Soil thermal properties

A Do-link ISMC working group project

ISMC Leads

Anne Verhoef (University of Reading, United Kingdom, a.verhoef@reading.ac.uk)

Yijian Zeng (University of Twente, Netherlands, y.zeng@utwente.nl)

Co-leads

Hailong He (Northwest A&F University, Yangling, China, hailong.he@nwsuaf.edu.cn)

Nan Wei (Sun Yat‐sen University, Guangzhou, China, wein6@mail.sysu.edu.cn)

Yongjiu Dai  (Sun Yat‐sen University, Guangzhou, China, daiyj6@mail.sysu.edu.cn)

Background

There is a longstanding tradition in soil science in developing and applying pedotransfer functions (PTFs) for the calculation of hydraulic properties; (e.g. Clapp and Hornberger, 1978; Cosby et al., 1986, Rawls and Pachepsky, 2004; Vereecken et al., 2010; Van Looy et al, 2017; Dai et al., 2019a). These functions use soil properties that are relatively simple to obtain such as texture, organic matter and bulk density in order to estimate parameters in mathematical functions (e.g. Brooks and Corey, 1964; Van Genuchten, 1980), that approximate the soil hydraulic properties (water retention characteristic, WRC, and the hydraulic conductivity curve, HCC), as these properties are pivotal in describing the water flow in soil and its availability for vegetation via root water uptake.

Besides applications in soil sciences, ecology, engineering and hydrology, hydraulic functions and their PTFs are used extensively to derive hydraulic parameters for land surface models (LSMs); LSMs are embedded in global climate models to enable global coverage and calculation of soil- and land surface related processes, pertaining to the water-, energy-, and carbon balance.

Information on soil texture and porosity is also used to calculate key parameters for equations that are used to estimate thermal soil properties (thermal conductivity and heat capacity) in LSMs, such as those developed by Johansen (1975) and Lu et al. (2007), derivatives thereof (see He et al., 2017, He et al., 2020a), or alternatives (see e.g. Lehmann et al., 2003; Ghanbarian and Daigle (2016), Dai et al., 2019b, He et al., 2020b). Although the term PTF is rarely used in this context, in this ISMC working group project these thermal parameters will be evaluated in a similar context. We will consider thermal property equations and their PTFs, but aim ideally to develop a harmonized underlying theory, that will yield both thermal and hydraulic parameters.

Often PTFs have been developed for specific regions (e.g. USA (Cosby et al., 1986)/Europe (Wosten et al. 1999) and locations, questioning the validity of their application beyond the region of development. For example, Marthews et al. (2014) developed hydraulic PTFs that are more suitable for tropical South American soils. Recently, Montzka et al. (2017) produced a global high‐resolution data sets of soil hydraulic parameters based on SoilGrids (Hengl et al., 2017), whereas Dai et al. (2019b) produced a global high‐resolution data sets of soil hydraulic and thermal parameters based on the Global Soil Dataset for Earth System Models (GSDE) and SoilGrids soil composition databases. These parameter sets were specifically designed for use in land surface models.  In Dai et al. (2019b) thermal parameters included the volumetric heat capacity of soil solids in a unit soil volume and the saturated (frozen and unfrozen) and dry thermal conductivities.

Soil thermal conductivity strongly depends on the thermal conductivity of the soil solids, TC_s [lamda]. All thermal conductivity models require this information, and most models simply use an estimate of quartz content (that has a TC_s of ~ 8.8 W m^-1 K^-1) for the coarse grain fraction, while assuming that the rest of the solid soil fraction is composed of finer ‘other minerals’, with a lumped TC_s of 2.0-2.2 W m^-1 K^-1. This approach assumes that all coarse material is quartz, while that may not necessarily be the case. Also, some of the ‘other minerals’ often have TC_s values that are considerably higher or lower than 2 (see also, He et al., 2020c).

Aims

Based on the above premises, the overarching aims of this ISMC working group project are

• to collate and generate global datasets of measured thermal property data (laboratory and field), conditions during the experiments (including soil moisture content and temperature, and ideally matric potential), and sample/field soil properties (texture, OM, mineralogy (if available), stoniness);
• to collate and test (using measured thermal property data, as mentioned above) existing, and design and test improved equations of thermal soil properties, that can be used in land surface models, at field to global scales;
• to link thermal theories with hydraulic theories, and to move away from empirical approaches where possible;
• to generate global datasets of parameters required in existing and proposed equations, based on soil texture, OM, as well as mineralogy and rock content, or proxies thereof ;
• to generate datasets of field-site driving data and thermal regime verification variables (soil temperature, soil moisture/matric potential, soil heat flux)  for testing of the equations at the field-scale (this includes FLUXNET-style sites, where these data are available).

