Hello everyone,
I got the device and the calibration went smoothly. But I want to measure soil carbon and was wondering if there is a link or an article that shows the step by step process to measure soil carbon using reflectometer? Also, what kind of data are we getting directly from reflectometer (unit)? Can we use the data to calculate how much carbon is available in lets say a square meter of a land?
Thank you very much.
Hi Binay,
That’s great that you got through the calibration procedure!
Our soil carbon model is still in development, but we are hoping to have something user-facing within the next couple of months. That said, we’d love to hear more about your project. It’s possible you could contribute to the development of our model as we continue to optimize it.
Without a user-facing estimate survey up, such as what we have for nutrient density estimates, you will only get spectral data as immediate feedback. This data cannot be used to directly calculate carbon levels, but could be correlated to lab-based carbon analysis as a way to build out our model.
If you’re interested in this collaboration, we can certainly add you to our SOC Pilot group in SurveyStack (our data collection software), so that you can contribute data to the relevant survey.
Please let me know what you think, and have a great Monday!
Best,
Nat
Hey nat, thank you for the reply.
I can measure and compare reflectometer data to lab based carbon analysis. I was wondering what is the procedure to measure the soil. Should I just put air dried soil in the petri dish? Which app would I use to measure the soil carbon?
For nutrient density, I first go to BFA survey stack though web browser and then open the surveystack app in the smartphone. What would be the procedure for soil carbon?
Thank you.
Hi Binay,
I can add you to the SOC Pilot group within the SurveyStack app. Accessing the app is quite similar to the BFA SurveyStack, but this time just go to surveystack.io in your Chrome browser (instead of bionutrient.surveystack.io).
What email will you be using for your account? This will allow me to add you to the group (feel free to dm if you don’t want to share on a public thread). Once added to the SOC Pilot Group, there is a survey called “Reflectometer Pilot Project Scans” that gives instructions about drying (air drying is fine) and homogenizing soil samples before scanning.
As for your samples, how many will you be scanning? And do you already have lab-based carbon estimates for each of your samples?
Thanks,
Nat
My email address is binay_sangat(at)yahoo(dot)com. I have also sent an email to Our sci hardware support using this email.
I dont know the sample numbers yet, may be 10-20 for now. I dont have the estimates but will be analysing both lab based and reflectometer samples simultaneously.
Also, I am located In Japan so will be analysing soil of Japan. Will the model that you are currently developing usable globally?
Thank you.
Edit
I found the protocol for measuring soil in groups-socpilot. I measured my soil three times and I got results for different wavelength. I am sorry but I don’t know how to interpret these results. There are 10 wavelength with their results. how can I compare these value with lab based combustion method? Am I suppose to average the three readings? Which wavelength am I suppose to compare with lab based method?
Thank you
Hi Binay,
You cannot directly get a carbon reading from the spectral results. That’s why we’re discussing having you participate in the model development at this time. Once our model is better optimized, we will release a user-ready survey similar to the Nutrient Density Estimate survey. That will have a more immediate carbon estimate for users to view as they complete their tests.
The purpose of these models is to correlate spectral data to lab-based carbon results. However, the process is more complicated than just averaging spectral results and trying to gain an understanding of carbon content from that.
Please let me know if you have further questions, and my apologies for the confusion!
Best,
Nat
Hi Binay,
I also want to follow up about your location in Japan.
The model that we are deploying at this time will be based on soil samples in the USA, and therefore would need to be checked against soils in Japan to see if it is useable.
For this reason, it may make more sense for you to look into creating your own local prediction model. Depending on the area and diversity in soils conditions that you are working with, that could require 10’s or 100’s of samples to develop. This is a question that we are interested in answering with the participants of our pilot project this year.
The SurveyStack app is flexible and allows users to create their own scripts using javascript. So, anyone with js experience could deploy their own prediction model. If you do not have the expertise to develop or deploy your own model, we could help.
All the best,
Nat
Thank you very much for your help. We will try to come up with a model.
Binay