Search

Both herbivorous and bacterivorous microfauna have been shown to influence root development, soil nitrogen (N) mineralization, and plant productivity. However, our knowledge of these effects is limited as multitrophic interactions remain largely unexplored. We investigated whether and how herbivorous nematodes (Pratylenchus zeae) and bacterivorous nematodes (Poikilolaimus oxycercus), alone and in combination, affect plant biomass (Lolium multiflorum) through root traits and/or soil N mineralization. Bacterivorous nematodes increased, whereas herbivorous nematodes decreased, plant productivity. We found that root trait coordination in response to soil microfauna was consistent with the concept of root economics space. The negative interaction between herbivorous and bacterivorous nematodes on plant productivity at high herbivorous nematode infestation could be explained by reduced N mineralization and variation in the root nitrogen concentration-root tissue density (RNC-RTD) axis aligned with increased herbivory intensity. This study revealed that herbivorous and bacterivorous nematodes moderated each other's effect on plant productivity via root trait coordination and N mineralization, and suggests, for the first time, the value of the root economics space concept for interpreting phenotypic root plasticity and functioning in response to local biotic factors.

Capillary forces in soil allow counter-gravity moisture transport from groundwater to unsaturated layers. This matric suction driven process is often disregarded in biogeochemical studies because the impact on topsoil is largely unknown (capillary rise is not included in soil C models & difficult if not impossible to control in empirical research). As climate change is expected to intensify the occurrence and severity of droughts, the control of groundwater depth and capillary rise on topsoil heterotrophic activity requires better understanding. In this study, a laboratory set-up is employed to (1) examine topsoil moisture supply by capillary rise and (2) consequently the effect on mineralization of added plant material. Undisturbed two-meter-long soil columns of three different textures (loamy sand, (sandy) loam and silt loam) were subjected to two artificial groundwater table depths (–115 cm and –165 cm) while moisture transport was followed by volumetric water sensors. Additionally, carbon mineralization of added plant material was monitored over a period of ten weeks. We observed an upward moisture flux of capillary rise towards the topsoil for all textures and both groundwater depths. However, higher moisture contents were observed for the shallow groundwater level treatment (–115 cm) as a result of more rapid fluxes, with the largest effect for silt loam soils. Contrary to our expectations, this had no effect on the mineralization of added plant material. In summary, capillary rise succeeded to reach topsoil at groundwater depths to –165 cm for all textures, but no differences were found in mineralization of added plant material when groundwater increased by 50 cm. Field studies with native soil C instead of a plant-biomass substrate are ongoing.

Vermicomposting converts organic wastes into high-quality nutrient-rich compost and has been considered as a suitable technology for developing countries, especially at the household and community level as it is a simple and natural technology that does not require sophisticated machinery and frequent process monitoring. In addition, attempts to address SOM and nutrient losses in a sustainable manner in sub Saharan Africa countries require careful optimization of use of organic amendments such as through vermicomposting. We performed a vermicomposting experiment using three-earthworm species, exotic (Eisenia fetida and Eisenia andrei) and indigenous (Eudrilus eugeniae), fed on six different substrates made from agricultural wastes (cow and donkey manure separately spiked with three types of equally mixed crop residues (maize with soybean (MS), maize with banana (MB) and soybean with banana (SB)) to assess the nutrient stabilization in the resulting compost; and performed an incubation experiment using a selection of these vermicomposts to study the net N mineralization of the composts, and how they affected soil microbial properties such as microbial biomass C, dehydrogenase and β-glucosidase activities, in Nitisols. Earthworm activities during vermicomposting of agricultural wastes accelerated the rapid decomposition of mixed substrates and resulted with a significant reduction of TOC (C mass loss > 65%) and C:N ratios (twofold compared to the control without earthworm) and an increase in N mineralization and the total concentration of P, K and S. The effect of vermicompost on N availability and soil microbial quality significantly varied depending on vermicompost sources. The vermicompost made from cow manure mixed with soybean and banana residues using Eudrilus eugeniae had the highest net N mineralization (32.4 µgNg-1 soil, 75% of N added). Regardless of vermicompost type, amended soils had significantly higher microbial and enzymatic activities than unamended control. Our results suggest that vermicompost sources tested could reliably use for long-term improvement of soil nutrient mineralization and microbial activity under Nitisols of Ethiopia due to their greatest microbial activities and nutrient dynamics during vermicomposting.

