Authors: Marta Vasconcelos Ottoni; Carlos Rogério de Mello; Márcio de Oliveira Candido and Quirijn by Jong Van Lier *
The complexity of nature manifests itself in the continuous transformation of terrestrial bodies-solid and liquid-during geochemical cycles and interconnected with biotic life and other elements of the planetary system. This intricate web of transformations and cycles makes life possible on planet Earth. A representative cut of its sphere would reveal, on a planetary scale, the interconnection of its elements, allowing the conceptualization and, later, the quantification of processes (e.g. physical, chemical, biological, hydrological, etc.) that operate on Earth, since their procedural and mathematical formulations are known. It is intuitive, therefore, to understand that the processes governing the earth develop in a particular structure or architecture, whose manifestation is largely dependent on the spatial and temporal observation scale.
To reinforce the idea about the relationship between the architecture’s architecture scale versus their corresponding processes, we will give an example: we imagine a small watershed being observed to the naked eye by an observer positioned on the earth’s surface. This surface becomes a point for an astronaut flying over and watching the planet. Thus, if every architecture has an associated process or functionality, we can then instinctively say that the processes represented on the scale observed by the observer on land should be quite different from those characterized on a planetary scale.
However, one may ask: the process on the watery scale is not occurring at the same time as the process represented on the planetary scale, so why are they different? They are different because when we are on the global scale, we take into consideration not only the small watershed of the example, but other land that are adjacent to that of the small basin and have other terrestrial structures and processes. Thus, the structure and process that manifest for the planetary observer is like an average of the architectures and terrestrial processes observed in the watershed of our example and its surrounding area. The issue of time dependence on the relationship arrangements versus terrestrial functions (or processes) is also very relevant, as biochemical cycles (water cycles, carbon, living organisms, etc.) are continually operating on Earth, making transformations in their architecture and, in turn, in associated features.
Given the importance of this theme in hydrology, we will address this subject in this article, aiming to highlight some concepts and challenges in characterizing the architecture of the soil terrestrial element and its hydrological processes, as soils are one of the main regulators of the water cycle and life on the planet.
Architecture and water functionality on global scale
Soil heterogeneity, on a planetary scale, is observed in different pedological units represented by a relatively thin layer on the earth’s surface. In this spatial dimension, seeing the soil with a human eye as an integrated body and related to terrestrial processes can be a challenge, even on a very small scale, where the complex solid-patient system manifests. To characterize the soil on the global scale, it is used to the use of
satellites that orbit the earth, performing constant images of the structure of our system.
The products of this technology are diverse and resolution compatible with several studies (in the order of meters to kilometers), generating information on the structure and behavior of the relief, the so -called digital lifting, soil use and occupation over time, behavior of water storage in basins (soil moisture and aquifers) and even inferences on the mineral constituents of rocks. Based on these images and field information, pedologists can make inferences on global pedological units of the earth’s surface, each presenting a conceptual architecture representing in depth soil variation.
Thus, through this architectural characterization of the soil, the processes that occur in this environment, including hydrologicals, are represented. However, to understand such processes, especially storage and water flows, it is necessary to determine the hydraulic properties of these environments on a global scale. They fall among the hydraulic properties, water retention (variants with the suction of water retained in the pores) and hydraulic conductivity (also variant with suction). These properties are part of the formulation of the Richards equation, which is our main model for the space-time description of moisture variation in porous means, but some studies raise the hypothesis that this equation is not valid to represent hydrological processes on the more global scale. However, we will remain here assuming how valid this formulation in the spacetime quantification of moisture variations in planetary soil.
Remote soil moisture sensors (Examples: SMAP, SMOS, and now NISAR) are one of the potential products to estimate the hydraulic parameters mentioned on a more global scale through reverse hydrological modeling (also using other hydrological data such as precipitation and evapotranspiration). This overall representation of hydraulic properties seems quite promising, as it can incorporate the effects of land use and occupation, hysteresis, preferential flows in the Vadosa zone and hydrophobicity, all largely affecting retention and water flows.
