AgTalk Home
AgTalk Home
Search Forums | Classifieds (54) | Skins | Language
You are logged in as a guest. ( logon | register )

How do you define a management zone?
View previous thread :: View next thread
   Forums List -> Precision TalkMessage format
 
dko_scOH
Posted 11/26/2014 10:28 (#4202485 - in reply to #4202257)
Subject: RE: How do you define a management zone?



39.48, -82.98

Beat me by three minutes, Clay! I'll blame it on the smart phone.   :-)

Another thought: I come from the grid perspective. I see a lot of variability within a soil type and would never assume that a composite soil test would accurately describe the entire zone, defined that way. What I want to do is find a better way to sample, based on historic yield/variability. The goal is to place the sample locations where they will do the best job of modelling yield when interpolated. I will likely have as many samples to send off; it is my contention that guessing is no substitute for measuring. I would probably not sample an area much less than 1 acre and would likely divide any zones over about 5 acres. Some more study will better define these limits.

Finally, the objective is to do this with machine learning algorithms, not a pencil. I want to remap these fields, not in latitude and longitude but in the dimensions that are commonly used to delineate zones. Then, I want to classify the points in such a way as to maximize information gain and re-project them in lat/lon and see if there is a coherent pattern that can be vectorized into management zones. Sounds like a fun wintertime project.

Top of the page Bottom of the page


Jump to forum :
Search this forum
Printer friendly version
E-mail a link to this thread

(Delete cookies)