To cope with climate change, the use of water-saving irrigation practices in rice production is increasingly promoted, instead of the traditional practice of continuous flooding. Yet, it is unclear how the periodic introduction of oxygen during drying events impacts some fundamental biogeochemical processes in paddy soils because of the difficulty to quantify them. This knowledge gap actually complicates the development of mechanistic biogeochemical models, which are key for the prediction of paddy soil nutrient dynamics over a range of edaphic, agronomic and climatic settings. However, stable isotopes may help to shed some light on dynamics of important soil nutrients, i.e., by exploiting their natural or artificial (by labelling) variations over different biogeochemical transformations. In an effort to improve their implementation in soil models, we experimentally investigated the effect of water-saving irrigation management on three prominent but understudied carbon and nitrogen transformations in paddy soils from Italy and Bangladesh by means of stable isotopes. In particular, we assessed the oxidation of methane (CH4) in situ, the decomposition of soil organic carbon (SOC) after the incorporation of organic residues, and the fixation and defixation of ammonium (NH4+) in clay interlayers, always as a function of irrigation management. As it turns out, the adoption of water-saving irrigation practices had the potential to relatively stimulate in situ CH4 oxidation, in spite of some methodological difficulties, which implies that the commonly observed lower CH4 emissions in paddy fields under water-saving irrigation probably do not only stem from lower CH4 production rates under less anaerobic conditions, but also from this mitigation mechanism. Secondly, the stimulation of native SOC dissolution after the incorporation of organic residues was restricted under water-saving irrigation, and redox-active iron oxides with sorbed SOC potentially play a role in this. However, this effect on SOC dissolution was not proportionally reflected in SOC-derived emissions of CH4 and carbon dioxide (CO2), so that irrigation management probably does not influence these emissions and the long-term SOC balance through the proposed mechanism. Thirdly, the effect of irrigation management on abiotic fixation and defixation dynamics of NH4+ into or from interlayers of clay minerals was rather limited, but these processes – which determine the bioavailability of NH4+ in paddy soil – were instead predominantly dictated by an equilibrium between bioavailable and clay-fixed NH4+ phases. Overall, with a well-considered experimental setup in which untargeted isotope contrasts can be quantified and the appropriate pools sampled, stable isotopes are apparently a powerful tool to quantify gross nutrient transformations in paddy soil.

The drumbeat to reach net-zero emissions by 2050 and foster a European circular economy has arguably thrust anaerobic digestion front and centre due to its ability to produce renewable energy and fertilisers from recycled nutrients in the form of digestate. In this context, it is of paramount importance to better understand the fate of the nitrogen (N) and carbon (C) from digestate when applied to soil. In this study, the C and N mineralisation rates of five digestates from various feedstocks, undigested pig slurry and compost were studied in back-to-back incubation experiments in which predetermined amounts of these products were applied to soil. The treated soil mixtures were placed in PVC tubes, packed to a bulk density of 1.4 g cm-3, and their water content adjusted to 50% water-filled pore space. For soil incubations aimed at assessing N mineralisation patterns, destructive sampling was carried out every 20 days (over 4 months), during which ammonium nitrogen (NH4+-N) and nitrate nitrogen (NO3--N) were measured. For C soil incubations aimed at monitoring organic carbon (Corg) mineralisation, each PVC tube was placed inside an airtight glass jar with an NaOH 0.5 M vial to trap CO2 from the treated soils. The treatments were incubated over 149 days, during which time the NaOH traps were periodically removed (17 sampling moments) and back titrated with 0.5 M HCl to measure the evolved CO2. Table 1 summarises the main results of the C and N incubations. Net N mineralisation (Nmin) was determined by subtracting the mineral N content of each treatment on day 0 from the mineral N at all subsequent measurements and expressed as percentage of applied organic N (Norg). As shown in Table 1, a positive Nmin (%Norg added) was observed for all products at the end of the experiment (day 127). Regarding C incubations, the amount of mineralised C (Cmin) from the treatments was calculated by subtracting the cumulative CO2 evolved from the untreated soil and was expressed as a percentage of added Corg from each product. A second-order kinetic model was fitted to the Cmin pattern of each of the tested products (Sleutel et al.,2005), from which the humification coefficient (Hc) of each treatment, defined as the amount of C remaining after one year, could be extrapolated. Results show that the Hc ranged from 50 to 81% for raw digestates (Table 1). The initial TC:TN ratio of the products was significantly negatively correlated (p < 0.01) to the amount of Cmin on the last day of C incubations. Also on the final day of N incubations, Nmin (p < 0.05) was significantly positively correlated with the initial NH4+-N:TN ratio of the products and negatively correlated with TC:TN (p < 0.05), as highlighted previously (Tambone & Adani, 2017). Thus, the TC:TN ratio of products proved a reliable predictor of C and N mineralisation rates in the incubated soils. Moreover, the Nmin rates (%Norg) of the studied digestates on the last day of the experiment ranged from 20 to 39% (Table 1). In other words, this suggests that anywhere between 60 to 80% of Norg was still present in the fertilised soil treatments after 127 days. Therefore, more attention might be given to the Norg fraction that still remains after application. In this respect, field trials would be warranted to assess the long-term fate of mineralised N when digestates are applied to a plant-soil system. Under this scenario, the synchrony, or possible asynchrony, between the mineralisation rates of the remaining Norg pools from the digestates and plant N uptake could be better monitored. Also of note, while digestates all exhibited an initial N immobilisation phase (not shown), the dynamics were very different from one product to the other, with some showcasing a comparatively faster remineralisation (from day 20 onwards) than others (from day 40 onwards). Thus, an understanding of the immobilisation/remineralisation pattern of a specific digestate could constitute a useful tool, among others, to reach an informed decision in the field regarding appropriate timing of N fertilisation, thereby reducing potentially counterproductive effects on crop growth. The amounts of Hc after one year, from the C incubation experiment, ranged from 50 to 81% for the digestates (Table 1). This result suggests that, without reaching thelevels of compost (99%), these products had an equally interesting C sequestration potential in the incubated soil, aside from their fertilising value.

Soil texture determines microbial activity and organic matter (OM) decomposition in several ways. Next to direct physicochemical stabilization of OM, the location of moisture in the soil matrix also depends on soil texture. At lower moisture content, the contact between soil moisture and OM might be critical for OM decomposition, because such areas of contact could function as a ‘bridge’ for transport of moisture, substrates and microorganisms. Additionally, water absorption by porous OM itself likely also depends on soil texture and this may thus bring about another indirect textural control on OM mineralization. However, the relevance of such indirect soil textural controls on OM mineralization has not been investigated. We therefore compared the mineralization of 13C-labeled maize particles (Cmaize-min) in a loamy sand and silt loam soil at three levels of water-filled pore space (20, 40 and 60%WFPS). The distribution of maize particles and moist soil volume was assessed at the microscale using X-ray μCT and contrast-enhancing agents. We hypothesized that a lower contact between OM and soil in the loamy sand soil would lead to a reduced OM decomposition compared with the silt loam soil. On the other hand, we expected that OM itself could attract more moisture from the surrounding loamy sand soil due to a higher water potential than in the silt loam soil at equal WFPS level. Contrary to our expectation, we found a lower contact between zones of moist soil and the added OM in the silt loam soil. But as expected, OM was able to attract more water from the loamy sand soil, and it preserved a moist state even when soil was at a low moisture content, which may explain the surprisingly limited inhibition of Cmaize-min in drier loamy sand soil. This finding suggests that such a ‘sponge-effect’ of OM may be a relevant yet largely overlooked element in studying the relationship between soil moisture and OM decomposition. Contrary to our expectation, Cmaize-min was higher in the silt loam soil texture across all three moisture levels (P

Degradability of organic matter (OM) in soil depends on its spatial location in the soil matrix. A recent breakthrough in 3D-localization of OM combined dual-energy X-ray CT-scanning with OsO 4 staining of OM. The necessity for synchrotron-based µCT and the use of highly toxic OsO 4 severely limit applications in soil biological experiments. Here, we evaluated the potential of alternative staining agents (silver nitrate, phosphomolybdenic acid (PMA), lead nitrate, lead acetate) to selectively enhance X-ray attenuation and contrast of OM in CT volumes of soils containing specific mineral soil particle fractions, obtained via lab-based X-ray µCT. In comparison with OsO 4, administration of Ag + and Pb 2+ resulted in insufficient contrast enhancement of OM versus fine silt (< 20 µm) or clay (< 2 µm) mineral particles. The perfusion procedure used in this work induced changes in soil structure. In contrast, PMA staining resulted in a selective increase of OM’s attenuation contrast, which was comparable to OsO 4. However, OM discrimination from other soil phases remained a challenge. Further development of segmentation algorithms accounting for grey value patterns and shape of stained particulate OM may enable its automated identification. If successful in undisturbed soils, PMA staining may form an alternative to OsO 4 in non-synchrotron based POM detection.

