ecowitt sensor autodiscovery and soilMoist units

[vince]

Gary - the autodiscovery the driver does is pretty magical.   I added two WH31 indoor temp/hum sensors and one WH51 soil moisture sensor and did 'nothing' in the sensor_map and weewx picked them up and added their values to the db right away as extraTemp1+2 and soilMost1.

Very cool.

One question - the WH51 reports as "34 cb" for (presumably) a 34 percent reading.   Where does the 'cb' unit come from and what does it mean ?

I'm just grabbing $current.soilMoist1 in the skin if that helps any.  The value matches what I'm seeing in the db.

Any ideas ?

[gjr80]

cb stands for centibar and is (I believe) a measure of the soil water tension or the level of moisture held in the soil. Davis soil moisture sensors read in centibars. Strictly speaking I guess that fields that contain Ecowitt soil moisture data should be moved from group_moisture to group_percent, though given there is no real need for inter-unit conversion (à la temperature or speed) and they use the same default format there is not much to be gained. True, the default unit label is different but that can be easily overridden.

Gary

[Greg Troxel]

gjr80 <gjrod...@gmail.com> writes:

> *cb* stands for *centibar* and is (I believe) a measure of the soil water

> tension or the level of moisture held in the soil. Davis soil moisture
> sensors read in centibars. Strictly speaking I guess that fields that

> contain Ecowitt soil moisture data should be moved from *group_moisture* to
> *group_percent*, though given there is no real need for inter-unit

> conversion (à la temperature or speed) and they use the same default format
> there is not much to be gained. True, the default unit label is different
> but that can be easily overridden.

Soil moisture is very complicated. It sounds like Davis sensors are on
the pro end of things with an absolute calibration.

I have a WH51, not hooked up to weewx, and find that the % readings are
useful over time to compare to other readings, but don't seem to have
absolute calibration.

When digging into this there are two moisture levels of primary interest
for any particular soil type and ay particular plant:

wilting level: the level below which the plant is unhappy due to dry

saturation: the level you get to when the dirt/plant is in a long
soaking rain, more or less

For my plant, those were about 22% and 40% with the default calibration
of the WH51 receiver. One can adjust those to make that range read 0 to
100, but I think that's only adjusting the display.


I do think it would be good that Ecowitt % values not be mislabled as
centibar, if my assumption they aren't centibar is correct.



Some reading for those with a lot of Copious Spare Time. I'm not
qualified to judge this, but it comes across as being from people that
know what they are doing.

https://www.metergroup.com/environment/articles/what-is-soil-moisture-science-behind-the-measurement/

[Greg Troxel]

The quick summary of that tl;dr page is that there are two separate
concepts:

Volumetric Water Content (VWC), which is the fraction of any volume of
soil that is water. This is dimensionless, and typically in %.

Water potential, reported as pressure

The wilting level of a plant depends on water potential, and the
relationship between VWC and water potential is highly dependent on soil
type.

Field capacity is the moisture after soaking and draining (but not
evaporation or plant use, I think).

With my WH51, it seems reasonably clear that it's trying to measure VMC,
and my experimentally determined 22%/40% makes sense.

The Davis sensor is very clearly water potential with a negative sign,
termed "suction" and in pressure units. The spec sheet says:

https://www.davisinstruments.com/product_documents/weather/spec_sheets/6440_SS.pdf
https://irrometer.com/sensors.html

The Soil Moisture/Temperature Station converts the electrical
resistance reading from the sensor into a calibrated reading of
centibars of soil water suction with a range from 0 (wettest) to 200
(driest) centibars.


The metergroup page gives an example of -0.3 Mpa, or -300 000 Pa.
Midrange in Davis is 100 cb which is 1 bar, and mulitplying by 100 000
Pa/bar gets -100 000 Pa. This doesn't quite line up as the example is
1.5x the top range, but it could be that typical soil for a
non-irrigated tree is much drier than for crops.


Architecturally, weewx should have separate column names with associated
units for VWC and either suction or water potential, deciding on
positive/negative. That's my theory right now and I'm sticking to it
for at least an hour!