John Koestel, Department of Soil and Environment
Soil is one of the natural resources that is crucial for the existence of a large number of microbial and the vast majority of higher terrestrial life forms, including humanity. Yet, we know surprisingly little about the details of soil functioning.
There are several reasons for this. Among them I consider three as especially important: i) the diverse structure and opaque nature of soils, ii) the complex interplay between soil physics, chemistry, biology and ecology and iii) and the emergence of large scale observation from small scale processes. In this presentation I give a short introduction into how modern three-dimensional imaging techniques and big data analyses will considerably advance our understanging of how soil systems function.
One fundamental property of soil systems is that they feature highly complex and diverse, hierarchically organized pore networks. They strongly impacts soil fertility, greenhouse gas emissions, nutrient retention, and contaminant degradation properties. The soil pore network also determines how oxygen, water and nutrients are transported and exchanged in soil. The architecture of the soil is therefore fundamental to soil functions.
By using three-dimensional X-ray imaging it is now possible to not only image the soil network architecture, but also to quantify the flow pathways of water through the pore network as well as the location of cation exchange sites, which bind positively charged solutes, e.g. metal contaminants. Such knowledge allows for a better understanding of the filtering capability of soil.
In another example I illustrate how time-lapse X-ray imaging reveals the evolution of soil structure with time and how roots, earthworms and ants are constantly reshaping the soil pore network. Thereby they are remodelling the habitat distribution for the soil microfauna and microorganisms. They are introducing new macropores that act as transport pathways for oxygen into the soil and they compress other parts of the soil, cutting them off from oxygen supply and making them susceptible for prolonged water saturation and anaerobic conditions.
Such isolated soil regions may become hotspots for N2O production, a powerful greenhouse gas. The above examples illustrate that physical, chemical, biological and ecological processes in soils are strongly intertwined. I therefore decididly advocate cross-disciplinary collaborations among soil scientists.
For practical applications of novel understanding of soil functioning, the detailed knowledge available at the scale of micrometers and millimeters needs to be translated to larger scales where it is required by stakeholders. As X-ray imaging is constraint to sample sizes of a few decimeters, many X-ray measurements are needed for this task and X-ray imaging has to be transformed into a high-throughput method.
During the last five years I have therefore developed a software that automizes X-ray image processing and analyses, which allows evaluating soil structures of a large number of soil samples. Databases with X-ray image data may then be used to parameterize, validate and improve simulation models for soil processes, especially when combined with complementary data.
For example, if combined with digital soil mapping, the soil structure and associated soil function may be inferred at a national scale using machine learning techniques to link both datasets. I am convinced that such ‘big data’ approaches will not only considerably advance the knowledge of our soils, but will also be a necessity to analyze the ever increasing amount of collected data.
Forskare, universitetslektor, biträdande vid institutionen för mark och miljö
Telefon: 018-67 24 10