Degradability of organic matter (OM) in soil depends on its spatial location in the soil matrix. A recent breakthrough in 3D-localization of OM combined dual-energy X-ray CT-scanning with OsO4 staining of OM. The necessity for synchrotron-based µCT and the use of highly toxic OsO4 severely limit applications in soil biological experiments. Here, we evaluated the potential of alternative staining agents (silver nitrate, phosphomolybdenic acid (PMA), lead nitrate, lead acetate) to selectively enhance X-ray attenuation and contrast of OM in CT volumes of soils containing specific mineral soil particle fractions, obtained via lab-based X-ray µCT. In comparison with OsO4, administration of Ag+ and Pb2+ resulted in insufficient contrast enhancement of OM versus fine silt (4. However, OM discrimination from other soil phases remained a challenge. Further development of segmentation algorithms accounting for grey value patterns and shape of stained particulate OM may enable its automated identification. If successful in undisturbed soils, PMA staining may form an alternative to OsO4 in non-synchrotron based POM detection.

Increasing interest in plant-root phenotyping has stimulated the development of X-ray μCT-based root/soil segmentation protocols. However, most scanning and CT volume processing protocols were only applied for detection of simple and small juvenile roots. We tested a new methodology for its ability to extract large mature maize roots from X-ray μCT volumes of undisturbed soil monoliths (10 × 20 × 15 cm) and compared its performance with two existing (semi-)automated segmentation algorithms. The X-ray μCT based root assessments were validated by means of the measured root biomass. Segmentation of maize root systems proved to be particularly challenging because regardless of soil type, within-sample vertical gradients in X-ray attenuation caused an overlap in CT-reconstructed grey values of roots, water and mineral phases. However, in contrast to the two existing algorithms, a new alternative methodology did allow for relatively fast and accurate segmentation of the mature roots. This was evidenced by a highly significant correlation between the CT-derived root volume and its measured root biomass. Despite the large size of the investigated soil monoliths, roots could be resolved down to a diameter of 200 μm. This methodology opens up new possibilities for investigating large and complex root systems, and may become an invaluable new tool in plant breeding research.

The soil environment impacts carbon turnover in a multitude of ways. It has been shown that C emissions are higher when exogenous carbon is locally concentrated than when it is spread in the soil matrix. Hence, the spatial separation between decomposers and substrate may cause ineffective dispersion of the produced enzymes within the soil matrix. It is not clear, however, if and how lower doses of exogenous C impact its degradability and if an interaction exists with native soil organic matter (SOM) level. Therefore, a seventy-day incubation experiment with six varying exogenous OM doses (0-6 g C kg-1) in soil mesocosms of loamy sand subsoil (WFPS 50%) with three different native SOM contents (i.e. 0.1, 0.5, and 1.0 %) was carried out. C emission and isotopic signature measurements of 13CO2 by means of a CRDS-analyser were used to track the degradation of added maize shoots with a δ13C=-12.7‰ that contrasts to native SOM (δ13C=-25‰). Mineral N was added, and the redox potential (Eh) monitored. Eh suggested that O2 may have not become limiting for C mineralisation, even at higher doses, but it did not clearly relate to the added C dose. Cumulative CO2 emissions increased proportionally with C dose and so our hypothesis that a lower dose dispersion limits microbial activity was disproved. However, a higher native SOM level led to higher relative degradability of added C for the highest doses (i.e. 3 and 6 g kg-1 soil). Furthermore, a negative impact of OM doses above 1 g kg-1 soil on native SOM degradation was found. We conclude that positive priming of native SOM mineralisation did not occur. A disproportional response of microbial biomass carbon and soil C mineralisation at lower doses suggests that dose does impact the microbial metabolic quotient. Ongoing PLFA analyses will help us interpret these observations.

