Soil Organic Matter Research and Climate Change: Merely Re-storing Carbon Versus Restoring Soil Functions

Over the last two decades, the sequestration of carbon in soils has often been advocated as a solution to mitigate the steady increase in the concentration of CO2 in the atmosphere, one of the most commonly mentioned causes of climate change. A large body of literature, as well as sustained efforts to attract funding for the research on soil organic matter, have focused on the soil carbon sequestration – climate change nexus. However, because CO2 is not the only greenhouse gas released by soils, and given the fact that the feasibility of large-scale carbon sequestration in soils remains controversial, this approach does not appear optimal to convince policy makers to invest in soils. In this perspective article, we argue that a far better strategy revolves around the effect of climate change on functions/services that soils render. In particular, since climatologists forecast less frequent but more intense rainfall events in the future, which may lead to food shortages, catastrophic flooding, and soil erosion if soils are not able to cope, a more suitable focus of the research would be to increase soil organic matter content so as to strengthen the water regulation function of soils. The different conceptual and methodological shifts that this new focus will require are discussed in detail.

The emphasis on carbon re-storing is understandable, given two key trends that have manifested in recent years. The first is that the public and political discourse on climate change has revolved predominantly around the concentration of CO 2 in the atmosphere, which has the advantage of being a simple enough metric to grasp. In this context, it is hardly surprising that researchers seeking funds to work on SOM have used as a sales pitch the reduction in atmospheric CO 2 that might result from carbon storage in soils (Minasny et al., 2017;Lal, 2018). The second trend is related to the development of carbon markets, which present great attractiveness for financial institutions and governments, and have received exponentially increasing attention in recent years (e.g., Abdul-Salam et al., 2019). Some view the inclusion of soil carbon sequestration efforts in this context as a cunning strategy not only to support agriculture economically and make it more sustainable in the long run, but also to incentivize farmers and land owners to preserve soil resources.
As long as, to meet political and economic expectations, research by SOM specialists remains fixated almost singlemindedly on the carbon storage -atmospheric CO 2 nexus, as seems to be the case now, we run the risk of missing important aspects of the climate change problem. Greenhouse gases (GHGs) include other chemical compounds than just CO 2 , some of which are far more potent than the latter. Methane (CH 4 ) is estimated to have a Global Warming Potential (GWP) that is between 28 and 36 times that of CO 2 over 100 years, whereas nitrous oxide (N 2 O) has a GWP between 265 and 298 over a similar 100-year timescale. Lugato et al. (2018) have argued that efforts to increase carbon storage in soils may result in the release in the atmosphere of significant amounts of N 2 O, an observation that is also supported by the fact that soil rich in C (for example, organic soils) tend to give rise to the highest N 2 O emissions (Pärn et al., 2018). Therefore, the net mitigating effect of soil carbon sequestration may be less than expected, yet unfortunately, there is no consensus at the moment on the issue. Several meta-analyses have evaluated how different C accrual strategies affect N 2 O emissions, but the results are not conclusive. For instance, the effect of adding organic amendments on N 2 O emissions is very dependent on the degree of stabilization of the material as well as many other factors (Chen et al., 2013;Cayuela et al., 2017). Reduced tillage, another strategy that has been promoted to store C, has been found to affect N 2 O emissions in different ways and may sometimes be associated to increased use of environmentally detrimental herbicides. In a meta-analysis comparing more than forty studies, van Kessel et al. (2013) concluded that N 2 O emissions relative to production (yield-scaled emissions) increased more than 50% in studies that had reduced tillage for less than 10 years, while in long-term studies a reduction of 27% was observed. The use of cover crops is also widely acknowledged as an approach to increase SOC. Basche et al. (2014) found great variability between studies, with an increase in emissions in 60% of the cases and a decrease in 40%. Their conclusion is that, when studies of a year or more are considered, the impact of cover crops on N 2 O emissions is negligible, but they stress the need for more long-term field studies to get robust conclusions.
The connection of soil carbon with GHG emissions gets even more murky when one looks not just at soils, but at entire agricultural systems. For example, the conversion of an agricultural field to a pasture may result in an increase in carbon storage in the subsurface over a certain period of time, before the system reaches a new equilibrium. However, as pointed out, e.g., by White et al. (2018), as soon as one puts grazing animals on that pasture, CH 4 emissions tend to increase drastically, leading to a net GHG abatement that can be less than predicted solely on the basis of soil carbon storage.

PERSPECTIVE SHIFT: RESTORING SOIL FUNCTIONS
Since the link between carbon sequestration in soils and the concentration of GHGs in the atmosphere is clearly not as easy to establish as was once thought, GHG reduction alone does not appear to be an appropriate driver to promote SOM research. A possible alternative approach to sell the research on soils to the general public would be to review in detail the various functions of soils and the crucial services they provide to human populations (cf Figure 1) 1 , to find out the extent to which each one of them could help mitigate climate change. However, our opinion is that a far better sales pitch for the research on SOM arises if we turn the looking glass around, and focus instead on the opposite perspective of the effects that climate change already has now, or is forecasted to increasingly have in the very near future, on the various soil functions on which society crucially depends. In this respect, the "4 per 1000 initiative, " launched by the French agriculture minister in 2015, included food security as one of its targets, even if, since, one could argue that it has been paid scarcely more than lip service in the related literature. Corbeels et al. (2019) opined recently that SOM researchers should focus on the production of food instead of on the release of GHGs. Undoubtedly, starvation and malnutrition are already fueling massive population exodus toward several industrialized countries, in particular from Sub-Saharan Africa to Europe and from Central America to the United States. In the longer run, the prospect of having to feed 9 or 10 billion people by the year 2050 is a daunting challenge, of which soil scientists ought to be constantly mindful. Our experience indicates that agro industries are very concerned by this problem. However, even though malnutrition has been a critical geopolitical issue FIGURE 1 | Schematic diagram of soil functions. This diagram is part of an uncommented "infographics" on soil functions put together by the FAO, with the subtitle "Soils deliver ecosystem services that enable life on earth." This particular diagram considers "carbon sequestration" and "climate regulation" functions separately (Adapted from http://www.fao.org/resources/infographics/infographics-details/en/c/284478/). for decades (e.g., de Castro, 1956), politicians appear to have demonstrated even less eagerness to tackle it seriously than they evince toward climate change. Therefore, one could argue that it would be wise for us to try to "sell" our research by emphasizing other soil functions/services, directly influenced by SOM content, that cause such societal concern at the moment that they would be more susceptible to incite politicians into action.
The downward trends in SOM contents in various parts of the world have been amply documented in recent years, as well as the significant effect that rising temperature can have in that respect (e.g., Conant et al., 2011;Johnston and Sibly, 2018). To understand the potential emergencies we face in this context, one needs to start by considering the effects that decreases in organic matter content have on a number of key soil properties. One could go down the list of physical, (bio)chemical, and biological properties, and demonstrate that the loss of organic matter has a significant impact on virtually all of them. But to make our point, it suffices here to consider physical properties.
With few exceptions (e.g., Santos et al., 1989;Radulovich et al., 1992), organic matter plays a key role in binding mineral particles together in most soils, and in making their architecture resistant, in particular to raindrop impacts at the soil surface or to protracted ponding (e.g., Rabot et al., 2018). Surface horizons depleted in organic matter tend to disperse more easily during rainfall events, and even in some cases tend to form largely impermeable crusts (Bresson and Boiffin, 1990;Bielders and Baveye, 1995;Bielders et al., 1996;Obale-Ebanga, 2001;Janeau et al., 2003), both of which accelerate the occurrence of ponding and runoff. This process diminishes the flood regulation function of soils, and in some extreme cases of erosion, where dense subsurface horizons become exposed, drastically decreases their potential for food production. Upon drying, dispersed clayey soils often have a tendency to become cemented, making it difficult for plants to germinate and grow.
As a corollary to the less stable architecture that results from the loss of organic matter, research has demonstrated that it also renders soils more vulnerable to compaction, which is increasingly regarded as one of the leading causes of soil degradation. This effect has become particularly acute in countries where mechanized agriculture is practiced and where, over the years, steadily heavier equipment has been used. In this respect, one might be tempted to think that the situation is not worrisome yet, and that it will take years for soils to get compacted to an extent that raises concern. But in some regions of the world, evidence to the contrary are in plain sight (Figure 2), and things are already now bordering on catastrophic, as soil researchers have repeatedly pointed out (Horn et al., 1995;Boivin et al., 2006;Gabarron-Galeote et al., 2019;Keller et al., 2019;Pöhlitz et al., 2020). Anecdotally, farmers one of us (P.C. B.) interviewed in the Picardie region in France (Northeast of Paris) have observed over the last two decades, as the organic matter content of their soils kept decreasing, that it became more and more difficult to grow their key cash crops, especially sugar beets. The reason is clearly due in part to increasing sensitivity of the soil to raindrop impacts, but predominantly to compaction by the hugely heavy "integral" machines now used (in the Fall, when the soils are wet) to harvest sugar beets. The biggest of these harvesters weigh more than 63 tons when their bunkers are full of beets. Many of the farmers in Picardie consider that they will no longer be able to grow sugar beets in a few (<10) years, at which point some of those who are old enough are planning to simply end their farming practice and retire.
These known facts about organic matter losses in soils take on significant urgency in the context of the ongoing climate change. One of the consistent predictions climate modelers have made over the last decade is that climate change will result in less frequent but more intense rainfall events in many parts of the world (e.g., Trenberth et al., 2003;Sun et al., 2007;Min et al., 2011;Ipcc, 2013;Intergovernmental Panel on Climate Change, 2014;Kendon et al., 2014;Berghuijs et al., 2017;Tang et al., 2019;Hess et al., 2020;Morán-Ordóñez et al., 2020;O'Donnell and Thorne, 2020). In North America, meteorological records indicate that heavy precipitation events have already increased in a large number of areas (Melillo et al., 2014). In the U.S. Midwest, for example, the quantity of precipitation occurring in the largest one percent of all daily events has increased by almost 40% over the last 60 years (Pryor et al., 2014). In the United Kingdom, modelers predicted a few years ago that heavy Summer downpours would become significantly more frequent in years to come (Kendon et al., 2014). A study carried out by the Met Office (2020) on Storm Desmond, which brought widespread flooding to Northern England and Southern Scotland in 2015, found such events have been made 59% more likely by climate change. This ongoing "rainfall intensification" process spells serious trouble in a number of ways on virtually all soils but especially so on soils that have lost a significant portion of their organic matter, because the problems that are already apparent in these soils now will get amplified in the future.
Protracted dry spells lead to significant wind erosion of soils in dryland areas, representing 40% of the world's land surface (Millennium Ecosystem Assessment, 2005) and 30% of the world population (UNCCD, 2014). This wind erosion gives rise to dust storms, which had not been seen since the 1930s but reappeared about a decade ago in various parts of the world (e.g., Baveye et al., 2011;McLeman et al., 2014;Barbosa and Olsson, 2019;Eekhout and De Vente, 2019;Middleton et al., 2019), and are forecasted to become even more frequent in the future. Beside dust storms, longer dry spells between rains could also cause severe water deficits for plants, e.g., crop failure in agricultural fields, and they could prolong the "hunger season" (the interval between two FIGURE 2 | Example of soil degradation brought about by compaction. (a) Field in Braine-l'Alleud (Belgium) where the use of heavy machinery, including sugar beet harvesters, over the years has resulted in soil compaction, lack of drainage, and difficulty for crops to germinate, (b) in this same field, evidence of dispersion of the soil at the surface, which at this point (April 4, 2020) is as hard as concrete when dry, (c) field soil in Lamairé (Deux-Sèvres, France) where pesticide application using heavy machinery has resulted in intense compaction. At various places, the tractor tracks are 20 cm deep, and the soil is once again extremely hard.
Frontiers in Environmental Science | www.frontiersin.org cropping seasons) in vulnerable and market-decoupled regions. Unless one builds massive numbers of artificial reservoirs, the water deficits will be impossible to compensate via irrigation since available groundwater or surface water resources are already severely overdrawn, causing a whole suite of environmental problems. In addition, more intense rainfall events generally mean quicker ponding at the soil surface and increased runoff, which in turn cause more severe soil erosion, an issue that has become critical in many countries (e.g., Nearing et al., 2004;Morán-Ordóñez et al., 2020;Wang et al., 2020). Beside via soil erosion, higher-intensity rainfalls may also have other effects on soil carbon storage, when the latter is increased through the incorporation of pyrolyzed organic material (e.g., agrichar or biochar). Physical modeling indeed shows that, under ponding conditions, agrichar or biochar particles have a tendency to rise to the soil surface because of their buoyancy, and to be transferred to runoff waters (e.g., Wang et al., 2013).
The higher runoff that one is likely to observe, either directly as a result of the higher rainfall intensity or indirectly, e.g., as a consequence of fires caused by drought (e.g., Candela et al., 2005;Papathanasiou et al., 2012), should also contribute to increase the frequency of severe floods and in particular "flashfloods, " at least in some parts of the world, causing significant damages to infrastructures, forcing 500 million people to flee their homes (UNCCD, 2014), and leading to the frequent loss of lives. The intensity and severity of the problem is heightened in many countries by the widespread sealing of soils, in particular because of urban sprawl (e.g., Miller and Hutchins, 2017;Skougaard Kaspersen et al., 2017;Rudd et al., 2020). In 2018, a team of researchers in Germany and the United States (Willner et al., 2018) concluded that in the next two decades, most of the United States, Central Europe, Northeast and West Africa, as well as large parts of India and Indonesia, will require intensive and onerous adaptation efforts, at least at double the current level, to keep high-end fluvial flood risk in check.

FUTURE RESEARCH AND PRACTICE
In this context, the imminent threats associated with rainfall intensification, yield stagnation, and land degradation seem to us to constitute a far better argument to convince decisionmakers to fund research on SOM, than the disputed connection between soil carbon sequestration and GHG emissions from soils. Such a shift of focus of the research will require changes in a number of respects.
First, the research will have to concentrate predominantly on the top portion of soil profiles. In the past decade, in part because plant roots often extend deep into soils (e.g., Baveye and Laba, 2015), some researchers have endeavored to estimate soil carbon stocks at depths between one and two meters (e.g., Baker et al., 2007;Rumpel and Kögel-Knabner, 2011;Börjesson et al., 2018). This makes sense from a carbon restoring perspective, as does the premise that any form of carbon, regardless of how much it contributes to the development or maintenance of a suitable soil architecture, is acceptable as input to increase stocks. However, neither one of these perspectives is meaningful from the viewpoint of making soils rapidly more resilient to rain intensification or of regenerating soil functions at an acceptable level. In fact, the focus on carbon storage may lead to the adoption of oversimplified practices, e.g., deep plowing (Alcantara et al., 2016(Alcantara et al., , 2017Keel et al., 2019), that appear convincing to decision makers but turn out eventually to be very deleterious to soils functions.
