Environmental Pollution 118 (2002) 259–272 www.elsevier.com/locate/envpol Atmospheric nitrogen deposition on the east coast of Spain: relevance of dry deposition in semi-arid Mediterranean regions M.J. Sanz*, A. Carratalá, C. Gimeno, M.M. Millán Fundación CEAM, Parque Tecnológico, c/Charles Darwin 14, 46980 Valencia, Spain Received 10 January 2001; accepted 31 August 2001 ‘‘Capsule’’: N deposition patterns on the east coast of Spain were characterized with special emphasis on dry deposition. Abstract Bulk deposition composition and pine branch washing were measured from April 1999 to March 2000 on the east coast of Spain. The main objective was to characterise N deposition patterns with special emphasis on dry deposition. Bulk deposition in the region is dominated by neutralisation processes by Ca2+ and HCO 3 , ClNa of marine origin and a high correlation between NO3 and 2 2 SO4 . SO4 concentrations show a decrease with respect to previous studies in the region in agreement with generalized sulfur emission decreases while the remaining ions, including NO 3 , are higher due to their general increase as well as to the inclusion of dry deposition in bulk collectors in the present study. An enrichment in NO 3 has been observed in dry deposition composition (branch washing) with respect to bulk deposition, while an impoverishment has been observed in the case of NH+ 4 . Annual bulk deposition varies between 7.22–3.1 and 3.5–1.8 Kg ha1 year1 for S- SO2 4 and N- NO3 , respectively. N total deposition goes from 9.78 to 6.8 Kg ha1 year1 at most stations, with the lowest deposition at the control station and Alcoi. The relative dry deposition with respect to the total was over 40% at most stations, going up to 75 % at the southern station. N-deposition is expected to be higher considering that N-NH+ 4 deposition has been underestimated in this study. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: Sulfate; Nitrogen; Bulk deposition; Dry deposition; Mediterranean 1. Introduction Recent emissions inventories have shown that N emissions are increasing in many areas of the world, suggesting that N saturation of terrestrial ecosystems may occur. It has been hypothesised that in the Western Mediterranean Basin the quantity of inorganic nitrogen deriving from the atmosphere may be much more important than estimated. With the growing industrialization and traffic increase in this region, some environmental problems have appeared which were unknown before the 1970s. Among them are the formation of ozone, nitric acid, nitrates and other nitrogenous compounds, including aerosols and particulate material, which generate from the emissions of NOx * Corresponding author. Tel.: +34-96-131-8227; fax: +34-96-1318190. E-mail address: mjose@ceam.es (M.J. Sanz). and organic vapors. For most of the year the regional climate is dominated by local atmospheric processes (sea breezes and up-slope winds), which cause the air mass to re-circulate over the same area and keep it much warmer than in other European regions (Millán et al., 1997). These conditions promote photochemistry and ozone formation, so that the values detected in this area greatly exceed European directives for vegetation and crop damage (Millán et al., 1992, 1997). Thus, to the list of environmental concerns in this region, from eutrophication of coastal areas and drinking water contamination to the problem of water pollution by nitrates employed in ‘‘intensive’’ agricultural practices (Mueller et al., 1995), we must now add the possible ‘‘extensive’’ problems arising from ozone damage and the deposition of nitrogenous compounds via the atmosphere. In Spain, the areas with the greatest identified problems, for both air pollution (Millán et al., 1992) and nitrate leaching produced by intensive agriculture practices (Varela, 1991), are the Mediterranean coastal fringe and parts of the central plateau. 0269-7491/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0269-7491(01)00318-9 260 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Nitrogen deposition in terrestrial ecosystems where photochemical pollutants dominate the pollution climate may be high or at least persistent. Thus, chronic nitrogen inputs are expected in remote sites, which can lead to various negative effects, including deficiencies, soil acidification, altered species composition, decreased mycorrhizal infection, increased susceptibility to environmental stresses, and elevated nitrate concentrations in soils, ground water and streams (Bytnerowicz and Fenn, 1996; Fenn et al., 1998; Fenn and Poth, 1999) in natural ecosystems and forests. Nevertheless, initial positive effects can be detected, such as enhanced tree growth when N deposition is not excessive. The other side of the problem, as earlier mentioned, is that intensive agriculture and the large N inputs result in heavy nitrate-leaching to the ground watersheds in some areas, such as the deltas and lower valley plains of the east coast of Spain where nitrate concentrations are very high (Varela, 1991; Ramos, 1996). Generally speaking, in Mediterranean Ecosystems the magnitude of the impact of nitrogen deposition and the susceptibility of specific ecosystems to dry deposition remain largely unknown. The greater magnitude of dry deposition is due to the temporal coincidence of high atmospheric pollutant levels and drought, a fact already evidenced in the Mediterranean ecosystems of California where N dry deposition can increase to 35–45 kg N ha1 year1 (Fenn, 1991; Bytnerowicz and Fenn, 1996). The objective of the present study is to evaluate atmospheric (Dissolved Inorganic Nitrogen, DIN) deposition on the east coast of Spain, where recirculatory processes dominate the air pollution dynamics, and dry deposition may be as high or even higher than wet deposition. Newly obtained data are compared to older data to see the time evolution, and sampling problems are also discussed. 2. Material and methods 2.1. Sampling network description In the Mediterranean Basin, the atmospheric air motions at the mesoscale level are strongly linked to topography and it is very well documented that recirculatory processes are present most of the year (Millán et al., 1997) and are responsible for the observed air pollution (Photochemical smog) patterns in the air quality networks (Seufert, 1997; Millán et al., 2000). Sampling stations are located along valley transects combined with north/south transects (Fig. 1) in order to characterize deposition patterns on the East coast of Spain. Co-ordinates and brief descriptions of each location can be found in Table 1. Sampling stations 4 and 7 are regular stations on the Valencian Community’s Air Quality Network, and station 5 is a MEDEFLU-Project site. 2.2. Sampling protocol 2.2.1. Bulk deposition To collect rain for chemical analysis, we used a 10-l bulk collector composed of a polypropylene bucket connected by a U-bend, to prevent evaporation, to a funnel with a 707 cm2 horizontal interception surface. The rim of the funnel was positioned at 1.5 m above ground level to avoid ground contamination. In addition, a device to prevent bird contamination was set up. Before collecting the sample, the funnel was rinsed with 300 ml of deionised water to wash away any matter deposited between the last rain episode and the collection time. Thus, such samples represent a combination of dry+wet deposition, with the dry component being mostly coarse particulate matter (Krupa, 2000). The sampling period was 2 weeks (14 days). At two sites, ‘‘El Saler’’ and Morella, precipitation amounts were continuously recorded by a precipitation gauge (ARG100, SKY), and most of the other sites were near National Institute of Meteorology precipitation rain gauges, which was useful to compare the precipitation volumes collected and the number and date of the rain events. Once in the lab, pH (Crison micropH 2002, Ag/AgCl electrode), conductivity (Crison microCM2201) and alcalinity (colorimetric tritiation) were measured for each composed sample. Afterwards, the samples were filtered with a Sartorious cellulose filter (0.45 mm pore size) and frozen (18 C) for a period of no longer than 2 weeks. 2.2.2. Branch washing Because of the inadequacy of any single method to evaluate dry deposition, we measured deposition to pine branchlets to be compared with the bulk deposition data. A leaf-washing technique developed by Lindberg et al. (1984) and modified by Bytnerowicz et al. (1987) was used. Aleppo pine (Pinus halepensis Miller) was selected as one of the most common conifer species on the East coast of Spain and in the Western Mediterranean Basin. Three 7-year-old potted trees supplied by the National Forestry Department (DGCN, Ministry of Environment) were placed on each sampling location. Four branches (5–8 cm long), one from each geographic quadrant, at about 1–1.5 m above ground level, were thoroughly washed at the beginning of the experiment and rinsed every 2 weeks with 50 ml of deioniseddistilled water dispersed from polypropylene spray bottles. Extended series of washes were done to check leaf efficiency for removing deposited ions, and it was found that the first 50 ml of wash removed more than 90 % of the deposited ions. The rinsing solutions were collected into a funnel in 50 cc polyethylene bottles and immediately placed in cool boxes at 4 C. Once in the lab, a composed sample of 200 ml was obtained for each tree, and the same analytical procedure as for bulk deposition samples was followed. M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 261 Fig. 1. Map of the locations. Total surface area exposed was calculated by measuring the two main axes of each needle, approaching half of the cylinder surface. The surface of the stems was calculated from the stem diameter and length measurements. 2.2.3. Surrogate surfaces Three Teflon filters (Whatman PTFE, 47 mm, 1 mm pore-size) were held in PVC double rings at a height of about 1.5 m above ground to avoid soil-dust contamination. Surfaces were exposed in a horizontal position. The filters were collected at the same time the conifer branches were rinsed. The exposed and blank filters were placed in 47 mm diameter Petri dishes (Poliestilen, Millipore) and additionally enclosed in zip plastic bags. Immediately after collection, bags were placed in cool-boxes filled with ice. Once in the lab, bags were kept at 18 C. Prior analysis filters were rinsed with 25 ml of deionised-distilled water (Milli-Q), shaken for 30 min and submitted to the same analytical procedure as the bulk deposition samples. 2.3. Analytical methods After samples were filtered and prior to freezing (18 C), the subsample for ammonium was acidified 262 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Table 1 Sampling stations (see footnotes for other available information)a Transect Co-ordinates and altitude (m) Stations and distance to the coast line (km) North-south coastal transect To characterise heterogeneity in the deposition along the coastal bend, background air mass at the mouth of the valley (0 90 5800 W, 38 580 22N)/12 (0 180 3900 W, 39 200 5000 N)/2 7. Gandı́a (Alicante)a 5. El Saler (Valencia)b,c (0 240 2300 E, 40 250 1000 N)/30 2. Benicarló (Castellón)/2.7b – 2. Benicarló (Castellón) (0 10 3600 E, 40 320 4900 N) / 700 (0 50 2800 , 40 380 1600 N)/1156 3. Vallivana (Castellón)/40b 4. Morella (Castellón)/51a – 5. El Saler (Valencia)/0.5 (0 260 3000 W, 39 170 1300 N)/32 6. La Peira (Valencia) 6.4a – 7. Gandı́a (Valencia)/3.2 (0 320 1700 W, 38 300 5800 N)/ 1014 8. Alcoi (Alicante)/29.6a (0 380 000 W, 38 280 2200 N)/503 9. Agost (Alicante)/17.4 (1 210 4600 W, 41 10 200 N)/895 1. Burbáguena (Teruel)/164 Valley transect I To characterise deposition along the sea breeze penetration path (Servol Valley). Gradients due to coastal cell penetration and deposition in areas influenced by reservoir layers (top of the mountain) Valley transect II To characterise deposition along sea breeze penetration path (Turia Valley) on the coastal plain of a wide valley with intensive agriculture Valley transect III To characterise deposition along sea breeze penetration path (Serpis Valley) Semi-arid area Southern semi-arid area where dry deposition is the most important component Control Non-influenced by coastal recirculatory cells a b c Temperature, speed and wind direction, distance to the sea, O3, NO, NO2, rainfall, RH. Temperature, speed and wind direction. Rainfall. with sulfuric acid and the one for cations with nitric acid, to prevent volatilisation and precipitation, respectively. Concentrations of anions were determined by ion chromatography (Dionex DX500 Ion Chromatograph, IonPac ASII 2mm and ASRS Ultra 2 mm suppressor column), concentrations of cations were determined by ICP (Optima 3000) and the concentration of ammonium was determined spectrophotometrically by the indophenol blue method (Eaton et al., 1998). 2.4. Attribution of marine source The major source for Chlorides is marine aerosols. Considering that there is no fractionating in sea salt aerosols and all chlorides in the samples come from marine origin ([Clm]=[Cl]), the percentage of marine source (%MS) at each station could be calculated using the marine ratios. In this case the ratios from the review of Keene et al. (1986; in meq), have been used in the following expression. 2þ %MS ¼ ½Cl m þ Naþ m þ SO2 4 m þ Mg m P P þ Ca2þ m þ ½K mÞ= ½cat þ ½ani ¼ð1 þ 0:86 þ 0:10 þ 0:21 þ 0:038 P P þ0:018Þ½Cl = ½cat þ ½ani : P P where ½cat þ ½ani are the total meq of cations plus anions and 1, 0.86, 0.1, 0.21, 0.038 and 0.018 are the of ratios the different ions to marine chlorides, i.e Naþ m =½Cl m ¼ 0:86. 2.5. Quality control/quality assurance The evaluation of the ionic balance and the estimation of conductivity were applied as quality checks for all samples: bulk deposition, branch washes and surrogate surface washes (Mosello et al., 1999). Na+/Cl ratios were also calculated. Assuming that sea salt is the dominant source of both ions, the ratio should resemble 263 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 the sea-water ratio, being close to 0.858 (Keene et al., 1986; Ivens, 1999). To assure reproducibility, 10 % of the samples were randomly selected and re-analyzed; 2 differences in the main anions (NO 3 , SO4 , Cl ) and cations were less than 10%. Ammonium differences ranged between 2 and 15 %. 3. Results 3.1. Average concentrations in bulk deposition and branch washing The volume-weighted means (VWM) for each location was calculated for bulk samples (Table 2). VWM for nitrates and sulfates in bulk samples is higher in sites with lower volumes of precipitation, with the highest being in Agost (106 and 144 me l1, respectively) and the lowest in Burbáguena (34 and 58 me l1, respectively). For ammonium the control site shows relatively high VWM values, although Agost presents the highest values. To compare with concentrations and ratios from this study, Table 3 shows previous data from studies on the chemical composition of rain at 27 locations in the Valencian Community in 1990–1991 (Carratalá, 1993), at other locations in the Mediterranean (Carratalá, 1993; Escarré et al., 1999) and in central-northern Europe (Pedersen et al., 1992). 4. Variations in dry deposition with respect to bulk deposition concentrations Table 4 presents the VWM of the dry deposition measured on pine needles during periods without rain. Its greatest interest is in verifying whether there is enrichment in any of the ions with respect to the bulk deposition, since in themselves the concentrations are dependent on the volume of wash used and on the area of the surface. Said volume was approximately the same in all the washes (200 ml), with surfaces of between 129 and 1470 cm2. The relation between both types of deposition was determined by using a linear adjustment between all the stations for each ion, and it is shown in Fig. 2 for Ca2+, 2 Na+, K+, NH+ and NO 4 , SO4 3 . To compare the results with those of bulk deposition, a comparison of the slopes of these linear adjustments was performed. In general, the best regressions are found for NO 3 and 2 + 2 SO2 (with r =0.7), followed by Na with r =0.6 4 + 2+ and by K and Ca . Finally, in the case of ammonium, no correlation is found between the dry and the bulk deposition concentrations. Worth noting is the fact that in the case of NH+ 4 , the concentration values in the washes are clearly lower (slope51). For the remaining ions, the slopes obtained are greater than 1, with the nitrates showing the maximum slope (3.2). The significance here lies not so much in the slope values as in the differences between them which are indicative of the enrichment or the impoverishment of some ions with respect to others in each type of deposition. Taking sulfate and calcium as references, both of which present a slope ffi 1, it can be seen that sodium and nitrate have bigger slopes, implying their enrichment in dry deposition with respect to bulk deposition. In the case of ammonium, the slope is less than 1 and thus shows an impoverishment. 5. Main sources of ions in bulk deposition and branch washing The percentage of marine sea salt (Tables 2 and 4, % MS) has been calculated using marine ratios as described in the Section 2. In bulk deposition it goes from 5% at station 1 (Burbáguena), the location farthest from the Table 2 Volume weighted mean (VWM in meq l1) for major ions in bulk deposition for this study (1999)a Site 1.BUR 2.BEN 3.VAL 4.MOR 5.SAL 6.PE 7.GAN 8.ALC 9.AGO Rainfall (mm) 382 530 570 436 464 361 369 238 184 Volume weighted mean in bulk deposition (meq l1) Ca2+ K+ Mg2+ Na NH+ 4 Cl NO 3 SO2 4 b SO2 4 nm HCO 3 184 286 189 203 212 379 342 440 483 6 10 3 7 11 11 12 11 14 14 34 17 18 46 46 64 41 38 20 103 35 34 178 127 195 94 92 55 48 38 48 33 34 28 40 80 12 95 29 30 168 139 198 76 114 34 44 44 51 40 58 52 57 106 58 70 63 75 72 104 122 81 144 56 60 59 72 54 89 101 73 132 145 191 122 156 156 303 219 336 381 %IMB %MS 5.67 8.96 4.11 0.63 4.60 0.62 4.01 6.30 2.70 5.0 24.0 12.0 10.8 40.6 25.7 35.6 14.4 17.4 P P P P The volume of rainfall collected is also presented (prec.). % IMB (% Imbalance) correspond to the ratio ( cat- ani)/( cat+ ani), and % MS (% marine ions) correspond to the sum of ions of marine origin using marine ratios from the literature (Keene et al., 1986). b SO2 4 nm correspond to non-marine sulphates. a 264 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Table 3 Volume weighted mean (VWM) precipitation chemistry (meq l1) in nine localities on the Mediterranean seaboard (1–9) and six localities from the European Network (EMEP; 10–15a) Reference 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 a Site Rainfall (mm year1) Morella (S) Vinaroz (S) Torrent (S) Gandia (S) Alcoi (S) Elx (S) Monseny (S) Prades (S) Almeŕa (S) Iraty (FR) Ispra (I) Birkenes (N) Witteveen (NL) Braganza (P) Kamernicki (Y) 677 628 555 829 522 186 876 551 675 668 1233 1861 844 420 5287.6 Volume weighted mean of rainfall samples (meq l1) Ca2+ K+ Mg2+ Na+ NH+ 4 Cl NO 3 SO2 4 127 226 212 144 320 201 57 59 123 21 17 7 19 34 82 12 12 26 18 22 10 4 9 15 1 2 4 – 4 19 16 48 48 58 29 38 10 11 34 8 4 17 25 11 26 23 109 95 137 69 60 23 26 37 13 7 76 113 12 53 42 38 75 40 41 51 23 29 32 57 71 33 139 18 87 34 134 123 180 87 75 29 30 34 13 10 91 132 11 69 24 34 34 33 32 38 21 23 19 22 62 34 51 10 36 75 118 105 90 122 99 46 56 58 23 35 27 87 25 75 pH HCO 3 (meq l1) 5.7 5.8 6.0 5.3 5.7 6.2 4.8 4.9 5.7 5.1 4.3 4.4 4.9 5.4 5.1 133 163 235 126 296 186 14 23 – – – – – – – Source: 1–6 (Carratalá, 1993), 7–9 (Escarré et al., 1999) and 10–15 (Pedersen et al., 1992). Table 4 Volume weighted mean (VWM in meq l1) for major ions in pine branch washing for this study (1999)a Site Volume weighted mean of branch washing samples (meq l1) 2+ 1. BUR 2. BEN 3. VAL 4. MOR 5. SAL 6. PE 7. GAN 8. ALC 9. AGO + Ca K 69 359 128 79 164 294 516 141 526 5 12 7 6 10 13 25 8 17 2+ + Mg Na 7 41 19 12 45 42 149 14 66 15 137 57 34 143 113 508 33 201 NH+ 4 13 12 9 7 10 14 13 9 12 Cl NO 3 SO2 4 b SO2 4 nm HCO 3 12 133 30 23 151 121 556 37 179 22 97 60 36 69 77 187 47 275 12 67 19 17 40 56 166 20 118 10 53 16 14 24 43 107 16 98 54 210 87 57 97 162 169 95 226 %IMB %MS 5.36 4.53 5.49 1.17 1.93 6.47 5.57 0.38 1.26 12.1 27.6 16.0 18.6 45.7 30.0 53.7 20.3 24.5 a Also presented is the, % IMB (% imbalance) corresponding to the ratio (catani)/ (cat+ani), % marine ions corresponding to the sum of ions of marine origin using marine ratios from the literature (Keene et al. 1986). b SO2 4 nm mm correspond to non-marine sulphates. seashore, to 40.6% at station 5 (Saler), the closest to the seashore. In branch washing the %MS were slightly higher, going from 12% in Burbáguena to 53% in Gandı́a. The relationship between marine source and distance to the sea can be seen in Fig. 3 and follows a clear potential gradient from seashore high values towards inland low values. Typical ions of terrestrial origin are associated with CO3Ca. Several ions such as Mg2+, K+, and SO2 4 also have a certain terrestrial origin. The amount of calcium and carbonates in this study is quite large. There is a + strong correlation between Ca2+ and HCO 3 and K 2 (r =0.98 and 0.80, respectively). These two ions (Ca2+ and HCO 3 ) show a strong negative correlation with volume (0.80 and 0.87, respectively) indicating their predominantly local origin. They are washed in the first fractions of rain and then diluted. A quite strong correlation has also been found between Ca2+ and 2 SO2 and NO 4 3 (r =0.79), as a consequence of neutralisation processes. The strong correlation between nitrates and sulfates of non-marine origin (r2=0.8 for bulk samples and 0.9 for branch washing samples), shown in Fig. 4, indicates their common anthropogenic origin. The differences between the two samples (bulk and branch washing) is that nitrates are enriched with respect to sulfates in branch washing. Ammonium shows a certain correlation with nitrates only in bulk samples (r2=0.5) and no correlation with the remaining ions from bulk and branch washing. M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 265 2 Fig. 2. Relationship between the VWM concentrations of Ca2+, Na+, K+, NH+ 4 , SO4 and NO3 in both types of deposition (bulk deposition and pine branches washing ) calculated by using a linear adjustment of all the stations in the study. The different slopes indicate the preferential way of deposition strategy for each ion. Fig. 3. Relationship between marine source (represented by Cl concentration) and distance to the sea (Km). Branch washing (p) and bulk samples (pl) shows the same potential relationship. Higher concentrations in branch washing are an effect of washing volume. 6. Deposition of sulfate and nitrogenous compounds 6.1. Bulk deposition The bulk deposition data for S-SO2 4 , N-NO3 and + 1 1 N-NH4 in kg ha year corresponding to this study are shown in Table 5. In order to compare them with previous data, Table 6 shows data corresponding to Mediterranean and European stations, taken from the literature. The annual precipitation at each locality is also included. In the case of S-SO2 4 , deposition is around 5.5 kg ha1 year1, oscillating between 7.22 kg ha1 year1 at Gandı́a and 3.1–3.5 kg ha1 year1 at Alcoi and Burbáguena. The lowest values are found at the control station and at the Alcoi station. N-NO 3 deposition oscillates between 3.5 kg ha1 year1 at Vallivana and 1.8–1.9 kg ha1 year1 at Burbáguena and Alcoi. The contributions from bulk deposition are quite homogeneous by valleys (Table 5), as is the precipitation (Table 2). It is interesting to note that the Burbáguena station (control station), while showing precipitation levels on the order of the El Saler and La Peira stations, presents a lower deposition, pointing to the lower average NO 3 levels in the area. With respect to N-NH+ 4 deposition, the behavior is similar to that of N-NO 3 , excepting the fact that for ammonium, the Burbáguena station does not behave like a background station. With this exception the levels oscillate around 3 kg N-NH+ ha1 year1 in the northern 4 1 valley and from 2 to 1 kg N-NH+ year1 in the 4 ha other valleys. Comparing our results with the values in Table 6, we find similar results in Spanish and European Mediterranean stations with similar environments and precipitation volumes. The stations with the highest ammonium deposition are those located in areas with high ammonium emissions and high annual precipitation, for example sites in the Netherlands (Table 6). 266 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 6.2. Dry deposition over Pinus halepensis leaves Dry-deposition contributions are calculated from the 14-day measurement period of deposition to pine foliar surfaces and taking into account the days of exposition since the last day of rain within the fortnight. A rain event is considered to wash the pine leaves completely, which means a reset to 0 in terms of deposition. In this way, the dry deposition only accounts for several dry periods of 14 days and several periods with 1–14 days of dry deposition. The days prior to the rain event are lost; thus it must be kept in mind that branch washing underestimates the annual dry deposition when rain events are present within a collection period. Table 5 also shows the results obtained for dry deposition measured in the yearly period considered in the present study (drym). As has been mentioned, the 2 1 Fig. 4. Relationship between NO 3 and SO4 (meq l ) in both types of deposition (bulk deposition, pl) and pine branches washing, p). AH high correlation indicates their common origin (anthropogenic). The higher slope in pine branch washing indicates that dry deposition is more effective than wet deposition for NO 3. dry deposition calculations use the data from the date on which the precipitations occurred, and this is obtained from the pluviometric data registered at the regional meteorological network stations located near the stations in this study. Concordances between precipitation volumes at homologous stations are shown in Table 7. The agreement observed is quite good (slope 1, r2=0.8), taking into account the great spatial heterogeneity in the precipitation of the region. Table 7 also shows the number of days with precipitations and the number of available days for estimating dry deposition to needle surfaces. The dry deposition measured for sulfates, including sulfates of marine origin, is seen to be quite variable and, in contrast to the bulk deposition, to follow no clear pattern. The lowest value is registered at the control station (Burbáguena, 0.16 kg ha1 year1) and the highest value is measured at the Gandı́a station (3.10 kg ha1 year1). In general, the levels are substantially lower than those obtained for bulk deposition; the percentages associated to dry deposition range from 5 to 37%. However, it should be pointed out that in Table 5 the annual dry deposition (drym) is underestimated and the bulk deposition includes dry deposition, although over a much less efficient collection surface. In fact, the dry deposition estimated over surrogate surfaces like teflon filters (data not shown) is also lower than for branches. The dry deposition of N-NO 3 oscillates between 5.17 kg ha1 year1 at the Agost station and 0.26 kg ha1 year1 at the control station (Burbáguena). In this case a greater homogeneity can be seen between the transects: from 1 to 1.6 kg ha1 year1 in the north followed by 1– 2.3 kg ha1 year1 in the Turia valley (middle transect); the southern transect shows the most heterogeneity, with Table 5 + + Annual deposition of S-SO2 4 , N-NO3 , N-NH4 and N-(NO3 +NH4 ) thorough bulk and dry (over pine branches) deposition measured in this a study Site Bulk deposition (kg ha1 year1) and dry deposition (kg ha1 year1) S-SO2 4 1. BUR 2. BEN 3. VAL 4. MOR 5. SAL 6. PE 7. GAN 8. ALC 9. AGO a N-NO 3 + N-(NO 3 +NH4 ) N-NH4 Bulk Drym Drye Bulk Drym Drye Bulk Drym Drye Bulk Drym Drye Bulk+Drye 3.52 5.98 5.71 5.23 5.33 5.99 7.22 3.10 4.24 0.16 1.32 0.38 0.66 0.73 1.95 3.10 0.57 2.46 0.29 2.37 0.76 1.56 1.27 2.79 5.00 0.91 3.33 1.81 3.30 3.52 3.09 2.59 2.93 2.69 1.89 2.73 0.26 1.61 1.04 1.27 1.14 2.29 2.98 0.93 5.17 0.48 2.89 2.09 3.01 1.97 3.28 4.81 1.49 6.99 2.94 3.54 3.07 2.93 2.14 1.72 1.46 1.35 0.94 0.16 0.18 0.14 0.23 0.15 0.43 0.25 0.21 0.28 0.31 0.32 0.29 0.53 0.26 0.61 0.40 0.33 0.37 4.75 6.83 6.59 6.02 5.66 4.65 4.15 3.24 3.67 0.42 1.79 1.19 1.49 1.30 2.72 3.23 1.13 5.44 0.78 3.20 2.38 3.55 2.23 3.89 5.21 1.82 7.37 5.5 10.0 9.0 9.6 7.9 8.5 9.4 5.1 11.0 Bulk deposition covers the period April 1999–March 2000 (less that 5% of samples have been lost). Dry deposition covers the same period; however, two values are presented for the annual dry deposition: Drym which only accounts for the deposition on the days after rainy events where dry deposition could have been measured (the exact number of days included in each station are in Table 6) and Drye which has been extrapolated to the total number of dry days in the yearly period. 267 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Table 6 + 1 Annual deposition of S-SO2 year1 from the literature at seven localities in Spain and seven from the EMEP networka 4 , N-NO3 , N-NH4 in kg ha Reference Site 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Rainfall (mm year1) Pyrenees (S) Montseny (S) Prades(S) Roquetas (S) Morella (S) Valencian C. (S) Almerı́a (S) Deuselbach (G) Iraty (F) Ispra (I) Birkenes (N) Witteveen (NL) Braganza (P). Kamernicki (Y). 1640 876 551 428 678 518 675 691 668 1233 1861 844 420 527 Deposition (kg ha1 year1) S-SO2 4 N-NO 3 N-NH+ 4 + N-( NO 3 + NH4 ) 13.4 6.4 4.8 3.3 8.1 12.3 6.3 2.5 2.5 6.9 8.0 11.7 1.7 6.3 4.1 2.6 1.7 2.8 2.3 3.1 1.8 2.9 2.1 10.7 8.9 6.0 0.6 2.7 5.1 2.8 2.3 3.2 4.0 4.1 3.0 3.2 5.3 12.3 8.6 16.4 1.1 6.4 9.2 5.4 4.0 5.9 6.3 7.2 4.8 6.1 7.4 23.0 17.5 22.5 1.6 9.1 a References: (1), Camarero and Catalan (1993); (2, 3 and 7), Rodà et al. (1993); (5 and 6), Carratalá et al. (1993); and (4, and 8–14), Pedersen et al. (1992). Table 7 Rainfall (mm) collected in the sampling period (1 year) in this project and at the closest station to a National Institute of Meteorology (INM) stationa Site 1. BUR 2. BEN 3. VAL 4. MOR 5. SAL 6. PE 7. GAN 8. ALC 9. AGO Rainfall (mm) Number of days Sampled INM Wet deposition Dry deposition 382 530b 570b 436b 464b 361 368b 238b 184 371 687 509 509 471 320 457 277 196 86 66 111 111 70 35 49 37 32 149 167 127 107 171 231 196 204 246 a Also presented are the number of days with rain and the number of the days in which it was possible to measure the dry deposition. b 1–2 fortnight periods have been lost. oscillations of between 2.98 kg ha1 year1 at Gandı́a and 0.93 kg ha1 year1 at the Alcoi station, which could be under special meteorological conditions. On the other hand, Agost presents the highest values (5.17 kg ha1 year1). The percentage of dry deposition with respect to bulk deposition is higher than in the case of the sulfates; it oscillates between 65% at the driest station and 12% at the control station. In the case of ammonium, very low deposition values were found in dry deposition: between 0.43 and 0.14 kg ha1 year1. Compared to the bulk deposition (0.94– 3.54 kg ha1 year1), ammonium values are considerably lower, in most cases below 10% of the bulk deposition values. 7. Estimation of annual dry deposition and comparison to annual bulk deposition As we stated above, the dry deposition data (drym) in Table 5 do not correspond to the set of annual dry days. This data was used to calculate the average daily deposition (Table 8), which, taking into account the annual number of dry days per station (365 minus the number of rainy days), was then used to obtain the annual dry deposition, drye (also shown in Table 5). These calculations show the largest deposition rates to 2 be for N-NO 3 , followed by S-SO4 , with ammonium levels very much lower. The calculated deposition rates between stations are relatively homogeneous; for total N they vary between 0.76 and 2.213 mg m2 day1. The minimum is found at the control station (Burbáguena) followed by Alcoi. The station showing the highest rates is Agost which has been shown to have the highest concentrations and the maximum number of days without rain. When annual dry deposition, drye, is calculated by applying the above-calculated rates, S-SO2 yearly 4 deposition ranges from 0.29 to 5 kg ha1 year1. The minimum appears at the control station and the maximum at Gandı́a. For N-NO 3 yearly deposition, the minimum also appears at the control site (0.48 kg ha1 year1), but the maximum is at Agost (6.99 kg ha1 year1). The northern valley transect shows small differences between stations from the seashore to the upper valley stations (0.92 kg ha1 year1). The Turia valley transect shows slightly larger differences even though the distance between the sites is smaller (1.21 kg ha1 year1). On the other hand, the southern valley transect shows even larger differences (3.32 kg ha1 268 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Table 8 Annual average of daily dry deposition rate of S-SO2 4 , N-NO3 , N + 2 1 and N-tot (N-NO + N-NH ) in mg m day calculated NH+ 4 3 4 from drym, and taking into account the days of dry deposition measured (in Table 7) Site 1. BUR 2. BEN 3. VAL 4. MOR 5. SAL 6. PE 7. GAN 8. ALC 9. AGO Daily deposition rate (mg m2 day1) S-SO2 4 N-NO 3 N-NH+ 4 N-tot 0.104 0.792 0.301 0.612 0.430 0.844 1.583 0.279 1.000 0.172 0.967 0.822 1.186 0.668 0.993 1.522 0.454 2.100 0.109 0.105 0.113 0.210 0.089 0.186 0.127 0.101 0.112 0.281 1.072 0.935 1.397 0.758 1.179 1.649 0.556 2.213 Table 9 Estimation of the percentages of wet and dry deposition of N-tot (N+ a NO 3 +N-NH4 ) considering dry deposition over pine branches Site Surface efficiency (%) Wet and total atmospheric deposition N (kg ha1 year1) 1. BUR 2. BEN 3. VAL 4. MOR 5. SAL 6. PE 7. GAN 8. ALC 9. AGO N-DB/DP N-WET N-TOT WET/TOT 82% 29% 44% 36% 47% 16% 23% 66% 17% 4.10 5.90 5.55 4.74 4.61 4.03 2.95 2.04 2.42 4.89 9.11 7.92 8.29 6.85 7.92 8.16 3.86 9.78 84% 65% 70% 57% 67% 51% 36% 53% 25% a Wet deposition has been calculated by subtracting dry deposition to bulk collectors, and dry deposition to bulk collectors has been estimated with the use of data from fortnight periods with no rain occurring. year1). Ammonium causes the total N (N-NO 3 + NNH+ 4 ) dry deposition to increase by a small fraction (0.26–0.61 kg ha1 year1). Thus, the total N (bulk plus dry, Table 5) deposition ranges from 5.53 kg ha1 year1 at the control station to 11.04 kg ha1 year1 at the Agost station. It should be noted that the total N deposition may be overestimated due to the inclusion of a small fraction of dry deposition in the wet fraction. In order to obtain an estimation of the total N deposition and of the fractions of wet and dry N deposition, we compared the deposition data from several (3–6) rainless fortnight periods and found that the N-NO 3 dry deposition to bulk collectors was 16– 82% of the deposition to pine branches (depending on the stations). No estimations were made for N-NH+ 4 due to the very low efficiency of the dry deposition. Taking these percentages into account (Table 9), the dry deposition to bulk collectors was discounted from the bulk deposition to obtain the corrected wet deposition (Table 9). Then the total deposition to pine leaf surface was calculated (see also Table 9). The final estimated values for total N deposition thus range from 3.86 to 9.78 kg ha1 year1, with the lowest at the Alcoi and the control station (3.86 and 4.89 kg ha1 year1 respectively) and the highest at the Agost station (9.78 kg ha1 year1). The wet/(dry+wet) ratio for N deposition ranges from 84% at the control station to 25% at the Agost station (Table 9), with 60% being the average for the northern transect and the Turia valley transect, and 44% for the southern transect, clearly linked to precipitation amount and number of wet days. Therefore, N dry deposition accounts for more than 40% of the deposition at most stations, increasing to 75% at the driest station. 8. Discussion In this work, higher ion concentrations in bulk samples than those reported by Carratalá (1993) are registered due to the fact that in the present study: (1) there was less rainfall, producing higher values of ions associated to the washout process (Ca2+, HCO 3 and to a 2 + lesser extent NO 3 , SO4 , NH4 ), (2) the dry deposition after the rain events was also collected by washing the collector funnel prior to the collection of the sample, and (3) the deposition may have increased since 1992. Thus, Burbáguena is confirmed as a background or control site because it shows the lowest values for all but ammonium whose levels are comparable to those of the EMEP network (Table 4). With respect to the 1990– 1991 period (Table 3), sulfate levels in the area are maintained or are slightly lower while nitrate levels are considerably higher. The general decrease in sulfate levels is in agreement with emission and overall deposition tendencies registered in recent years on the East coast of Spain (Avila, 1996; Escarré et al., 1999). On the other hand, NOx emissions are generally increasing in western Europe due to a sharp increase in gas and oil consumption (Dignon and Hameed, 1989; Pacyna et al., 2000) and, as a consequence, higher nitrate concentration is expected. In fact, long-series for overall deposition are showing a slight increase for nitrates in remote sites (Avila, 1996) with respect to previous studies in the area. However, it should be noted that the inclusion of dry deposition to the surface of the funnel resulted in a definite increase in levels, indicating a possible earlier underestimation of the dry deposition component in the bulk deposition calculated by other studies in the area (Carratalá, 1993). Regarding ammonium, there is no clearly observable tendency with respect to previous data, especially if we take into account: (1) that locations do not coincide M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 exactly, (2) the primary character of this pollutant with variations at the local level which may be important, and (3) the difficulties in evaluating possible losses due to volatilisation in our climate with long collection periods. Other studies have documented that the sampling period is especially important for ammonium: long sampling periods (week compared to event sampling) resulted in lower NH+ 4 concentrations and deposition values of about 11% (De Pena et al., 1984). The most significant differences in NH+ between weekly and 4 event samples were found by another group of authors in Chicago (Sisterson et al., 1985) who attributed them mostly to volatilisation, whereas researchers in Prague attributed the differences they found to bioconsumption (Vesely, 1990). Others concluded that biases are sitespecific (Ramundo and Seastedt, 1990). