Look for complete geospatial metadata in this layer's associated xml document available from the download link

* Metric Name: Fire Ignition Probability, Human-Caused *  

* Tier: 3  

* Data Vintage: 1992 to 2015  

* Unit Of Measure: Probability, 0-1  

* Represent element and pillar: Yes  

* Type and distribution of data: Bimodal distribution representing probability. **** Raw data values range from 0.001 to 0.713.  

* Translation: Where is human-caused fire ignition probability low?  

* Translation method and outcome: Raw data values translated to range from +1 to -1 representing more favorable to less favorable conditions, respectively. The proposition of more favorable is based on the objective of lower human-caused fire ignition probability (negative linear slope). More favorable (translated to +1) set at 0.001, less favorable (translated to -1) < 0.713 for human-caused fire ignition probability. Alternative approaches could be to set -1 at the top value, but that would result in a more favorable interpretation of human-caused fire ignition probability.

 ****

Figure . Histogram and scoring of predicted human fire ignition probability in
Southern California

 ****

Figure . Histogram of translated predicted human fire ignition probability in
Southern California

 ****

Figure . Maps displaying raw metric and translated metric predicted human fire
ignition probability in Southern California

  

* Metric Definition and Relevance: The rasters depicts the predicted human-caused ignition probability for the state of California.  

* Credits: Bin Chen and Yufang Jin, University of California Davis, bch@ucdavis.edu \--University of California Davis