Analyze data
Many of our analysis features are under construction!
Analysis Features
The analysis features and capabilities built into our website include descriptive statistics, the ability to extract pixel values from base layers of predictive raster layers (raster data extraction) and predictive spatial modeling capabilities. To begin, you must first create spreadsheets to use as input data for your analyses. The types of analyses we currently offer include simple linear regression, multiple linear regression, simple logistic regression and multiple linear regression (used for binary data such as presence/absence), and regression tree analyses.
Descriptive Statistics
We are planning on implementing basic descriptive statistics that may be used to evaluate normality and examine the data in columns within the spreasheets you create. These features will help you make the best choices for your modeling methods. These features are under development.
Raster Data Extraction
We are building features to allow you to upload raster datasets and use them to extract data for field points. We also provide low resolution remote sensing datasets including:
- Average annual precipitation PRISM Data
- Average annual temperature PRISM Data
- Minimum average annual temperature PRISM Data
- MODIS three year average Enhanced Vegetation Index - EVI (range)
- MODIS three year average Enhanced Vegetation Index - EVI (mean)
- MODIS three year average Enhanced Vegetation Index - EVI (date of peak)
- Elevation (in meters)
- Distance to water (in meters)
These remote sensing datasets are available anywhere within the contiguous United States and are all at approximately 1000 meter resolution. See the 'Spreadsheets' and 'GeoRasters' links on our navigation menu for more details. You will need to login to use these features.
Predictive Spatial Modeling
Using online statistical tools available here, you can develop predictive models using (1) simple linear regression, (2) multiple linear regression (3) simple logistic regression, (4) multiple logistic regression, or (5) regression trees, and generate a predictive surface based on your selected model. Next, you can overlay your predictive surface onto our map application for your study area and visualize the spatial distribution of your species along with your original species occurrence data you used to create your predictive species. Finally, you may also save your map and/or print it as a PDF file if you wish.
Data sharing agreement | Webmaster | Data usage agreement |