NPK Soil nutrients measurement sensor for Mesh network

Hi,

I am aware of a similar topic in another thread which is now closed and seen the good and bad responses in that thread. There seems no viable solutions out there yet or not created, hence I wanted to recheck if there is anything out there.

For those, who think this is unnecessary, overkill, pay for lab test etc. I politely request them to not comment on this thread as I understand you have not solution to post or positive suggestions to answer the request please. And for the record I do have a use case for it.

I see a cost effective NPK (or multi nutrient) soil sensor to be a game changer is lot of places in the world especially developing and poorer countries. This could help farmers optimize the yields and take actions before it is too late.

Now, my use case is for a multi crop/tree plantation that has high value plantation that needs monitoring. So a Mesh network with Multi Nutrient soil sensor will be fantastic. I have look at Particle.io and need to find a soil sensor.

I have look at the suggestion in other posts which does not resolve my problem. I have also downloaded the research papers provided on this topic but there is no practical way to get this implemented i.e. a sensor which can provide values. I also saw the colour detection but does fall short of how we will detect under the soil and so on.

Since this will have huge benefits for farming I am resurrecting this topic so see if there are any new developments in this area.

Thanks and appreciate your positive inputs.

Not a hardware suggestion, but wonder if you have discussed this with soil scientists at an agriculture college/university, such as Oregon State University?

Paul

I see a cost effective NPK (or multi nutrient) soil sensor to be a game changer is lot of places in the world especially developing and poorer countries.

Of course it would be, and there is an unbelievable amount of money to be made, if someone could just come up with one.

Unfortunately, there is nothing simple about measuring just the "N" (let alone the P and the K), as you will learn if you ask a soil chemist.

Paul_KD7HB:
Not a hardware suggestion, but wonder if you have discussed this with soil scientists at an agriculture college/university, such as Oregon State University?

Paul

Paul_KD7HB:
Not a hardware suggestion, but wonder if you have discussed this with soil scientists at an agriculture college/university, such as Oregon State University?

Paul

I have asked around but have not got any response yet. I may have to pursue this harder...hmm

jremington:
Of course it would be, and there is an unbelievable amount of money to be made, if someone could just come up with one.

Unfortunately, there is nothing simple about measuring just the "N" (let alone the P and the K), as you will learn if you ask a soil chemist.

It is not impossible and it is not required to be 3 decimal places precise either for most agricultural projects. Teralytic has a wifi probe costing $1000 each which does it.

I am looking for the mechanics of measuring it then we can talk about manufacturing cost and selling price etc. It seems like for the sacristy of the measuring technique the pricing is sky high and looking at the benefits the wider farming community can benefit from such measurement, it is worth pursuing this to the end imo.

As you know, Teralytic is a relatively new startup, which claims to have developed the first ever wireless NPK sensor. Let us know if you find any reliable, independent scientific reviews of the NPK probe performance under typical use cases, as a quick search turned up only advertising and ag news stories.

I suspect that review won’t happen soon, because Teralytic sensor systems are available only on lease. You are required to upload all your data to Teralytic servers, it is forbidden to send the raw data anywhere else and finally, Teralytic can modify your data at will, even before you see it.

The Device collects Data when properly installed and uploads this Data to Teralytic’s servers. During the Term, you consent to allow Teralytic to receive and store your Data on the Teralytic servers and clean your Data by removing perceived errors and omissions. You may not direct your Device to transmit Data to any other location other than Teralytic. Your ability to access, share, and delete Data is explained in Teralytic’s Data Use Agreement.

https://teralytic.com/device-lease-data-upload-agreement.html

This does not pass the briefest sniff test.

I see there are commercial drone services that do video recording and mapping of farmers fields, so they can look at the crop and see what additives need to be added to the soil. No need to put 1,000 sensors in a 80 acre field and wait for rain to provide the liquid so the measurements can take place.

Paul

Update: I sent the info about Teralytic to a chemist friend who works on developing biological sensors. Here are his very interesting and enlightening comments.