One of the limitations to the use of these moisture sensors in the estimation of global hydraulic properties is that they can only illustrate a small fraction of soil depth, a few centimeters, losing an important portion of the soil in the quantification of water processes. In addition, reverse modeling does not seem to be a trivial task and is not always successful in the result.
It is necessary to explain here the hysteresis of soil hydraulic properties, before continuing the subject on the agenda, given its importance in hydrological processes. Hysteresis represents a discrepancy between the values of hydraulic retention or conductivity, in a certain suction (or energy), in case of determination during drying or soil wetting. If the soil loses water by drying (water comes out of the soil), the water content retained in the porous system under a certain suction will be different from the content retained by the moistening process (water enters the soil system) for the same suction. This theme has a lot of relevance in hydrology, as some hydrological processes are moistened (eg infiltration) and others by drying (eg percolation, evaporation). Thus, the hydraulic properties in the soil (water retention and conductivity) will vary according to the process being represented. We will not expand the discussion of hysteresis here, as it is not the purpose of this article, but soil science and related areas should invest more in this matter.
Alternatives and challenges for the characterization of hydraulic properties
An alternative to the use of satellites to quantify globally hydraulic properties in soils and taking into consideration in depth variations, is to characterize the hydraulic properties in representative profiles of pedological units. The main disadvantage is that the values would be average per unit and not variants in space as are the results obtained by moisture sensors. Another problem concerns the pedological units themselves: would they be representative and can be considered coincident with hydrological units? However, we are dealing with a global scale, and soil variations within the same pedological unit do not appear to be very significant to quantify hydrological processes on this scale.
However, this proposal faces several challenges:
1- Representation of variability:
It is not enough just a single representative profile per pedological unit, since in each of these units there are textural, use and coverage variations, which affect the hydraulic properties of the soil. This fact already increases the time and effort in field work.
2- LIMITATIONS OF LABORATORY METHODS:
To minimize the operational work of field tests, hydraulic properties are determined in laboratory independent soil samples that, depending on their size, may not reproduce porous macrostructures
of the soil. Laboratory methods also do not take into account the phenomenon of hysteresis. Most measures water retention and hydraulic conductivity by the drying method.
A partial way out to overcome these difficulties is to measure the average properties in the soil profile in the field
Infiltration tests (today there are automatic technologies for this measurement), taking into consideration an area not very small to represent the porous macrostructures of the soil. The volume of water required in these tests remains a challenge, especially when it comes to areas far from water sources. However, the average hydraulic properties obtained by this method also do not take into account hysteresis, as field tests represent the properties by moisture.
3- Temporal variations:
How to consider the variations of soil properties in time, since the effect of use and coverage has proved to be quite variable in space and time?
Despite the limitations mentioned, the representation of hydraulic properties by representative profile of global pedological units, considering the laboratory analytical results in well -characterized samples in the field, is, in our understanding, the easiest and practical procedure to represent hydrological processes on this scale.
The infiltration tests in the field, although more representative of macrostructural units, demand a lot of time, water and are very laborious. It is crucial that the profiles are represented by typical uses and coverage of pedological units, as we know their effect on the determination of these variables.
The issue of volume representation that should be properly sampled to measure water properties is a subject that deserves greater investigation. The usual samples of 100 cm3 may be far from representing this representative volume of the soil called Rev, and there are suggestions in the literature to indicate the best sample size.
The issue of hysteresis is still a challenge, and the scientific community also needs to devote more to this theme, especially in laboratory methods, with previously said. Hydrophobicity is still a term little treated in Brazil and can have a relevant effect on the generation of runoff when rain begins in very dry periods.
