Where Lewis and Clark finished the rum | Well the neat/frustrating thing about prec ag is that there is no lack of data.
Your thought about lodging and Cerone seems logical to me.
Conductivity - we have been in a 10 year drought here, so have had trouble getting good VERIS data. Some fields have been collected 3 times. We try to do with uniform moisture, and that has been our focus, so we can see mainly soil differences, rather than moisture differences. But once we have a decent conductivity map with "uniform" moisture, and then have the opportunity to get a differential water map, it might be good idea to give conductivity another go - sure couldn't hurt to see the differences. Still waiting for that rain that causes runoff event though.
I've been working on one field trying to do zones using several methods, including a couple of the satellite based guys. In fact, I started PA with satellite/aerial images. (www.UMAC.org has been very helpful - you should be able to find historic LS, ASTER< MODUS images at their site)
In the one particular field of about 160 acres, I used a June 26 Landsat image and the yield map from harvest about a month later. I gridded the data point data and did a regression of yield vs NDVI and got only about a 0.2 to 0.3 R squared - was expecting more like a 0.7 or so. Need to look at that again to make sure methodology was appropriate, but a zone map done independently by one of the satellite guys didn't "look" anything like my zone map based upon VERIS, yield, and 73 test holes - hmmm. I now have this year's yield and protein map, as well as several other years worth of yield data, so have some winter work to do, but have been trying to get some fall work done, so can work around the campfire later. Plan to do some yield stability analysis, and also use the protein data to help out.
Re protein - we are running a couple Zeltex units so we get a protein data point about every 6 seconds, and man does it vary quite a bit. Just as yield monitors have shown us that yields ofter vary bay a factor of 3 to even 4 (15 to 60 is not uncommon), proteins vary by a factor of 2 within a field (8 to 16 isn't uncommon). In the field mentioned above, we also collected over bin samples by hand during harvest, so we should have a nice comparison between the Zeltex and the Infratech we used to analyze the hand samples.
I'm trying to move my screenshots to www.srvscience.org, but right now some protein histogram examples can be seen at http://www.3rivers.net/~chuckm/index.htm Oh yes, also look at the GreenSeeker screenshots there. Should have some comparisons between nTech's GreenSeeker and Holland Scientific's Crop Circle scanner over the winter too. And in fact, Kyle Holland showed me an interesting Crop Circle vs OM - very good agreement - could be interesting.
More to come. |
elevation issues