Soil moisture strongly affects the balance between nitrification, denitrification and N2O reduction and therefore the nitrogen (N) efficiency and N losses in agricultural systems. In rice systems, there is a need to improve alternative water management practices, which are designed to save water and reduce methane emissions but may increase N2O and decrease nitrogen use efficiency. In a field experiment with three water management treatments, we measured N2O isotope ratios of emitted and pore air N2O (δ15N, δ18O and site preference, SP) over the course of 6 weeks in the early rice growing season. Isotope ratio measurements were coupled with simultaneous measurements of pore water NO3-, NH4+, dissolved organic carbon (DOC), water-filled pore space (WFPS) and soil redox potential (Eh) at three soil depths. We then used the relationship between SP × δ18O-N2O and SP × δ15N-N2O in simple two end-member mixing models to evaluate the contribution of nitrification, denitrification and fungal denitrification to total N2O emissions and to estimate N2O reduction rates. N2O emissions were higher in a dry-seeded + alternate wetting and drying (DS-AWD) treatment relative to water-seeded + alternate wetting and drying (WS-AWD) and water-seeded + conventional flooding (WS-FLD) treatments. In the DS-AWD treatment the highest emissions were associated with a high contribution from denitrification and a decrease in N2O reduction, while in the WS treatments, the highest emissions occurred when contributions from denitrification/nitrifier denitrification and nitrification/fungal denitrification were more equal. Modeled denitrification rates appeared to be tightly linked to nitrification and NO3- availability in all treatments; thus, water management affected the rate of denitrification and N2O reduction by controlling the substrate availability for each process (NO3- and N2O), likely through changes in mineralization and nitrification rates. Our model estimates of mean N2O reduction rates match well those observed in 15N fertilizer labeling studies in rice systems and show promise for the use of dual isotope ratio mixing models to estimate N2 losses.

Mineral nitrogen (N) availability to heterotrophic micro-organisms is known to impact organic matter (OM) decomposition. Different pathways determining the N accessibility depend to a large extent on soil structure. Contact between soil mineral and OM substrate particles can facilitate N transport toward decomposition hot spots. However, the impact of soil structure on N availability to microbes and thus heterotrophic microbial activity and community structure is not yet fully understood. We hypothesized that carbon mineralization (Cmin) from low-N substrate would be stimulated by increased N availability caused by closer contact with soil particles or by a higher moisture level, enhancing potential for N-diffusion. Under opposite conditions retarded heterotrophic activity and a dominance of fungi were expected. A 128-days incubation experiment with CO2 emission monitoring from artificially reconstructed miniature soil cores with contrasting soil structures, viz. high or low degree of contact between soil particles, was conducted to study impacts on heterotrophic activity. The soil cores were subjected to different water filled pore space percentages (25 or 50% WFPS) and amended with either easily degradable OM high in N (grass) or more resistant OM low in N (sawdust). X-ray μCT image processing allowed to quantify the pore space in 350 μm around OM substrates, i.e., the microbial habitat of involved decomposers. A lower local porosity surrounding sawdust particles in soils with stonger contact was confirmed, at least at 25% WFPS. Mineral N addition to sawdust amended soils with small particle contact at 25% WFPS resulted in a stimulated respiration. Cmin in the latter soils was lower than in case of high particle contact. This was not observed for grass substrate particles or at 50% WFPS. The interactive effect of substrate type and soil structure suggests that the latter controls Cmin through mediation of N diffusion and in turn N availability. Phospholipid fatty acid did not reveal promotion of fungal over bacterial biomarkers in treatments with N-limited substrate decomposition. Combining X-ray μCT with tailoring soil structure allows for more reliable investigation of effects on the soil microbial community, because as also found here, the established soil pore network structure can strongly deviate from the intended one.