To incorporate the issue of the temporal variability of hydraulic properties in soil hydrological modeling, a simple proposal would be to use the worldwide inventoried data from soil hydraulic properties related to the current use and coverage of the simulation. That is, if in the same place, the coverage vary, the registered data of the current use obtained from another location of the same pedological unit is used.
TERRITORIAL TERRITIONS
Another option to characterize hydraulic properties in the representative profiles of global soils is the use of pedometry techniques. In this scenario, it is essential to access a broad database of physical-hydrical properties so that the predictor variable variable models (here may be the physical, chemical and environmental variables of processing of digital terrain models), also called pedoffunctions, are developed. The challenge is to have access to these data, in particular, in tropical soils. However, scientific publications have increasingly charged transparency in editorial processes, requiring access to the databases used in studies, which has motivated the scientific community to share their analytical results, especially in existing data repositories around the world. We have made significant advances in recent years in sharing open data. In Brazil, we already have Soildata and Hybras, the first with compilations of pedological surveys conducted in Brazil and the second with information from hydraulic properties in the country’s soils. However, we hope to increasingly evolve on this subject by incorporating data from regions still in need of access to soil information.
Now, in the case of other territorial scales, in this case of our national and regional territory (eg in the order of 50,000 to 100,000 km2), the architecture and determination of its water functionality can follow the same patterns of structure representation and water proposed properties. When it is, however, microbasins (eg <10,000 km2), the sample soil challenge is greater in order to represent the variation of soil architecture more detail. However, remote sensing resources can still be very useful on this scale, such as determining soil digital maps, as sensing product resolutions can be in the order of a few meters. Pedometry on this scale has been a successful tool in these mappings, reducing costs and speeding up productions.
Hydraulic properties in these scales can be defined on samples randomly distributed or following some sample methodology, but the sample and analytical difficulty, either in the laboratory or in the field, still persists, as already mentioned. A faster solution would be to use some developed pedoffin that has the same geoenvironmental conditions and use and coverage of the study area. Access to these pedofofunctions, in the case of Brazil, is not yet facilitated, and when it is, origin works do not always make calibration methods, models mastery, calibration and validation performance errors, nor clearly explicit Machine Learning models or models.
Estimates of hydraulic properties estimates by pedoffunctions can be high. Thus, the caution when using existing pedofunctions of the literature in estimating the hydraulic properties of soils in the area of interest is necessary, and the ideal is to follow for a field investigation.
Proximal sensors (eg PXRF, X -ray fluorescence spectrometer) should be exploited as an alternative to estimate the hydraulic properties of the soil as they are non -destructive techniques with rapid processing in the results.
Exploited the possibilities of representation of soil architecture and water functionalities in the global, national/regional and microbasins, there would be comments on the theme Architecture x soil process on the scale of Catena, Profile and Poro, but we will leave this exploration for another opportunity.
Conclusions
Given the difficulties and diagnosis of the state state in the context of the representation of soil architecture and its hydraulic properties, the challenge of scientists in the areas of soil physics and hydrology is still very expressive. In the hydrological context, for example, simulation models for the impact of soil use and coverage on the behavior of the flow of a watershed still lack much development and research for significant advance towards more reliable results.
For this, we need better databases of soil hydraulic properties, especially the tropical, the use of remote sensors that allow to identify properties of plant coverage correlated with the hydraulic properties of the soil, and the construction of physical models that can be implemented with hydrologicals without prejudice to its parcimony.
Our goal with this article was to present concepts and how water in the soil still lacks innovative research in different territorial scales that can be implemented in different hydrological studies to improve our understanding of land systems. Thus, we hope to have contributed to those interested in the area to a better understanding of the challenges raised here.
About the authors:
*Marta Vasconcelos Ottoni and Márcio de Oliveira Candido are researchers of the Geological Service of Brazil
*Carlos Rogério de Mello is a professor at the Federal University of Lavras
*Jong Van Lier’s Quirijin is a professor at the University of São Paulo
No tags for this post.
