing that was discovered about the Garmin eTrex H receiver is its inability to save any type of timestamp when collecting waypoint data (see waypoint data sample above). This renders waypoint data unusable for photo georeferencing. Timestamp data is only available in eTrex H track log records. In the end, because I took only a small photo set and recorded field notes at each location, I was able to easily georeference the photos manually without the use of software.
Seemingly simple data acquisition tools such as a low-cost handheld GPS receiver and digital camera can provide accurate and useful data for input into a GIS. This project successfully utilized the data collected with these tools to produce an accurate map showing the location and distance of selected valley oak trees within a survey area.
Cusick, J., 2008. Quick Start Guide Using Any Digital Camera, Garmin and GPS PhotoLink.
Mayolo, K.A., 2000. Investigating the Oak Community. Appendix 3:104-105.