Cosmic Ray Neutron Sensors do fill the gap. A single sensor is capable of measuring integrated water content in a footprint of 600m diameter, non-invasively. Follow publications of Zreda etal 2012 for instance. In 2013, they even published a mobile campaign with these sensors in order to easily cover remote sensing pixels at a much higher resolution.

But since the measurement indirectly depends on every hydrogen atom in the vicinity of the sensor, there is some calibration effort necessary, such as soil sampling or vegetation correction.

I had already posted this question that triggered an interesting discussion (67 answers). Unfortunately, the question was accidentally deleted. Therefore, I believe it would be interesting to re-start the discussion.

Actually, I would like that the persons who contributed add their main comments again as I lost all these important information . For instance, just add the technology, procedure, method you suggested for addressing my question.

I did't participate to the original discussion ... but I think that GPS measurements can provide interesting complementary data (see Larson et al. : "Use of GPS receivers as a soil moisture network for water cycle studies", 2008, GRL), but still at a small scale. And you need a network of GPS receiver in the area of interest.

The reflection of GPS signals at the ground lead to noise in the GPS measurement, and allow an estimation of the soil moisture near the surface.

The question needs clarification. What is the objective of soil moisture measurement in large areas? What is the practical utility of such data? The answers to these questions would provide clear focus to suggest techniques for soil moisture estimation.

Thanks for the clarification.

The objective is to have soil moisture measurements that could be used for studies carried out over large areas, e.g. for hydrological modelling. This means that we need techniques able to measuring soil moisture with good spatial resolution (~0.5-1 km²) and quickly over large areas (~500-1000 km²), for having also a good temporal resolution (~daily). This would allow to obtain soil moisture estimates that might be effectively used for constraining and improving hydrological modelling at catchment scale.

Remote sensing technology coupled with ground truth verification would serve the objective set for.

I think the only technology available that can meet you requirements for resolution is a spaceborne Synthetic Aperture Radar (SAR). There is considerable literature available on the retrieval of soil moisture from SAR. Also noteworthy is the SMAP (Soil Moisture Active Passive) NASA mission, planned to be launched in 2014, that combines a SAR and microwave radiometers to retrieve soil moisture. Other remote sensing (such as scatterometers) do not provide retrievals at the resolution you want. Hope this helps.

It is worth practically to wait until 2014 as the data would be more realistic and practical on soil moisture status of a given area with SMAP.

What about using a Thornthwaite-Mather type model to generate a temporal estimate of soil moisture for your study area? The model uses a series of iterative calculations to determine volume of water available for runoff and for infiltration over a given time interval. Actual data inputs include topography, precipitation, land use, ground cover (including plant typing), and soil type. Most of this information can be obtained or estimated using readily available data sources without the need for additional investigative expenditure. Potential and Actual Evapo-Transpiration, PET and AET, respectively; can be estimated but model results are more reliable with site specific values for AET. As part of the model process the comparison of AET and PET generates soil moisture demand; ie: when soil moisture is not limiting AET=PET. Soil moisture demand is based upon the maximum soil-moisture holding capacity. Using an estimate of porosity for a given soil type you could then extrapolate the percentage of available porosity containing water ie: soil moisture. You can even use this type of model to gain insight into other factors such as how much precipitation must fall before any runoff is generated, what the rate of aquifer recharge is and the maximum sustainable yield of groundwater resources, potential impacts of changing land use, impacts of irrigation on the water balance, etc. Hope this is helpful.

Yes, likely the integration of in situ and satellite soil moisture measurements with modelling approaches should be adopted to meet the target ...

Is it possible to get back what has been invested in terms of currency on the study of water resources of the areas when these advanced technologies are used to derive the data. Economics of science and technology is vital for the growth and development of the world.

Georgi Stoimenov

You can use canopy temperature of the areas. Calibrate it in one small area, and use the different of canopy temperature. For details you can ask me by E-mail - [email protected]

It depends on the scale of interest, for sure. Anyway in many cases geophysical mapping by integrating georadar at radio-frequencies or micro-wave and time domain reflectometry (TDR) could provide an useful tool for mapping soil moisture. Network of TDR are adopted to monitor time-lapse changes of soil moisture, covering large areas (hectares or decades of hectares) for agricoltural purposes; example of georadar (GPR) investigation for soil moisture evaluation are widely reported in Journal of Applied Geophysics, Water Research and Journal of hydrology, for instance. I recently published on Journal of Applied Geophysics (20112) a paper on the integration of TDR and GPR for snow water equivalent estimate on a scale of small basin.

We measure the soil moisture by gypsum blocks, designed In our institute and compare this data with measuring canopy temperature by IRT(infrared thermometer), correlation is more 0.95 this is enough for agricultural practice.

Canopy temperature is related to the life or dead of trees and plants, and only indirectly to the soil moisture; I am wondering how is it possible to use as a good estimator of the soil moisture: it is really nice if you have found robust relationships.

Of course to determine the soil moisture by IRT, you have to calibrate it - this depend of plant, of soil surface.

Cosmic Ray Neutron Sensors do fill the gap. A single sensor is capable of measuring integrated water content in a footprint of 600m diameter, non-invasively. Follow publications of Zreda etal 2012 for instance. In 2013, they even published a mobile campaign with these sensors in order to easily cover remote sensing pixels at a much higher resolution.

But since the measurement indirectly depends on every hydrogen atom in the vicinity of the sensor, there is some calibration effort necessary, such as soil sampling or vegetation correction.

@ Martin Schrön

The India is not yet covered under the probes?

http://cosmos.hwr.arizona.edu/

Any suggestion for estimating soil moisture data ?

I suggest all of you interested to obtain large scale soil moisture data these two recent studies:

1) Ochsner, T., M. Cosh, R. Cuenca, W. Dorigo, C. Draper, Y. Hagimoto, Y. Kerr, K.M. Larson, E.G. Njoku,E.E. Small, and M. Zreda, The state of the art in large-scale soil moisture monitoring, Soil Science Society of America Journal, Vol. 77(6), 1888-1919, doi:10.2136/sssaj2013.03.0093. http://xenon.colorado.edu/portal/publications/Ochsner_etal_2013.pdf

2) Vereecken, H., J.A. Huisman, Y. Pachepsky, C. Montzka, J. van der Kruk, H. Bogena, L. Weihermüller, M. Herbst, G. Martinez, J. Vanderborght (2014) On the spatio-temporal dynamics of soil moisture at the field scale, Journal of Hydrology, in press, http://dx.doi.org/10.1016/j.jhydrol.2013.11.061.

It would depend also on the spatial variability of soil properties and meteorological conditions. Where the soil is largely homogeous a smaller representative volume would suffice than in highly heterogeneous regions.

Gor God sake, please do not attempt to estimate soil moisture to cover large areas. It is the most futile exercise besides wastage of time, energy and human resources. Soil moisture is a dynamic resource never remains constant.

I strongly support the use of Cosmic Rays Neutron Probes (COSMOS). USA and UK are building large country-wide networks of such instruments with the aim of ground-truthing remotely sensed data to drive irrigation over large agricultural areas (mostly homogeneus) and to calculate soil water content to help forecasting flood risk. Few of such devices are used also elsewhere. We got some deploied in the field in Italy and in India for an EU funded FP7 project.

Agronomic soil water balance models can help: once calibrated on measured spots, they can be applied to wide areas where rainfall data and soil and crop maps are available. We do this in Emilia-Romagna on a routine basis, see for instance our agrometeorogical bulletins here: http://www.arpa.emr.it/sim/?agrometeo/bollettino_settimana

Cosmic ray neutron probe provides soil moisture estimates at an intermediate scale (34 hectares). 

I'd like to comment on the last answer and to update my own answer regarding to the measurement scale. Recently, the representative area of the Cosmic-Ray neutron probe has been revisited. It is not 34 ha and not 600m diameter. Instead it is various from 6 to 18 ha depending on moisture conditions and the average is highly sensitive to the closest tens of meters. See Köhli et al. 2015 for details: http://onlinelibrary.wiley.com/doi/10.1002/2015WR017169/full

For large areas soil moisture will be very different, average evaluation  can be done measuring the plant water stress by infrared thermometer.