Skip to content
tagtuna edited this page Sep 8, 2015 · 1 revision

Feeding model results to a remote sensing data extraction routine

Xtractomatic is a set of free tools, written in Matlab or in R, that allows client-side access of environmental data served on the ERD/CoastWatch ERDDAP Server. The xtractomatic tools were originally developed for the marine biology tagging community to extract satellite data coincident to the tracks from tagged animals. The packages have been extended to extract a 3D cube of data and time series from the region defined by the polygon.

Preparing model results as lon-lat input data for extraction

The extraction routine essentially requires a longitude and latitude range to define a "box" to download remote sensing data. Since our geolocation models have errors, we use the confidence interval for longitude/ latitude for defining the spatial extent. Assuming your results are in the fit object,

### Longitude range (xlow to xhigh) to extract data from:

    xlow <- fit$most.prob.track[, "x"] - 2 * sqrt(fit$var.most.prob.track[, 1])
    xhig <- fit$most.prob.track[, "x"] + 2 * sqrt(fit$var.most.prob.track[, 1])

Xtractomatic download and usage

Please refer to http://coastwatch.pfel.noaa.gov/xtracto/ and https://github.com/rmendels/xtractomatic

This guide is especially useful and comprehensive: http://coastwatch.pfeg.noaa.gov/xtracto/R/using_xtractomatic_r.pdf