http://www.courtyard-partnership.co.uk/
http://www.soilquest.co.uk/
http://www.soyl.com/
Any others I have missed other than DIY?
http://www.soilquest.co.uk/
http://www.soyl.com/
Any others I have missed other than DIY?
I think you are correct here Clive. The sales pitch and the coloured maps make it look extremely accurate, as if there is a distinct dividing line between this patch and that. It's all extrapolation of data , nothing more. If you run 10 P, K, pH samples in a 10ha field (each costing £10 each), you're still working based on patches that are 100m x 100m, that's a pretty big pixel. You could divide it down to 100 samples, but the sampling cost would then be £900 for the field, excluding the labour involved in sampling.
Then, even if you did this greater sampling accuracy.....you're then applying fertiliser in 24m wide lines up and down the field, a boom width spanning two and a half pixels. It's like painting the Mona Lisa using a paint roller!
Yes there is a sample taken from every hectare, but it is made up from 20 or so other sub-samples from within that hectare.
I've said it before, when we first started using it the saving were huge, it paid for the sampling, the GPS box and the extra to go to a VR ready spreader
How accurate the sub sampling is depends on the dilligence of the sampler. I caught one with a spade right beside his quad bike not doing the sub sampling at all, just filling the bag with one sample only. I had been watching him do this on a number of occasions. He was thrown out & that firm had to resample the fields this individual had messed up.
Junk information in, junk information out!
Rob - How did you quantify your saving? Product saved vs blanket doses?
Thats not quite how the sampling works,
Yes there is a sample taken from every hectare, but it is made up from 20 or so other sub-samples from within that hectare.
I've said it before, when we first started using it the saving were huge, it paid for the sampling, the GPS box and the extra to go to a VR ready spreader
Maybe i'm getting seduced by the sales hype but i think the whole principle of VR inputs is sound.
We only do VR P+K at the moment but i really want to move onto VR N and seed.
There has has to be an element of inaccuracy in everything that we have to accept, like @Clive said, for a true VR N calculation you'll have to do a live tissue test on EVERY plant and for the correct another of N onto that plant to be truly VR, there has to be some averaging done.
Savings are easy - if I apply no N this year I will save a load of money - they doesn't make it the right thing to do though
Composite samples are worse than single ones as that are not actually a sample if any point in the field
Zoning is the best way if you want to vra IMO but even that is limited by resolution and the pretty contor maps are no more than a mathematical equation
Composite samples give an average for that ha grid section. It's back to how much detail you require, what the extra sampling costs & how precise your spreaders/sprayers/drills can apply the different rates.
Zoning based on conductivity scans every 24m backed up by ground truthing & historical yield/canopy maps is better than the basic 1 ha grid system IMO.
Composite samples give an average for that ha grid section. It's back to how much detail you require, what the extra sampling costs & how precise your spreaders/sprayers/drills can apply the different rates.
Zoning based on conductivity scans every 24m backed up by ground truthing & historical yield/canopy maps is better than the basic 1 ha grid system IMO.
But surely there are more variables to yield than P and KI don't like the idea of mixing samples - you need point samples to look at what you have
zoning is an improvement on the grid method but all are flawed by the gross over simplification of the model they base application upon - non are looking at nutrient ratios and just apply to maintain or correct indices - 21st century application methods based on 100 year old soil nutrition models !
VRA application of base nutrition focuses on P, K Mg and Ph - CA/Mg ratios would be a better starting point in most cases. I was advised on the basis of grid samples year ago to apply VRA lime (calcium) - I know know that some of those fields were too high in Ca and lime was the last thing they needed - ph is not a measure of Ca its a measure of H in the sample !
............the entire concept of PF is oversimplified to make it both possible and therefore a saleable product, look closer and its no better than your grandfather
I maintain a proper single sample per field will tell you more than many basic ones and set you on the right road to fixing nutrient ratios and be a lot cheaper to implement
N sensing is another thing that most who use it don't really understand IMO as well ! it measures chlorophyll content of plants NOT nitrogen levels but varies N rates based on information it collects - there are many reasons why chlorophyll content varies between plants with just one of them being N levels ..................again a gross over simplification of something much more complex but with a liberal sprinkling of great marketing !
all that said there are ways to use the tech in a useful way ! use yield maps to replace P& K off take for example to ensure you maintain levels once you know they are correct. I like what I have seen and done with EC scanning so far - I think it might help targeted samples (full samples) to maybe break filed down into zones that then with a good application model might be a better way forward
I have been using various PF and VRA for over 10 years now - as I said I think its a very interesting area and these things have to start somewhere but i'm not going to kid myself it saves or makes me a penny extra (yet ) ! The technology works just fine, the agronomy behind the models sucks though frankly in most cases !
But surely there are more variables to yield than P and K
I maintain a proper single sample per field will tell you more than many basic ones and set you on the right road to fixing nutrient ratios and be a lot cheaper to implement
What do you define as a proper sample? Albrecht?
Is anyone seeing the benefits of variable rate drilling? Everyone seems to have a different opinion on this and I am yet to decide if we are doing the right thing?!
The results come from looking at the aggregated yield data of Rhiza farmers using VRA seed compared to Rhiza farmers who are not. The farmers are in control of what rates they use and as we work with farmers all over the UK, these will vary considerably depending on location and soil types.How many rates we're used in your example?
We have not looked at grain quality, just yield. In the WW yield maps above the farmer cut back the seed rate in the deep valley running through the middle of the field where lodging had been a problem and increased rates on the shallow, stony bank to the west of the field where poor establishment had been an issue.Do you have any results for variations in grain quality before and after VR seed? My VR seed fields seem more homogenous in ear numbers but I have yet to see a yield boost vs a flat rate control field next door.