Specific Tasks

Specific tasks to achieve these aims are

• compile combined datasets of thermal properties, soil physical properties (such as texture, pore size distribution, porosity/bulk density, SOM, stoniness), as well as soil hydraulic properties and mineralogical information (or its proxies), if available, in a database. A very good starting point is the dataset already generated by Hailong He, that can be cross-checked with the list of papers already compiled by Wei and Dai, and others found by other members of the WG. Ideally water retention curve data together with thermal conductivity data are available for the same samples;
• Calculate bulk TC_s from methods proposed in He et al. (2020 c) with or without explicit inclusion of mineralogy information;
• Collate code of existing equations (e.g. as generated by He, Dai, Verhoef and no doubt many others) to describe the dependence of soil thermal conductivity on soil moisture content or matric potential, to compare with measured thermal properties and propose a sub-set of best-performing equations for testing;
• These efforts ideally include those methods that have used machine learning/AI to improve soil thermal conductivity equations and/or their parameters (e.g. Rivzi et al. (2019, 2020a,b), Wen et al., (2020) and Zou et al (2019);
• Sensitivity and verification analyses to test the effect of the different parameterisations on soil heat- and water flow via implementation of all equations in an independent soil physical program such as Hydrus or SWAP, rather than independently in a range of LSMs;
• To compose global datasets of the parameters required for the new equations;

Considerations

• What to do with organic soils and what about rock/gravel content? Currently, there is no universal model for the consideration of OM, and some models have used erroneous approaches. Effects of rock or gravel may be challenging to be considered, because barely any data is avaiable and only steady-state methods (e.g., guarded hot plate) can make such measurements. However, the steady-state methods give errors in measurements of unsaturated soils due to the water redistribution driven by temperature gradients, and phase change (see He et al., 2018)
• Throughout the procedures we think about a unifying theory (see e.g. Lu and Dong, 2015; He et al., 2020b) that takes into account (in no particular order) specific surface area, adsorption isotherms, corner- and thick-film flow, bound water etc. etc. and how these are affected by the particle (clay-type/roughness) and ‘pore properties’, and therefore affect thermal/hydraulic properties
• Alternative/parallel approaches will consider getting ‘in-situ’ field-level data of thermal conductivity, diffusivity or thermal inertia, either directly measured using K-type needle probes or indirectly from remote sensing (see e.g. Verhoef (2004), Verhoef et al., 2012); see e.g. Montzka et al. 2011, for hydraulic data).
• We will also consider inverse modelling using measured soil temperature, moisture and soil heat flux data

References

Best, M., Pryor, M., Clark, D., Rooney, G., Essery, R., Ménard, C., Edwards, J., Hendry, M., Porson, A., Gedney, N., Mercado, L., Sitch, S., Blyth, E., Boucher, O., Cox, P., Grim- mond, C., Harding, R., 2011. The Joint UK Land Environment Simulator (JULES), Model description – Part 1: Energy and water fluxes. Geosci. Model Dev. Discuss. 4, 595–640.

Brooks, R. H. and Corey, A. T. (1964) Hydraulic properties of porous media, Colo. St. Hydrol. Papers, 3, 1–30.

Clapp, R. B. and G. M. Hornberger, (1978), "Empirical Equations for Some Soil Hydraulic Properties," Water Resources Research, 14: 601-604.

Cosby, B.J., Hornberger, G.M., Clapp, R.B., Ginn, T.R., 1984. A statistical exploration of the relationships of soil moisture characteristics to the physical characteristics of soils. Water Resour. Res., 20, 682-690.

Dai, Y., Xin, Q., Wei, N., Zhang, Y., Shangguan, W., Yuan, H., et al. ( 2019a). A global high‐resolution data set of soil hydraulic and thermal properties for land surface modeling. Journal of Advances in Modeling Earth Systems, 11, 2996– 3023. https://doi.org/10.1029/2019MS001784

Dai, Y., Wei, N., Yuan, H., Zhang, S., Shangguan, W., Liu, S., et al. ( 2019b). Evaluation of soil thermal conductivity schemes for use in land surface modeling. Journal of Advances in Modeling Earth Systems, 11, 3454– 3473. https://doi.org/10.1029/2019MS001723

Ghanbarian, B., and H. Daigle (2016), Thermal conductivity in porous media: Percolation-based effective-medium approximation, Water Resour. Res., 52, 295–314, doi:10.1002/2015WR017236.

He, H., Zhao, Y. , Dyck, M., Si, B., Jin, H., and Lv,  J. et al.,  (2017) A modified normalized model for predicting effective soil thermal conductivity, Acta Geotech. 12 1281–1300

He, H., Dyck, M.F., Horton, R., Ren, T., Bristow, K.L., Lv, J., Si, B., 2018. Development and application of the heat pulse method for soil physical measurements. Rev. Geophys., 56(4): 567-620. DOI:10.1029/2017rg000584

He, H,  Noborio, K,  Johansen, Ø,  Dyck, MF,  Lv, J.  (2020a) Normalized concept for modelling effective soil thermal conductivity from dryness to saturation. Eur J Soil Sci.  ; 71: 27– 43. https://doi.org/10.1111/ejss.12820

He, H., Dyck, M., Lv, J., (2020b). A new model for predicting soil thermal conductivity from matric potential. J. Hydol., 589: 125167. DOI:https://doi.org/10.1016/j.jhydrol.2020.125167

He, H., Li, M., Dyck, M., Si, B., Wang, J., Lv, J., (2020c). Modelling of soil solid thermal conductivity. Int. Commun. Heat Mass Transfer, 116. DOI:10.1016/j.icheatmasstransfer.2020.104602.