Thanks for sending this. There's lots of crazy stuff out there on nutrient monitoring. Most of them just use off-the-shelf water/pH/salinity sensors, which have no selectivity for NPK. They just try to correlate conductance measurements to NPK-levels, which might work for a home garden, but are not really useful. The data is very valuable to the companies, though. When soil models get good enough, the sensors won't be necessary anymore for prescribing soil amendments. That license agreement is particularly troublesome.

The Climate Corporation is trying to develop a digital farmer's almanac, and they still need good sensors for soil NPK. They probably have good enough sensors for aqueous measurements (ours had pretty good nitrate-N selectivity). They definitely are a "big brother" that is trying to collect a ton of growers' data for their models. I hope they are doing that in a more forthright way through their sensor networks and customers.

I still think there's a good market for a home-use NPK sensor for gardens.

jremington:
Update: I sent the info about Teralytic to a chemist friend who works on developing biological sensors. Here are his very interesting and enlightening comments.

Thanks, that is informative. Yes, I do not trust any of these closed systems like Teralytic on this very controversial topic.

I found this UAV based mapping and interesting thing.

I have seen in my search, there is claim of using a camera to measure RGB values. I do not know how reliable any of these are but intend to see if RGB can be used here is a link .

And finally, I think these guys seem to have cracked it :slight_smile:
For less than $100 this will be great kit but manual.
http://deshpande.mit.edu/portfolio/project/site-soil-nutrient-analysis-system-smallholder-farmers 

I will continue looking to find one that can be in the soil using a mesh network (e.g. Learn how to build In-Plants, a mesh-connected soil monitoring system – Particle Blog), so the soil health can be monitored real time and adjustments can be done more quickly rather than doing it manually.

Paul_KD7HB:
I see there are commercial drone services that do video recording and mapping of farmers fields, so they can look at the crop and see what additives need to be added to the soil. No need to put 1,000 sensors in a 80 acre field and wait for rain to provide the liquid so the measurements can take place.

Paul

My requirement is different, I need a sensor for mesh network. Once in a few years scanning the field may be fine for other requirements not for mine. I have high value trees which need to be optimally fed to reach yield.

maxpayne_:
My requirement is different, I need a sensor for mesh network. Once in a few years scanning the field may be fine for other requirements not for mine. I have high value trees which need to be optimally fed to reach yield.

Then your requirements are more extensive. You need to layer the sensors vertically as well as horizontally and have them in the ground permanently.

How large of a root mass will your trees develop? How many sensors do you think you will need per tree?

Paul

Paul_KD7HB:
Then your requirements are more extensive. You need to layer the sensors vertically as well as horizontally and have them in the ground permanently.

How large of a root mass will your trees develop? How many sensors do you think you will need per tree?

Paul

I just need one sensor per tree near its root to monitor the soil health.

The ancient rule-of-thumb is the radius of roots is equal to the height of the tree. What kind of tree has one root?

Paul

This is a huge research area and is complicated by the need to understand bioavailable nutrient forms vs. recalcitrant forms. In the soils I work with, most of the P is not bioavailable so simply measuring the concentration is not very useful. Nitrogen is subject to the very complicated nitrogen cycle in soils and the form of nitrogen in the fertilizer. Most farmers here apply nitrogen as urea because it it inexpensive but it needs to undergo transformation in the soil before uptake by the trees, pasture, or crops. It is also lost from the soil as N2O or ammonia.

If you want to do it through a mesh network, maybe your best bet is to use fixed sensors similar to that used on the drones to detect nutrient deficiencies in the tree leaves.

Good luck!

Paul_KD7HB:
The ancient rule-of-thumb is the radius of roots is equal to the height of the tree. What kind of tree has one root?

Paul

I am not sure you understand. My requirement is to measure the soil nutrients at the root area of the tree so my irrigation system can mix the correct nutrients balance to improve the growth/yield. If you take for example Teak tree, this may help get the optimized growth and yield in 10 years