In a context of increasing food demands and a changing climate causing decreased water availability, alternative irrigation management (AIM) in paddy rice production is broadly being applied in Asia and other rice producing regions. In combination with efficient soil fertilization management, AIM aims at sustainably intensifying rice production while using less water. However, AI ’s increasing popularity may impose a trade-off from saving water to increasing GHG emissions (mainly of N2O) and affecting micronutrient availability in paddy soils. The general objective of this research was to assess the effect of AIM on micronutrient availability (Fe, Mn, Cu and Zn), GHG emissions and other relevant related soil and plant parameters. Soils from the plough layer of 3 Philippine farmers’ paddy fields with differing soil textures were collected for a greenhouse pot experiment. The samples were subjected to three commonly applied irrigation techniques, differing in the number and duration of draining events along the growing season: continuous flooding (CF), alternate wetting and drying (AWD) and mid-season drainage (MSD). Micro-nutrient availability to plants (DTPA extracts), GHG emissions (CH4, CO2 and N2O) and various biogeochemical parameters such as exchangeable NH4, dissolved OC (DOC), pH and soil redox potential and temperature were assessed during the entire growing season. Other plant related parameters (plant height, tiller and leaf number and mean biomass) were measured during harvest. Results indicate a reduced availability of Fe and Mn under an AWD compared to CF regime, particularly during the reproductive growth phase of the rice plants. During the earlier growth stages, however, Cu and Zn concentrations show an opposite trend for all soil types. As for mineral nitrogen concentrations, a strong decline was observed for all three soil types, regardless of the water management applied. By the end of the ripening stage, contents close to zero kg ha-1 were generally observed, probably due to enhanced volatilization of NH3 towards the end of the growing season, caused by the high ambient temperatures in the greenhouse. DOC concentrations did not show clear differences between any of the treatments. CH4 fluxes were surprisingly high at the onset of the growing season but they quickly declined within a time span of three weeks. Due to considerably low CH4 emission values during the rest of the growing season (not exceeding 10 mg m-2 d-1), no significant differences could be observed between the three irrigation treatments. Analyses of measured N2O and CO2 emissions are still in progress. Rice plant height, tiller and leaf number and mean biomass were all favoured by CF when compared to AWD management in all three soils. The above mentioned results indicate that while water saving management in rice farming reduces emissions of CH4, it may as well limit micro-nutrient availability for rice plant uptake, and in our pot experiment also crop growth.

The role of soil structure in organic matter (OM) stabilization has been primarily investigated through physical fractionation studies operative at the scale of aggregates and smaller organo-mineral particles. By narrowing down soil structure to an arrangement of mineral and organic particles, the majority of studies did not explore the spatial organization of the soil pore network, the actual habitat of microorganisms. In a lab experiment we incubated a sandy loam soil (with application of ground grass or sawdust) in 18 small aluminum rings (Ø 1 cm, h 1 cm). Bulk density was adjusted to 1.1 or 1.3 Mg m-3 (compaction) and 6 rings were filled at a coarser Coarse Sand:Fine Sand:Silt+Clay ratio.

Soil moisture strongly affects the balance between nitrification, denitrification and N2O reduction and therefore the nitrogen (N) efficiency and N losses in agricultural systems. In rice systems, there is a need to improve alternative water management practices, which are designed to save water and reduce methane emissions but may increase N2O and decrease nitrogen use efficiency. In a field experiment with three water management treatments, we measured N2O isotope ratios of emitted and pore air N2O (δ15N, δ18O and site preference, SP) over the course of 6 weeks in the early rice growing season. Isotope ratio measurements were coupled with simultaneous measurements of pore water NO−3, NH+4, dissolved organic carbon (DOC), water-filled pore space (WFPS) and soil redox potential (Eh) at three soil depths. We then used the relationship between SP × δ18O-N2O and SP × δ15N-N2O in simple two end-member mixing models to evaluate the contribution of nitrification, denitrification and fungal denitrification to total N2O emissions and to estimate N2O reduction rates. N2O emissions were higher in a dry-seeded + alternate wetting and drying (DS-AWD) treatment relative to water-seeded + alternate wetting and drying (WS-AWD) and water-seeded + conventional flooding (WS-FLD) treatments. In the DS-AWD treatment the highest emissions were associated with a high contribution from denitrification and a decrease in N2O reduction, while in the WS treatments, the highest emissions occurred when contributions from denitrification/nitrifier denitrification and nitrification/fungal denitrification were more equal. Modeled denitrification rates appeared to be tightly linked to nitrification and NO−3 availability in all treatments; thus, water management affected the rate of denitrification and N2O reduction by controlling the substrate availability for each process (NO−3 and N2O), likely through changes in mineralization and nitrification rates. Our model estimates of mean N2O reduction rates match well those observed in 15N fertilizer labeling studies in rice systems and show promise for the use of dual isotope ratio mixing models to estimate N2 losses., Biogeosciences, 16 (2), ISSN:1726-4170