Second, to make sure that soils and their functions are apprehended in all their inherent (and often daunting) complexity, research teams will have to be more interdisciplinary than they generally are at present. A break with tradition is direly needed in that respect (Baveye and Wander, 2019;Baveye, 2020). Clearly, soil physicists and soil fauna specialists, who at the moment tend to be only peripherally if at all associated with the research on soil carbon sequestration, will have to be intimately involved in the work that will need to be carried out to understand key processes at the microscale in soils. Apart from anecic earthworm burrows aiding water infiltration, little to nothing is known mechanistically about the interactions between soil fauna and physical or chemical soil characteristics. It has been known for a long time that grazing microarthropods stimulate microbial activity and thus speed up OM degradation (Uvarov, 1985). But it is not clear which species of either group are involved, if this phenomenon holds true in soils of different mineral makeup, and how this will impact plant growth, plant-microbe interactions, or SOM stocks over the longer term. Soil physicists, zoologists, microbiologists, (bio)chemists, as well as agronomists will have to collaborate closely to resolve such questions (Briones, 2014;Filser et al., 2016). In particular, measurements will need to be carried out, and experiments designed, to assess not merely the amount of carbon re-stored in soils at any given time, but also how and to what extent this stored carbon improves the hydrology of soils, which molecular compounds are most effective in this context, and which organisms are instrumental in the transformation of added organic residues into these molecular compounds. From a broader perspective, to determine how these microscale processes affect the capacity of soils to regulate the cycle of water, it will be necessary to monitor simultaneously the dynamics of SOM as well as the water regulation function of soils under different conditions. Research carried out recently by Chalhoub et al. (2020) shows for the first time how direct measurements of soil water regulation services can be carried out in the field. All that will be needed to move forward will be to combine these measurements with a detailed analysis of the composition and evolution of SOM.
The focus on functions and services of soils in SOM research will necessitate interdisciplinary collaboration of a different kind as well. Experience has shown that in situations related to soil use that require a change in human attitude, soil expertise is in general ineffective, unless it is complemented by assistance from sociologists or psychologists. In this case, not only will experts from these disciplines need to be involved in the research, to gain insight into the human side of things, but stakeholders themselves (in particular farmers) will also have to be intimately associated with the work or even lead the questioning it will encompass. To achieve effective agri-environmental management, one will need to reenergize the research on indicators of what constitutes proper management, and their data requirements (Burton and Schwarz, 2013). The motivation of farmers to increase the organic matter content of their soil is a key condition of success. Soil researchers must acknowledge that decision making occurs at farm scale under heavy pressure of the value chain, and that, compared to that pressure, pure soil-related technical considerations or norms are of little relevance to trigger change (Seguin et al., 2007;Biardeau, 2018, 2019). In that context, our contacts with farmers suggest that increasing the soil productivity, and eventually maintaining it over the long haul, is a much stronger motivation for farmers at the moment than C sequestration. In any broadening of the scope to include other soil functions than just food production, farmers will have to be brought on board from the onset.
A third key change will also stem directly from the focus on functions and services, which by their very nature are very much affected by local conditions. As a result, the practical solutions that the research will try to identify will require intensive knowledge acquisition at the local scale, and will likely defy attempts at one-size-fits-all approaches imposed by global C markets.
Even though relatively profound changes will be required in research programs focusing on the effect of climate change on key soil functions, that does not mean that traditional research ought to stop altogether on carbon storage in soils or on practical approaches currently being investigated to sequester carbon. Cover crops, relay-or intercropping, conservation agriculture practices, and a return to mixed forms of agriculture (including both crop production and animal husbandry), which are currently advocated in various areas as convenient ways to incorporate more fresh plant residues and nitrogen in soils, are potentially useful also to restore the flood-regulation function of soils or their capacity to produce food. Undoubtedly, new approaches, which have not yet been developed or tested, will be needed as well, and will likely differ from region to region.

AUTHOR CONTRIBUTIONS
All authors contributed equally and substantially to the discussion that led to this perspective manuscript. PCB wrote a first draft of the manuscript, integrated editing inputs from all the other co-authors, and submitted the final version with their approval.

ACKNOWLEDGMENTS
Sincere gratitude is expressed to Dr. María Luz Cayuela (Spain) for providing several useful leads to the literature on N 2 O emissions from soils, and for general comments on the manuscript.