I have backscatter (radar image value) measurements of corn fields taken at multiple points along the growing season. I can estimate the expected backscatter value of corn by plotting the mean backscatter response at each time point. For any one time point I can calculate the standard deviation, which is useful to know because at some times during the year there is a wider range of plant behavior than at other times. The model of corn behavior can be used along with models for other crops to do crop classification.

Now I want to take the model I built during 2016 and try to apply it to images collected during 2017 (without having to collect a bunch of 2017 field data and build a new model from scratch). The radar images won't be collected on exactly the same dates in the two years, so I would like to estimate the average backscatter response at new dates. For example if I used images from May 1st and May 20th in the 2016 model, but for 2017 had an image from May 5th. To estimate the mean backscatter response I can do a simple linear interpolation between the two nearest time points in the model. Can I do the same thing for the standard deviation measurements? I'm trying to not accidentally make totally wrong mathematical assumptions with my standard deviation values (not being a statistician by profession).

In this image I've illustrated my original data (top), and the model I am trying to build (bottom).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.