In summary, this PhD study investigated the relative contribution of below- versus above-ground biomass to the buildup and maintenance of stable SOC stock. We found that the relative stability of below- and above-ground biomass measured over a decadal period differs by a factor of 3.5. This implies that management oriented at storing SOC needs to be root-oriented and this knowledge could further help to improve the prediction of SOC dynamic by biogeochemical modeling. It also remains to be tested to what degree current biogeochemical models are capable of capturing these trends correctly. Furthermore, in this PhD it was also suggested - but not confirmed - that a delayed physical disturbance of root residues in the field and post-harvest remnant belowground biomass derived C could be the potential reasons for apparent relative stability of belowground biomass. Root biomass is mostly predicted based on aboveground yield by fixed R/S ratios, often leading to over- or underestimation under various soil conditions. Therefore, fast and reliable quantification of root biomass remains another big challenge to improve simulations of dynamics of SOC.

In many agricultural systems, nitrogen (N) is the most common nutrient limiting plant growth, and widespread application of N fertilizers has contributed to huge gains in productivity across the world. This increase in crop production is necessary to feed a growing population and to mitigate the conversion of wild lands to crop lands. However, in parallel with increased N application and crop yields, N pollution has also increased rapidly. The ease of access and price of N fertilizers in many parts of the world, as well as the expense or time involved in the implementation of improved practices, results in over application of N in relation to plant demand. Excess N is highly susceptible to loss and contributes to a host of environmental problems such as eutrophication of water bodies, contamination of drinking water, loss of biodiversity and global warming. N pollution largely contributes to global warming through the production of nitrous oxide (N2O), since approximately 60% of global N2O emissions come from agriculture. N2O is a powerful greenhouse gas, with a global warming potential 280 times that of CO2. N2O emissions are difficult to manage because they represent a small fraction of applied N, are produced by multiple pathways and are highly heterogeneous in space and time. The case of irrigated rice production exemplifies well the challenges and trade-offs of managing for N2O emissions. For example, the efforts made with alternate water management strategies, aimed at mitigating methane emissions (10% of global emissions) and reducing water consumption (40% of global irrigation water consumption), can also lead to higher N2O emissions, thereby negating any net reduction in greenhouse gas emissions. This thesis has three foci: a) to evaluate agricultural management strategies to mitigate N2O emissions across different scales; b) to improve our mechanistic understanding of the processes underlying N2O emissions or more general N cycling under alternative water management in irrigated rice production and c) to advance natural abundance N2O isotopocule methodologies used to identify the processes involved in N2O production, N2O consumption and their relative contribution to total N2O emissions. In the first study (Chapter 2), I performed a meta-analysis to evaluate the efficacy of biochar to mitigate N2O emissions at the field scale in different cropping systems. Biochar is a carbon (C) rich, charcoal-like product formed during the pyrolysis of organic matter. In this meta-analysis, I found that biochar did decrease N2O emissions by 9.2 to 12.4% across cropping systems. However, the magnitude and significance of this reduction was affected by the weighting function (either by variance or by number of observations). Indeed, currently available field data is biased by a disproportionate number of studies in flooded rice systems (41% of observations), a low number of studies with a comparatively high number of treatments, and an uneven representation of biochars and N fertilizers used. My second, third and fourth studies (Chapters 3-5) were conducted over the course of two growing seasons in the Northern Italian rice-growing region. These studies set out to improve our understanding of how N cycling, N2O production and consumption processes change along a soil profile in relation to flooding and drainage events and ultimately how surface emissions are affected. To this end, I utilized an experimental platform with three water management treatments: water-seeded + conventional flooding (WS-FLD), water-seeded + alternate wetting and drying (WS-AWD), and direct dry seeding (DS-AWD). Soil environmental parameters as well as N and carbon (C) substrates were measured in parallel with N2O concentrations and isotopocule signatures of emitted N2O and pore air N2O. N2O emissions in both years were higher in the AWD treatments. These emissions were more affected by the mineralization of native soil N or availability or pre-season fertilizer N in the top soil rather than by fertilizer applied within the growing season. N2O production was most likely limited by nitrification and the production of NO3- or NO2-, but was ultimately derived from coupled nitrification-denitrification, from denitrification or from nitrifier-denitrification. Results of a dual isotope mapping and mixing model approach with site preference (SP) and δ18O-N2O ratios showed extensive N2O reduction in all treatments regardless of more aerobic conditions in the DS-AWD treatment. The isotope modeling approach allowed me to estimate N2O reduction to N2, a notoriously unconstrained aspect of the N cycle. Mean N2 emissions ranged from 245-333 g N ha-1d-1 (~ 29-40 kg N ha-1 per 120d season), thus representing a considerable loss relative to applied N. My final study (Chapter 5), evaluated the role of interactions between iron (Fe) and N reduction-oxidation reactions on NH4+ availability. NH4+ fixation was highest in the WS-FLD but more dynamic in the WS-AWD treatment, demonstrating an influential role of water management. Fe reduction and the likelihood of Fe-N redox reactions clearly increased in the WS treatments but I could not directly relate these processes to NH4+ concentrations. In summary, this thesis demonstrates the ability to detect significant impacts of management practices such as biochar additions or AWD irrigation on N2O emissions. I also showed that there is a real potential to use the SP-N2O and δ18O-N2O to source-partition N2O production and estimate N2 emissions, thereby better constraining N2O emissions and N2 losses. In irrigated rice systems, nitrification in the top soil appears to be a bottleneck for N2O production and N2 losses. Therefore, management to reduce nitrification such as delaying pre-plant fertilizer application or the use of slow release fertilizers may help to reduce N losses. Future work should investigate if processes such as NH4+ fixation can be beneficially manipulated to further regulate NH4+ availability in relation to plant demand, thereby mitigating N2O emissions and improving the nitrogen use efficiency of cropping systems.

