Hi,
Basically I'm tired of drlling soil samples up and down just to determine soil water content. That's why I'm trying to understand if I can replace the oven dry method with a Fieldscout TDR. I guess best thing I can do is try to find some relations between them. So, as you can see in the picture, I found an R2 value of 57 after 6 months of double sampling. My question is: how can I determine if this is acceptable or not? Any suggestion, advice, comment?
Use tensiometers. Irrometers as comertial name. go to www.irrometers.com. You can find it in english
If you sample on the same soil type then salinity test may improve your correlation. In other cases drift calibration may be required, as in http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.715.4278&rep=rep1&type=pdf
TDR is OK in sandy soils but does not perform well for smaller grain sizes. I have several dozen in my lab that have been taken out of service because of their problems. Even with calibration, they will have large error. If you have the resources and permitting, neutron probes are the standard.
Dear Matteo,
You can find application notes for this topic on The Soil Water Compendium:
http://www.edaphic.com.au/soil-water-compendium/
When calibrating with the Fieldscout it is better to use the raw data option rather than the %VWC option. The raw data option gives the TDR pulse time and this will show a linear response to oven dried samples. I have done this previously with different types of soils and always found a very good R2 and linear calibration.
The %VWC on the Fieldscout, which it appears you are using, is based on a calibration for an ideal or average type of soil - in other words, who knows on what type of soil the manufacturer did the calibration! When calibrating soil moisture sensors, never use a manufacturer's VWC but always use the raw data. If the manufacturer does not, or won't provide, raw data then they are probably hiding something and best to find a better manufacturer.
Also, the preparation of your soil samples is crucial for accurate data. In the link to the Soil Water Compendium, see the application note "How to calibrate soil moisture sensors" and follow the steps therein.
Lastly, do not use a tensiometer as suggested by Carlos. This measures soil water potential and not water content and has nothing to do with calibration.
Michael
http://www.edaphic.com.au/soil-water-compendium/
I am using a Diviner 2000 Probe with PVC access tube. It worked well in light soils. I have got R2 of 81% against gravimetric samples. You can repeat readings on the same spot without any radiation hazards as it works with frequency.
Now a days using drip irrigation and other methods, It is known that most crops must be irrigated as frequently as possible. That means that tension level should not go further than 20-30 kilopascals. Texture, structure, bulk density are important but less important than tension level when the moment of a new irrigation is decided.
Dear Matteo,
I refer to this paper where thermal inertia is used to measure soil moisture. Validation was perfomed with TDR probes.
Cheers,
Frank
URL:
https://www.researchgate.net/publication/233034705_Soil_moisture_content_retrieval_based_on_apparent_thermal_inertia_for_Xinjiang_province_in_China
Article Soil moisture content retrieval based on apparent thermal in...
- Badges
- Science topic
- Similar topics
- Agriculture
- Soil Science
- Soil