Calflora Logo About Observation Data


Objectives
  • Collect and integrate plant observations data from disparate sources.
  • Provide centralized access to this data through a variety of web interfaces.
  • Provide online visualization tools for this data .
  • Allow users to download as much of this data as possible for their own analyses.
Sources of Observation Data

Observation data in Calflora comes from many sources: records contributed by individuals from smart phone applications, records from land management agencies, herbarium specimen records, checklists compiled by botanists for known locations, plant lists made by environmental consulting companies for development projects, rare plant occurrences documented by the California Department of Fish and Wildlife (CDFW) Natural Diversity Database, survey data from the CDFW Vegetation Mapping Project, et al.

Herbarium specimen records are contributed by the member herbaria of the Consortium of California Herbaria (CCH). Specimen data can additionally be viewed and downloaded directly from here. A list of all data sources is available here.

By providing ready access to all these types of occurrence data, Calflora seeks to facilitate research on questions related to biodiversity, ecology, and conservation, and help researchers get the full benefit of geographic analysis and modeling tools.

Access to Data

  • Use Observation Search to search for observations and view them on a map. This application will show photos attached to individual observations when available.

  • Use Plant Distribution to see the extent of a particular plant. This application will show lines and polygons attached to individual observations when available.

  • Use What Grows Here? to view all plants at a particular location. This application can show the distributions of several plants on a map at the same time.

  • Use Observation Download to download observation data for further analysis in various formats.
  • How Data is Assimilated

    The intent is to provide a core set of data about each observation and enough information for the user to select only those records that are suitable for their purpose. When possible, links are provided to the data source for additional information. When we receive bulk data sets, we examine them to extract:
    • a plant name, a location, and a date
    • basic information about the site if available
    • the source institution and individuals responsible for the observation
    • the method used by the observer or collector to identify this observation
    Individual contributors answer these same questions by filling out a form, either in a smart phone application or in a web application.
    Quality Control

    When you see an observation of a plant at a certain location on Calflora, does it mean that the plant is (was) really there? Because Calflora collects data from diverse sources, each with their own strengths and weaknesses, there is some filtering of the data which you, the user, must do to make the best use of it.

    Here are some error conditions to watch out for:

    • the plant is erroneously identified
    • the record is incorrectly georeferenced (eg. the coordinates are not in the stated county)
    • the observer or collector did not realize that the plant in question was actually cultivated

    Whenever possible, we try to identify such records and to label them as suspect. This is an ongoing task as new data continues to come in. Meanwhile, here are several strategies for determining the level of evidence for a plant at a location:

      1. The presence of a plant in an area will be reinforced by other nearby observations of the same plant. Look at the number of records for a particular plant in a particular area. As the observation database grows larger, this becomes an increasingly effective technique. For instance, if you see that there is only a single record for a particular plant in the area of interest, and it is far away from other observations of the same plant, you may choose to discount it. Some Calflora web applications are set up to enable this kind of filtering:
       
      • From the map on a Taxon Report page you can quickly get an idea of how many records there are of the plant in a county or part of a county.
      • In the What Grows Here? application, if you set "Minimum Record Count" to two or more, you will eliminate singleton plant records from the plant list for the target area.
       
      2. Pay attention to the individual observers and collectors responsible for a record. Some are more reliable at plant identification than others.
    All that said, if you see a plant observation on Calflora that looks like it has a problem, you can put a comment on the record explaining what is wrong. (We do pay attention to comments!) To add a comment to a record, make sure you are registered as a contributor and signed into Calflora, then go to the DETAIL page for the record. Open COMMENTS, then click ADD A COMMENT, then enter your comment and press SUBMIT.
    Scientific Name Changes

    Calflora preserves the scientific name chosen by the contributor. A contributor can change the scientific name on a record at any time, but Calflora does not automatically change scientific names. If a scientific name changes, and the change in unambiguous (for instance, from Rhamnus californica to Frangula californica in 2012), Calflora will keep the old name, but make it resolve to the new name.

    As an example, here is the DETAIL PAGE for a record that Michael O'Brien contributed in 2010. He chose the scientific name Rhamnus californica. When the record was first entered, Rhamnus californica was an accepted name, and so clicking on the Rhamnus californica link went to the Taxon Report page for Rhamnus californica. Currently, clicking on the Rhamnus californica link goes to the Taxon Report page for Frangula californica (which shows Rhamnus californica as a synonym or alternate name).

    Location Quality

    When GPS enhanced devices are used to make observations, the device typically produces an error radius understood as the accuracy of the device at the moment the observation was made. If an observation arrives with an error radius of 2 meters, it indicates that the plant was growing somewhere in an area of 12.5 square meters around the given coordinates.

    Some specimen records from the Consortium of California Herbaria were georeferenced after the fact (eg. from written location descriptions), and are assigned a large error radius. If an observation arrives with an error radius of 2 miles, it indicates that the plant was growing somewhere in an area of 12.5 square miles (8,000 acres) around the given coordinates.

    In order to integrate these diverse record types, Calflora classifies the quality of location data in several levels as follows.

      Radius (meters) Area (square meters) Area (acres) Quality
      1 ≤ 5.6 ≤ 100 ≤ 0.025 HIGH
      2 ≤ 44 ≤ 6,000 ≤ 1.5
      3 ≤ 76 ≤ 18,000 ≤ 4.5
      4 ≤ 139 ≤ 60,000 ≤ 15 MEDIUM
      5 ≤ 489 ≤ 750,000 ≤ 185
      6 ≤ 798 ≤ 2,000,000 ≤ 494 LOW
      7 ≤ 3090 ≤ 30,000,000 ≤ 7,413
      8 > 3090 > 30,000,000 > 7,413
    Location Quality (HIGH, MEDIUM, LOW) is available as a selection criteria in the Observation Search application and the What Grows Here? application. When Error Radius is available, Location Quality is assigned from Error Radius according to the table above. When Error Radius is not available, Location Quality is assigned based on other factors such as the method used to geolocate the record.

    Internally, Calflora uses these quality levels for various purposes. One purpose is to select which points are appropriate for making climate and soil profiles, according to the resolution of the available climate and soil map layers. (If the accuracy is better than 15 acres (60,000 square meters), the point is usable for a soil profile. If the accuracy is better than 185 acres (750,000 square meters), the point is usable for a climate profile.)

     
    Calflora  •   1700 Shattuck Av #198, Berkeley, CA 94709  •  510 883-3148  •   Contact Calflora