Soil organic carbon (SOC) is involved in food security, climate change mitigation and other ecosystem services. Assessing SOC dynamics is receiving more and more attention in the context of global changes. SOC consists of organic compounds with a continuum of decomposability and turnover rates which can be divided in conceptual discrete pools that are functionally homogenous. Recently, SOC fractions that correspond to these theoretical pools were isolated through physico-chemical extraction. These fractions are usually based on (de-)stabilization mechanisms that influence SOC dynamics. Our hypothesis is that the evaluation of SOC dynamics across scales could be enhanced by means of SOC fractionation. Thus SOC fractions contribute to the characterization of SOC dynamics in response to anthropogenic forcings in different settings over timescales ranging from years to decades. This thesis aims to: (i) Evaluate the suitability of organic carbon associated with clay and fine silt as an indicator for SOC decadal evolution under different residue management practices in a long-term experiment in Belgium. (ii) Evaluate the dynamics of functional soil organic carbon pools during vegetation recovery after land abandonment in Andalucía. (iii) Evaluate the usefulness of SOC fractions to understand SOC stocks distribution in a hilly catchment after recent conversion to agriculture in southern Brazil. (iv) Improve the assessment of SOC state and related soil ecosystem functions using SOC fractions in meaningful units for agricultural management in Wallonia. We used these different study sites as they were selected according to the availability of data and their potential to demonstrate the effect of a single driving factor. We demonstrated that the number of fractions to be distinguished is a function of both the complexity of the system in terms of soil disturbance and the number of samples required to cover the study area. (SC - Sciences) -- UCL, 2016

Elevated depositions of non-marine sulphate (SO42-) and inorganic nitrogen (N), as a consequence of air pollution, resulted in a progressive acidification and eutrophication of Flemish forests. Since the 1980s emission abatement reduced the acidifying and eutrophying emissions and depositions in Flanders. This thesis aimed to evaluate the impact of the evolution in depositions on soil solution chemistry, using long-term data collected in 5 plots of the ICP Forests monitoring (Level II) network in Flanders. The sharp decrease of SO42- and ammonium (NH4+) depositions made that abiotic N status started to improve and acidification slowed down during the past two decades. However, N depositions are still far above the critical loads for ectomycorrhiza and epiphytic lichens. Given the still very low soil pH (3.5‒4.5) unfavourable for microbial life, the generally observed tendency of increased dissolved organic carbon (DOC) and nitrogen (DON) mobility is likely a direct result of lowered ionic strength and partly rise in pH. Abiotic recovery is delayed by a simultaneous decrease in the deposition of base cations (Ca2+, K+ and Mg2+) and SO42- desorption. Biotic recovery is lagging behind on the changes in soil solution chemistry, as indicated by the stable but unbalanced tree mineral nutrition. Acidification and eutrophication will likely continue to produce after-effects for many decades. The results from this thesis indicate that the Programmatic Approach to Nitrogen (PAS) is partly missing its target for oxidized N compounds and that extra measures will be necessary to bring NOx emissions at an acceptable level.