Effects of
15N deposition on long-term
15N retention in two
gleysol-dominated coniferous forest areas in the Alptal
Valley in Switzerland.
BSc Project – Earth Sciences
Alptal Valley, Switzerland
University of Amsterdam
Author:
Lukas Oosterbaan
First supervisor:
Liz Veerman
Second supervisor: Albert Tietema
Table of Contents
Abstract……….………3
1. Introduction………..4
2. Methodology………6
2.1 Study area and design…….………..……….………..6
2.2 Laboratory work ….………..………..….7 2.3 Data Analysis...……….………..……….7 3. Results……….……….8 4. Discussion………12 5. Conclusions……….13 Recommendations………..13 6. References……….……….14 7. Appendices……….15
Abstract
Scientific summary
The abundance and distribution of 15N in the soil after long-term retention in a study area, located in the Alptal Valley in Switzerland, is investigated in this research. The relevance mirrors in the fact that nitrogen (N) is a key element in many natural systems, but in the last decades the emissions have increased in a significant way due to anthropogenic influences such as agriculture and the combustion of fossil fuels. It is important to create a better understanding of the processes and flows of N to be able to prevent potential impacts of increased reactive N emissions. Due to the strong correlation between N uptake and carbon (C) sequestration in the soil it could also involve the potential of facilitating climate change mitigation, since nitrogen in the soil could slow down the rate of CO2 increase. The two forest catchments in the study area were separately labeled with pulses of ammonium (15NH4) and nitrate (15NO3) 20 years ago and one is extra fertilized with 14N. Samples were
collected this year to be analyzed in the lab to see whether there are significant differences between the 15N retention of the different soil horizons (LFh, A and B) after different time periods and if fertilization and vegetation cover has significant influence on the 15N retention. The results showed that on the short-term, the vegetation and organic layer will be the primary N sink, which can be better explained by the chemical complexity of plant and microbial residues that stimulates retarded microbial degradation. On the long-term, a slight transition to the mineral soil is present and this acts as a sink of the deposited N due to the presence of associations of minerals and 15N containing organic matter, which brings stability there. Fertilization had a negative influence on the 15N retention in the top layers, because of 14N uptake by plants, which stimulates an increase of 15N retention in the mineral layers. Vegetation seemed to have no significant influence on the recovery rate of 15N in the soil.
Summary for the general public
Long-term 15N retention in the soil of two coniferous forest catchments, which were labeled 20 years ago, in the Alptal Valley in Switzerland, is investigated here. Since nitrogen (N) is a key element controlling the species composition, diversity, dynamics, and functioning of many terrestrial, freshwater, and marine ecosystems, it is relevant to obtain more knowledge of all the processes and flows of N. It appeared that 15N retention makes a transition from the organic layer to deeper, mineral layers after long-term retention due to stabilization processes. Fertilization had significant influence and altered the pathway of 15N in the short- and long-term. Vegetation had no significant influence on the 15N retention.
1. Introduction
Nitrogen is a key element controlling the species composition, diversity, dynamics, and functioning of many terrestrial, freshwater, and marine ecosystems. Increased anthropogenic influences such as agriculture, combustion of fossil fuels and other human activities will cause higher emissions of nitrogen (N). This results in a substantial alteration of the global cycle of N, which positively affects the availability and the mobility of N in many areas on earth (Vitousek, P. M. et al., 1997). It is important to create a better understanding of the processes and flows of N to be able to prevent potential impacts from N enrichment such as eutrophication, acidification and loss of biodiversity (Bingham et al., 2015). Nadelhoffer et al. (1999) even stated that humans have altered this cycle in such a way that more atmospheric N2 is being fixed into biologically reactive forms by anthropogenic
activities than by all natural processes combined.
This increase of reactive N has many consequences and one affects every single human being on this planet, since it has the effect of increasing the amount of N2O, which is a well-known greenhouse gas
and this will stimulate climate change.
Besides, it influences the amount of carbon (C) uptake in an area. Although regional climate changes can have both positive and negative effects on carbon uptake, recent modeling studies point to reduction of global land uptake in the future primarily because plant and soil respiration increase
with elevated temperatures (Churkina et al., 2009).
