I have 45 cores from 3 fields and I also have Veris P4000 Vis-NIR spectra up to 100 cm deep at those 45 locations.
I would like to take calibration samples from the cores but the question is from which depths?
The final aim is to be able to predict the soil in all depths in all of the 45 cores and then interpolate them in 3 dimensions making a cube map of the field.
I am thinking about Latin Hypercube on the spectra of each of the 45 cores and then take let's say 3-4 calibration soil samples to do lab analysis on.
Any suggestions?
Dear Salman
Greetings
The attached content shall be useful to you.good luck.
Dear Salman
Greetings
The attached content shall be useful to you.good luck.
Hello Salman,
I have published an article about doing the same thing.
attached the article.
Dear Kulvir,
Thank you very much for the thesis.
BTW: the presentation you attached is my own presentation I presented in SSSA 2016 in Phoenix Arizona
@Yaron Ogen
Hello Yaron,
Thanks for sharing your paper.
I'll read it and probably get back to you with more questions :-)
Hi Yaron,
Yes, I did it right after you sent it to me. Thanks for that.
It was an interesting paper. I liked the depth cross sections you produced.
I am going to use most probably fuzzy c-means on spectra in depth to detect different soil horizons and then take calibration samples at those location. Then use the model to predict the rest of the cores.
Does that make sense to you?
That sounds good. let me know when the paper is gonna be published so I can enjoy it.
Good luck!
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Soil Science
- Soil Analysis