Characterization of nitrogen deposition by means of atmospheric biomonitors

An increase of nitrogen deposition resulting from human activities is not only a major threat for global biodiversity, but also for human health, especially in highly populated regions. It is thus important and in some instances legally mandated to monitor reactive nitrogen species in the atmosphere. The utilization of widely distributed biological species suitable for biomonitoring may be a good alternative. We assessed the suitability of an ensemble of atmospheric biomonitors of nitrogen deposition by means of an extensive sampling of a lichen, two mosses, and a bromeliad throughout the Valley of Mexico, whose population reaches 30 million, and subsequent measurements of nitrogen metabolism parameters. In all cases we found significant responses of nitrogen content, C:N ratio and the δ15N to season and site. In turn, the δ15N for the mosses responded linearly to the wet deposition. Also, the nitrogen content (R2 = 0.7), the C:N ratio (R2 = 0.6), and δ15N (R2 = 0.5) for the bromeliad had a linear response to NOx. However, the bromeliad was not found in sites with NOx concentrations exceeding 80 ppb, apparently of as a consequence of exceeding nitrogen. These biomonitors can be utilized in tandem to determine the status of atmospheric nitrogenous pollution in regions without monitoring networks for avoiding health problems for ecosystems and humans.


Introduction 4 4
Nitrogen deposition is one the most predominant forms of atmospheric pollution 1 . This 4 5 phenomenon results from the release of nitrogenous compounds to the atmosphere, both in 4 6 cities, and in the country, including oxidized (NOx) and reduced (NHx) species, which are 4 The δ 15 N ranged from -6.6 to 7.8‰. We found the most negative value in a semi- found between rural areas and the urban ones. In contrast, the season showed no effect on 1 4 5 the isotopic composition of this moss (Table 1). We found a weak relationship between the 1 4 6 rate of nitrogen deposition and the δ 15 N values (R 2 = 0.2; Figure 5). The nitrogen content for Tillandsia recurvata ranged from 0.8 to 3.6% during the ppb, contrasting with the lowest nitrogen content that reached 0.8% in a rural site where 1 5 7 NOx presumably was lower to 5 ppb, based on the lowest recorded concentration in 1 5 8 Mexico City. Wet deposition had no effect on the C:N ratio of this bromeliad, but it was   Figure 6c), while the wet deposition had no effect on δ 15 N values (R 2 =0.01). the data from 1 6 6 two sites (Figure 6c open circles) were excluded from the analysis because topographic and 1 6 7 exposure to pollutants were non-representative, clearly skewed the isotopic signature of is crossed by a stream that was visible polluted. In turn, the second site were from a small  The distribution of nitrogenous pollution and deposition in Mexico City generally In this case, NHx originated from both agricultural and industrial activities are measured  The nitrogen content of lichens has been utilized as an effective indicator of

31
.5 kg N ha -1 year -1 above which they could not take up additional nitrogen. We observed has been already observed in other species of lichen 21 . Their isotopic composition 1 9 3 responded more to wet than dry deposition as was observed for mosses. Additionally,  The nitrogen content of moss tissues is affected by the rate of deposition and 1 9 6 responds to the distance to urban centers 14,22-24 . We also observed an important influence  NHx or NOx,), being more commonly negative in mosses from rural areas 25-28 . The suggesting that they more likely take up NO 3 derived from NO x of fossil fuel burning and influencing the isotopic composition of these mosses is the gas-particle conversion process, Mosses take up NH 4 + preferentially over NO 3 because less energy is needed in its 2 1 0 assimilation 32 . Additionally, high deposition rates can cause the inhibition of nitrate atmosphere, the nitrogen content of the mosses can increase more than from wet deposition.

3
However, in Mexico City the prevalent nitrogenous gas pollutant is NOx, suggesting that 2 2 4 nitrogen content and the δ 15 N of the mosses responded more to wet deposition than gaseous 2 2 5 pollutants 27,38 .

6
Despite the high rates of wet deposition recorded in Mexico City which exceeded 50 2 2 7 Kg N ha -1 year -1 in some areas, neither the nitrogen content nor the C:N ratio of Tillandsia 2 2 8 recurvata were directly affected. This occurs because raindrops cannot be absorbed by the bromeliad which grows after the rainy season. Indeed, the NOx were dragged from the 2 3 7 atmosphere to the ground surface during rainy season reducing its biological availability Both the nitrogen content and the isotopic composition of T. recurvata were was not found in sites with concentrations of NOx higher to 80 ppb. inherently limited by their small area to volume ratio that in turn reduces water loss 47 .

5 7
Also, their rates for taking up water are low, so exposure to more benign environments does to consider are pathogens and herbivores 49 . For these reasons, specific biomonitors need to  Especially in mid-sized cities and surrounding areas where the saturation thresholds of 2 7 5 these organisms has not been reached. Finally, the utilization in tandem of these organisms can inform an early alert for avoiding health problems for both ecosystems and humans. The Mexico City environmental authority has deployed an air quality network of 16 2 9 1 monitoring stations for wet deposition (Fig. 1). This network collects data during the rainy We selected four species of three types of atmospheric organisms, the lichen Anaptychia 3 0 5 sp., the mosses Grimmia sp. and Fabronia sp., and the bromeliad Tillandsia recurvata (L).  The tissue samples were dried at 60°C in a gravity convection oven until reaching  reported in parts per thousand, were calculated relative to atmospheric air standards. The analytical precision for the δ 15 N was 0.3 ± 0.07‰ (SD). The natural abundances of 15 N 3 2 4 were calculated as: where, R is the ratio of 15 N/ 14 N for nitrogen isotope abundance for a given sample 53,54 .

2 8
We utilized linear regressions to determine the relationship between total wet 3 2 9 nitrogen deposition and the nitrogen content (% dry weight), the C:N ratio as well as the NOx. Data accomplished normality in both cases 30 . We performed two way ANOVAs  We utilized the ordinary Kriging method (a geostatistical gridding tool for 3 3 9 irregularly spaced data 55 ) to determine the rates of wet deposition, and the concentration of  The datasets generated and/or analyzed during the current study are available from the 3 4 7 corresponding author on reasonable request.