However, due to the fact that temperate forests are often limited in the amount of nitrogen they can uptake (Nadelhoffer et al., 1999), N inputs from atmospheric deposition can increase C storage by increasing primary production and C storage within plants. In soils, N additions can increase C storage in some ecosystems by reducing decomposition and respiratory C loss (Templer et al., 2012). This creates an alternating effect on the carbon-cycle (C-cycle), because an accumulation of forest
biomass is stimulated (Nadelhoffer et al., 1999).
So due to the strong correlation between nitrogen and carbon, an increased nitrogen deposition stimulates an increase in carbon sequestration of forests, which therefore is expected to partially compensate for a potential increasing CO2 level in the atmosphere (Oren et al., 2001). More
specifically, the way in which the C-cycle will be altered by an increased deposition of N can be determined by looking at the N retention and the associated C sequestration in the area. Templer et al. (2012) concluded that growth enhancement and potential for increased C storage in aboveground biomass from atmospheric N deposition is likely to be modest, because the largest sinks for 15N were belowground.
Effects of anthropogenic nitrogen deposition and the ability of terrestrial ecosystems to store carbon depends in part on the amount of N retained in the system and its partitioning among plant and soil pools (Templer et al., 2012). So part of the relevance of the research involves the potential of nitrogen in the soil slowing down the rate of CO2 increase and facilitates climate change mitigation (Churkina et al., 2009).
A way to test the correlation between N and C has been introduced by several studies by manipulating the N deposition of forest ecosystems by adding 15N and see how the environment responses. This is done to evaluate the complex interactions of processes in the N-cycle and to measure N-cycling within these ecosystems (Providoli et al., 2005). The nitrogen that is deposited in forms of NO2 and NH3 could react with hydrogen or oxygen to exist as ammonium (15NH4) or nitrate (15NO3) in the soil. This could end up in different layers and the recovery after 20 years per layer depends on many factors. It could behave as a major N sink after short-term retention and have close to zero N recovery after long-term retention, because much N might be taken up by trees or microbes (Currie et al., 2004) or nitrified and leached out as NO3 out of the system. N retention in
these systems almost only occurs through biotic processes, with the exception of cation exchange retention of ammonium. Biotic uptake of mineral nitrogen or through the incorporation of nitrogen into decaying soil organic matter will determine the N retention for a great part (Aber et al., 1989). Kögel-Knabner et al. (2008) stated that correlations between grain size and organic carbon (OC) content have been ascribed to stabilization of organic matter (OM) in soils by close associations
between OM and mineral surfaces, which is more likely after long-term retention. The total mineral surface area can be seen as a predictor for the amount of OM that is stabilized in a soil and after this stabilization the decomposition processes for OM are retarded. This means that the interaction of
the OC with the mineral surfaces influences the potential amount of C that is stored in a soil (
Kögel-Knabner et al., 2008). This also accounts for the adsorption of nitrogenous compounds to mineral surfaces, of which Kleber et al. (2007) suggests to be even more preferential. Many European sites were manipulated with 15N and this project is called NITREX (see figure 1 for other European locations of this project). On the
short-term (within the first year) all sites of the project showed an immediate response of nitrate leaching and in the coniferous forests the responses of vegetation and soil were delayed. However, this study will be focusing on the Alptal site in Switzerland, which is part of NITREX and represents N-limited temperate forests and mountain ecosystems (Providoli et al., 2005). A lot of interesting data can be derived from these tests but here the focus will be on the long-term retention (> 20 years) of this deposited
15
N and look at a possible change from the organic layer to the mineral soil as major N sink. More specifically, it is interesting to look at whether the long-term retention of
15
N at the sample site in Switzerland is a widespread phenomenon or not, since many different factors can determine the amount of nitrogen retained in a terrestrial area. The NITREX sites differ in climate, vegetation and soil
type, which might affect long-term N retention due to differences in soil moisture, texture, pH, clay types, CEC and temperature. But also local differences within one site and between different soil horizons can be of influence on the amount of 15N retained in the soil and thus storing of C. Despite considerable information on various aspects of the nitrogen cycle on the Alptal site, a lot about the nitrogen accumulation in the ecosystems is still unclear and a deeper understanding is needed of the single processes and flows (Providoli et al., 2005).
1.1 Objective and Hypotheses
The aim of this study is to investigate the effects of 15N deposition on long-term 15N retention in soil profiles of two coniferous forest catchments in the Alptal valley, 20 years after 15N label application, and in what way vegetation or additional 14N application can influence this. A comparison will be made between the 15N concentrations in the various soil horizons.
The first hypothesis is that the major 15N sink will be in the mineral soil instead of the more upper layers after long-term retention, due to associations of minerals and 15N containing organic matter. This is based on the unpublished results in Ysselsteyn (one of the NITREX locations), where the highest 15N concentrations were found in the deeper part of the soil, which consists of minerals (Tietema et al., unpublished results).
Secondly, there is a hypothesis that the difference in treatment (fertilized and labeled versus non-fertilized and labeled) has a significant effect on the 15N retention in the soil, due to a decrease of the
15
N/14N ratio.
Lastly, the hypothesis is made that the vegetation cover influences the amount of 15N that is retained in the deeper part of the soil, because it has a stabilizing effect on the soil and keeps the
nitrogen in the more upper parts. Besides, plant vegetation could take up 15N that leaches out from other areas, which stimulates the N retention in the LFh-horizon (Kronzucker et al., 1997).
2. Methodology 2.1 Study area and design
The study area, the Alptal valley, is located in the northern edge of the Alps of central Switzerland. The climate is cool and wet and the soils are often covered with snow. The soil profile generally consists of a LFh (layer of raw humus), an A and a B horizon and the major soil types are heavy gleysols (Providoli et al., 2005). The variability in the plot is related to differences in vegetation cover and slope angles.
To be able to conduct this research, samples are required from the study area to obtain knowledge of the concentrations of 15N in the different soil horizons after long-term retention. A few days in Switzerland was enough time to collect the samples from various spots in the coniferous forest of the Apltal valley. The area consists of two forest catchments (see fig. 2), which were both labeled with
15 N and catchment two is fertilized with additional 14N. Frequent small additions (approx. 200 per year) were applied that resulted in a deposition increase of 30 NH4NO3 kg per hectare per year (Schleppi et al., 1999).
Figure 2: Experimental set-up of the forest catchments (1 and 2) in Alptal, Switzerland (Providoli et al., 2005).
Catchment 1 was first labeled with 0,17 mm/m2 K15NO3 in 2000, and then with 0,7 mmol/m2 15NH4Cl
in 2002, after a one-year chase period. This second label was stronger to mask the residual effects of the first application in 2000 (Providoli et al., 2005). The second catchment, 2, was first and only labeled with 219 mmol/m2 15NH4 and 15NO3, in 1995
(Schleppi et al., 1999). Due to the impermeable gleyic sub-soil the water budget in this area is approximately balanced. Both these catchments were never artificially fertilized before (Providoli et al., 2005). Although 15N tracer applications alone involves N additions, this was not
considered to be fertilization since 15N tracer masses are very small compared to the other fertilized catchment.
The catchments have grids of 8 x 8 m existing of 20 and 24 cores of 25 cm deep that will be separated in the lab based on an LFh (raw humus), an A and a B horizon.
Soil samples were taken with the help of a soil corer (5 cm inner diameter, 25 cm depth), filled with a plastic cartridge that traps the sample. The corer is put in
Figure 3: a tilted soil corer in action
the soil straight and can reach into the B-horizon (Providoli et al., 2005). Once it is pulled out it should be tilted slightly to make sure nothing falls out of the filled cartridge (see fig. 3). Samples were transported and stored cool (at 2 oC) at the University of Amsterdam.
The data of the vegetation cover was already present, which was retrieved in the previous years by other studies. The means were taken of the pooled samples and this data ranges from 0 to 1, meaning respectively no vegetation to a fully vegetated grid.
2.2 Laboratory work
Once the samples reached the laboratory back at Science Park in Amsterdam these 44 cores were analyzed and separated in the three horizons (LFh, A, B) and pooled in groups of 2 or 3 cores, based on resemblances in humus type (anmoor, other anmoor, intermediate and mor), resulting in 8 cores per catchment. Pooled samples were dried in an oven at 40 oC, after which the mineral layers were crushed and sieved through a 2 mm sieve. This was done to prepare the samples to go into the ball mill machines (horizons A and B) or on the centrifuge grinder (LFh-horizon). After each sample very proper cleaning was required to make sure the samples could not contaminate each other.
Moisture content was also determined by drying a small part of each sample at 105 oC (A- and B-horizon) and at 70 oC (LFh-horizon).
Subsequently, the samples were put in a machine that measures the N and 15N concentration in all the different horizons. This is done with a new machine that provides a significant improvement over conventional extraction methods for the determination of isotope ratios in geological materials, called the elemental analyzer-isotopic ratio mass spectrometer, or EA-IRMS. This technique makes the sample that is folded into a small tin capsule undergo combustion in an oxygen atmosphere for an analysis of carbon and nitrogen (Grassineau, N. V., 2006). Only the latter is important for this research. The folding of the samples into the tin capsules (10 mg for LFh; 40 mg for A and B) was done manually to use a minimal amount of tin they were folded in triplicate to obtain more reliable results.
2.3 Data analysis
Calculations
To be able to compare the different catchments with each other, this can best be done in terms of recovery rates. This is necessary, because it makes it independent on the amount of tracer that was applied (Providoli et al., 2005). The steps and formulae used for the calculation are explained below. These rates can be achieved by starting with the δ15N ‰, which represents the 15N/14N ratio of the sample compared to the 15N/14N ratio of the air, written in the formula:
δ15N = Rsample / Rstandard – 1 (1)
The natural abundance of 14N in the atmosphere is used as a standard with Rstandard = 0.0036765. Then, the molar fraction of the sample can be calculated as follows:
Rsample = (δ15N + 1) ⋅ Rstandard = 15N/14N (2)
When this is known, the fractional abundance of 15N in the sample can be derived from:
Fsample = Rsample / (Rsample + 1) = 15N / (15N + 14N) (3)
With the fractional abundance Xsample can be calculated. This is defined as the molar ratio of the tracer N that is applied to the total N content of a pool, given in the formula:
Freference is the fractional abundance of the pre- or non-labeled sample and Ftracer is the fractional abundance of the tracer that is applied, which differs for the two catchments.
Finally, calculations if the tracer recovery (%) can be done by calculating the Zpool:
Zpool = Xsample ⋅ Npool / Ntracer (5)
This represents the tracer recovery and is a fraction of added tracer. It is usually expressed in percentages [%], so multiplied by 100. Npool and Ntracer are masses of respectively the total Npool in the sample and applied tracer in [mmol] per unit area [m2] (Providoli et al., 2005). The total Npool per sample is calculated by the several steps when having the moisture content of the samples. The complete calculation can be found in the appendix (1).
Boxplots were used to explore the data distribution and potential outliers.
Subsequently line plots, bar graphs and scatter plots were made to create a more visualized image, of which conclusions can be drawn. Within the catchments the comparison between the horizons is made in terms of δ15N (‰) and between the catchments the recovery rates (%) are used. Standard deviations are only taken into account for the data from 2016, and put in the graph as percent deviation of the δ15N (‰) per triplicate of the soil samples. This is also made accountable for the recovery rates.To see if the amount of vegetation cover is correlated with the 15N retention in the soil, a correlation command in MATLAB is used, which returns a corresponding Rho and P value. If P is very small, then the correlation RHO between the two arrays is significantly different from zero, saying there is a kind of correlation. The sample points are also visualized in a scatter plot to get a better idea of the potential correlation.
3. Results
Figure 4 on the next page shows the means and outliers of all the sample points per catchment per horizon. Most data points have values between the 0- and 100%, which are plausible outcomes, since the calculations give the percentages of 15N that is left of what is applied after long-term retention in the soil horizons. However, some values exceed the 100% and the 0% border. These plots show relatively higher values of recovery rate for the humus layer compared to the deeper layers, but it is not very clear. When time passes, a small switch of this balance to the deeper layers is noticeable. This, together with the influences of N-addition by fertilization, is better visible in the next figures, since bar graphs are better ways of comparing the different variables with each other and notice any potential transitions between them over time.
The same data is better visualized in figure 5. This graph can only be used for comparisons within a catchment (control and N-addition), because these show δ15N ‰. Very little δ15N ‰ was found in the control catchment in 2001, and more in 2009 and 2016, especially in the humus layer. Almost the same accounts for the catchment with extra N-addition, except for the high values in the LFh layer in 2001. Also, both catchments show a small transition from the upper layer to deeper layers over time.
To say something about the influences of fertilization with natural N to the 15N retention, the different catchments need to be compared in terms of recovery rates, as explained in the previous paragraph.
In figure 6 a graph is given that shows the two
catchments in all the different years. It is remarkable to see that the yellow bars, which represent the N-addition catchment, have relatively higher values than the control catchment, except for the LFh layer. Here the transition per horizon over time is also good visualized. For the years after 2001 the control catchment has higher rates of recovery in the top layer than the other catchment. It is the other way around in the mineral layers, and especially in the A layer. In the B-horizon the recovery rates are more or less the same. The standard deviations are shown with little error bars on the top of each bar. This is done to be aware of the margins of error.
Figure 4: Distribution of the recovery rates of all the sample points per catchment (Control & N-addition) per horizon (LFh – A – B).
Figure 5: mean δ15N ‰ of the different horizons over time in the control (top) and N-addition (bottom) catchment
Figure 7 shows the spatial distribution of the recovery rates (%) of all the pooled sample points in the two catchments, instead of taking the means and ignoring the data distribution. The more reddish the dots are, equals more 15N retention, which is mostly noticeable in the LFh-horizon of especially the control catchment. It is also clear that the N-addition catchment surpasses the control catchment in terms of recovery rate in the A-horizon. Samples 1 and 2, 3 and 4, 5 and 6 and 7 and 8 are pairs of two that share the same humus type anmoor, other anmoor, mor and intermediate respectively. Potential correlations between the recovery rates and humus types were hard to compute due to a limited amount of samples.
Then, scatter plots were made, shown on the next page in figure 8, to show potential correlations between the amount of vegetation cover on a grid and the 15N retention. The plots showed a bit chaotic results. The R (RHO) and P (PVAL) values are given next to the plots. As stated before, when P is very small (< 0.05), the correlation coefficient R between the two arrays is significantly different from zero, saying there is a correlation. However, these values seem high, which would indicate that there is no significant correlation between the variables.
Figure 6: bar graphs of the mean recovery rates of the different catchments in different years.
Figure 7: recovery rates of all 8 samples per catchment over depth.
Figure 8abc: scatter plots a (2001), b (2009) and c (2016) with the associated R and P values that show the relation between recovery rates (%) and vegetation cover (0-1) in different years and over the different horizons.
4. Discussion
This part will cover the potential discussions about the interpretation of the results of the data that is shown in the graphs. It includes the discussion about the outliers, the 15N retention on the short- and long-term and the influence of fertilization and vegetation cover.
Outliers
To begin with, it is strange that some of the results in terms of recovery rates, shown in figure 4, contained outliers that exceed the 100% border and others have negative values, which basically should not be possible due to the fact that you can simply not retain more 15N than what was applied or in the case of negative numbers, you can’t lose more 15N than what was applied. However, some of the grids of the sample points were located in a convex or concave, which represent the slope shapes, area. A possible explanation for the extra 15N that is found could be in the fact that extra tracer was leached from one grid to another one through erosion processes caused by precipitation events (Kreznor et al., 1989). This creates the possibility for one grid to retain more and for the other grid to lose more 15N than what was applied. The latter could occur, since all the results of δ15N values per sample were fixed for a reference year that involved the δ15N of the tracer that was applied before the last appliance. So when some of that old tracer is also leached out, this could result in negative numbers, because it was basically reset to zero 15N before the new appliance. 15
N retention on the short-term
The relatively high values of δ15N in the LFh-horizon, and especially on the short-term as seen in figure 5, can be explained by the high biological activity and biotic uptake that is present here and in the ground vegetation. This way mineral nitrogen becomes retained here, causing a progressive decline of the C:N ratio (Providoli et al., 2005). However, total ecosystem N retention can increase with mineral soil C:N, consistent with the idea that factors controlling microbial N uptake can shape the long-term sink for N in soils and thus total ecosystem N retention (Templer et al., 2012).
Also, the chemical complexity of plant and microbial residues or woody debris could be an important
15
N sink due to their potential to immobilize N during decomposition (Lamontagne et al., 2000 & Bingham et al., 2015).
The upper horizons of the control catchment in 2001 show very low and even negative numbers, because in this year in this catchment only very little tracer was added yet. A higher quantity was added in 2002 (Providoli et al., 2005). The other catchment was already labeled in 1995, and therefore has that high peak in 2001.
15
N retention on the long-term
Then, the slight transition to deeper layers can be ascribed to the fact that 15N containing organic matter can leach out of the organic layer and associate to mineral particles lower in the soil profile, called organo-mineral associations, which is more likely to happen after long-term retention ( Kögel-Knabner et al., 2008). This provides conditions there for higher stability. This is best shown in figure 5, which shows the mean δ15N over time.
This would be in line with the unpublished results in Ysselsteyn (Tietema et al., unpublished results). It would also be interesting for future studies to find out if the N retention in the mineral layer here especially occurs in the heavy fraction (>1.6 g cm-3), because here 15NH4 is possibly fixed in the
peripheral zone of clay minerals (Nieder et al., 2011). Influences of fertilization with 14N on the 15N retention
Then, the comparison between the catchments was made with the help of the recovery rates of 15N in the soil, visualized in figure 6. The control catchment has relatively higher 15N recovery rate in the LFh-horizon over the years compared to the fertilized catchment, if you neglect the results in 2001
(because little tracer was applied yet by then in control), which is in line with the hypothesis that it would. This is better explained by the fact that the 14N-addition to the area creates a lower 15N/14N ratio. 15N could directly leach out in forms of nitrate, without isotopic dilution due to exchange processes in the soil (Schleppi et al., 1999).
However, the A-horizon shows a significantly higher rate of recovery in the fertilized catchment. This increase of 15N recovery there might have to do with the fact that the 14N-addition is mainly taken up by plants and litter, with most tracer moving into aboveground biomass or to the mineral layers (Templer et al., 2012).
Correlation between vegetation cover and 15N retention
The plots showed a bit chaotic results, and not much could be concluded from this. However, with the help of the MATLAB command the correlation coefficients of the vegetation cover and the 15N retention of the pooled samples could be calculated. The values of R and P came out too low and too high, respectively, and varied greatly, which indicates that there is no significant correlation between the two variables. This means that vegetation cover has insignificant influence and that N settles
quickly in the soil.
Whether humus type had a correlation with the recovery rate of the sample on that location was hard to tell, due to a limited amount of samples, which would give unreliable results.
5. Conclusions
After all this research, including fieldwork, lab work and statistical analyses and further homework, some conclusions can be drawn.
This tracer study showed by various graphs of δ15N and recovery rates per catchment per horizon in different years that both catchments have the ability to retain most of the 15N, especially in the LFh-horizon. On the short-term most is retained in the top layers, because most biological activity and biotic uptake is present here. This makes a transition to the deeper layers, which contain minerals, on the longer-term. This can be ascribed to the fact that there organo-mineral associations, which is a chemically intimate binding, occur lower in the soil profile.
On the basis of figures 6 and 7 it can be concluded that the additional 14N in the fertilized catchment altered the pathway of the tracers containing 15N in the short- and long-term. It decreases the 15N retention in the humus layer and increases it in the mineral layers, especially in the A-horizon. The additional 14N is taken up by plants and litter in the top layer, which stimulates the 15N to move elsewhere, in this case to lower parts of the soil profile.
What also can be stated is that the vegetation cover is unrelated to the 15N retention in the soil in this ecosystem. 15N settles in the soil and is not significantly influenced by the vegetation cover.
Recommendations
Not a lot of information could be found about the influences of additional N by fertilization to the 15N retention in a soil. This study provides some new knowledge on this subject, but it asks for more research in the future to be able to make more trustworthy statements.
Besides, the used data for the vegetation cover might be out-dated and newly retrieved fresh data should be used in the future for another test. This could give better results.
Also, more samples should be available to see if there is a potential correlation between humus type and 15N retention.
6. References
Aber, J. D., Nadelhoffer, K. J., Steudler, P., & Melillo, J. M. (1989). Nitrogen saturation in northern forest ecosystems. BioScience, 39(6), 378-286.
Bingham, A. H., & Cotrufo, M. F. (2015). Organic nitrogen storage in mineral soil: implications for policy and management. Soil Discussions, 2, 587-618.
Churkina, G., Brovkin, V., Von Bloh, W., Trusilova, K., Jung, M., & Dentener, F. (2009). Synergy of rising nitrogen depositions and atmospheric CO2 on land carbon uptake moderately offsets global warming. Global Biogeochemical Cycles, 23(4).
Grassineau, N. V. (2006). High-precision EA-IRMS analysis of S and C isotopes in geological materials. Applied Geochemistry, 21(5), 756-765.
Kleber, M., Sollins, P., & Sutton, R. (2007). A conceptual model of organo-mineral interactions in soils: self-assembly of organic molecular fragments into zonal structures on mineral surfaces. Biogeochemistry, 85(1), 9-24.
Kögel-Knabner, I.; G. Guggenberger; M. Kleber; E. Kandeler; K. Kalbitz; S. Scheu; K. Eusterhues; P. Leinweber (2008) Organo-mineral associations in temperate soils: Integrating biology, mineralogy, and organic matter chemistry. Journal Plant Nutrition and Soil Science 171, 61-82. DOI: 10.1002/jpln.200700048.
Kreznor, W. R., Olson, K. R., Banwart, W. L., & Johnson, D. L. (1989). Soil, landscape, and erosion relationships in a northwest Illinois watershed. Soil Science Society of America Journal, 53(6), 1763-1771.
Kronzucker, H. J., Siddiqi, M. Y., & Glass, A. D. (1997). Conifer root discrimination against soil nitrate and the ecology of forest succession.Nature, 385(6611), 59-61.
Lamontagne, S., Schiff, S. L., & Elgood, R. J. (2000). Recovery of 15N-labelled nitrate applied to a small upland boreal forest catchment. Canadian Journal of Forest Research, 30(7), 1165-1177.
Nadelhoffer, K. J., Emmett, B. A., Gundersen, P., Kjønaas, O. J., Koopmans, C. J., Schleppi, P., ... & Wright, R. F. (1999). Nitrogen deposition makes a minor contribution to carbon sequestration in temperate forests. Nature,398(6723), 145-148.
Nieder R.; Benbi D.K.; Scherer H.W. (2011) Fixation and defixation of ammonium in soils: A review. Biol. Fert. Soils 47, 1–14.
Oren, R., Ellsworth, D. S., Johnsen, K. H., Phillips, N., Ewers, B. E., Maier, C., ... & Katul, G. G. (2001). Soil fertility limits carbon sequestration by forest ecosystems in a CO2-enriched atmosphere. Nature, 411(6836), 469-472.
Providoli, I., Bugmann, H., Siegwolf, R., Buchmann, N., & Schleppi, P. (2005). Flow of deposited inorganic N in two Gleysol-dominated mountain catchments traced with 15NO3− and 15NH4+. Biogeochemistry, 76(3), 453-475.
Schleppi, P., Bucher-Wallin, L., Siegwolf, R., Saurer, M., Muller, N., & Bucher, J. B. (1999). Simulation of increased nitrogen deposition to a montane forest ecosystem: partitioning of the added 15N. Water, Air, and Soil Pollution, 116(1-2), 129-134.
Templer, P. H., Mack, M. C., III, F. C., Christenson, L. M., Compton, J. E., Crook, H. D., ... & Emmett, B. A. (2012). Sinks for nitrogen inputs in terrestrial ecosystems: a meta-analysis of 15N tracer field studies. Ecology,93(8), 1816-1829.
Vitousek, P. M., Aber, J. D., Howarth, R. W., Likens, G. E., Matson, P. A., Schindler, D. W., & Tilman, D. G. (1997). Human alteration of the global nitrogen cycle: sources and consequences. Ecological applications, 7(3), 737-750.
7. Appendices
(1) Calculation of Npool:
First, the total mass of dry soil per sample needs to be calculated:
TotalSoilDry = TotalSoilMoist ⋅ (1 – MoistureContent) [g]
This is multiplied by the percentage of N content (which is given by the EA-IRMS) to get an idea of how much N it contains.
Ncontent = TotalSoilDry ⋅ (PercentN/100) [g]
This is again multiplied the molar mass of N to gain the results in moles. However, because the
15N/14N ratio is slightly different for each sample and the molar masses of 15N and 14N differ, the
molar mass ratio is also different. Thus this needs to be calculated for each sample, before the Ncontent can be defined in moles. Due to the fact that it is important that there is a common denominator in the calculations, it should be in mmol (just like the Ntracer). This is done by the following formula:
Gram2mmolratio = (1 ./ ((Fsample .* N15molarmass) + ((1 - Fsample) .* N14molarmass))) .* 1000
Then, to get the molar mass of N per sample:
Ncontent [mmol] = Ncontent [g] .* GramNtommolratio [mmol] Now the Ncontent is known in [mmol/horizon/sample]. We need to get the Ncontent in [mmol/m2]. This is reached by dividing this value in [mmol] with the surface area of the cartridge.