Hengl T, Mendes de Jesus J, Heuvelink GBM, Ruiperez Gonzalez M, Kilibarda M, Blagotić A, et al. (2017) SoilGrids250m: Global gridded soil information based on machine learning. PLoS ONE 12(2): e0169748. https://doi.org/10.1371/journal.pone.0169748

Johansen, O., 1975. Thermal conductivity of soils. PhD thesis, Uni-versity of Trondheim, 236 pp.

Lehmann, P., Stähli, M., Papritz, A. et al. A Fractal Approach to Model Soil Structure and to Calculate Thermal Conductivity of Soils. Transport in Porous Media 52, 313–332 (2003). https://doi.org/10.1023/A:1023595527999

Lu, S., Ren, T., Gong, Y., Horton, R., 2007. An improved model for predicting soil thermal conductivity from water content at room temperature. Soil Sci. Soc. Am J. 71, 8–14.

Lu, N., Dong, Y. (2015) Closed-Form Equation for Thermal Conductivity of Unsaturated Soils at Room Temperature. Journal of Geotechnical and Geoenvironmental Engineering, 141, 04015016-1.

Marthews, T.R., Quesada, C.A., Galbraith, D. R., Malhi, Y., Mullins, C.E., Hodnett, M.G., Dharssi, I., 2014. High-resolution hydraulic parameter maps for surface soils in tropical South America. Geoscientific Model Development 7, 711-723

Montzka, C., Moradkhani, H., Weihermüller, L., Hendricks Franssen, H-J,  Canty, M., and H. Vereecken (2012) Hydraulic parameter estimation by remotely-sensed top soil moisture observations with the particle filter, Journal of Hydrology, 399, 410-421

Montzka, C., Herbst, M., Weihermüller, L., Verhoef, A. and Vereecken, H.  (2017) A global data set of soil hydraulic properties and sub-grid variability of soil water retention and hydraulic conductivity curves.  Earth System Science Data, 9 (2).   pp. 529-543. doi: 10.5194/essd-9-529-2017.

Mualem, Y., 1976. A new model predicting the hydraulic conductivity of unsaturated porous media. Water Resour. Res., 12, 513-522.

Pachepsky Y. and W.J. Rawls, 2004. Development of Pedotransfer Functions in Soil Hydrology. Springer Verlag, Volume 30. 1st Edition

Rizvi, Z., H. Zaidi, A. Shoarian Sattari, and F. Wuttke (2019), Effective thermal conductivity of unsaturated sand using artificial neural network (ANN) and lattice element method (LEM), Int. J. Therm. Sci.

Rizvi, Z., S. Husain, H. Haider, and F. Wuttke (2020a), Group Method of Data Handling (GMDH) for computation of effective thermal conductivity of sands, Materials today: proceedings.

Rizvi, Z., S. Akhtar, W. Sabeeh, and F. Wuttke (2020b), Effective thermal conductivity of unsaturated soils based on deep learning algorithm.

Van Genuchten, M. Th. 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J., 44:892-898.

Vereecken, H.,  Weynants, M.,  Javaux, M.,  Pachepsky, Y.,  Schaap, M.G., Genuchten, M. Th.  Using pedotransfer functions to estimate the van Genuchten–Mualem soil hydraulic properties: A review Vadose Zone Journal 9 (4), 795-820

Verhoef A. (2004) Remote estimation of thermal inertia and soil heat flux for bare soil. Agricultural and Forest Meteorology, 123: 221-236.

Verhoef, A., Ottle, C., Cappelaere, B., Murray, T., Saux-Picart, S., Zribi, M., Maignan, F., Boulain, N., Demarty, J., and Ramier, D. (2012) Spatio-temporal surface soil heat flux estimates from satellite data; results for the AMMA experiment at the Fakara (Niger) supersite. Agricultural and Forest Meteorology, 154-155: 55-66.

Wen, H., J. Bi, and D. Guo (2020), Calculation of the thermal conductivities of fine-textured soils based on multiple linear regression (MLR) and artificial neural networks (ANN), Eur. J. Soil Sci., https://doi.org/10.1111/ejss.12934

Wösten, J.H.M., Lilly, A., Nemes, A., Le Bas, C., 1999. Development and use of a database of hydraulic properties of European soils. Geoderma, 90, 169-185.

Zou, H., N. Zhang, and A. J. Puppala (2019), Improving a thermal conductivity model of unsaturated soils based on multivariate distribution analysis, Acta Geotech.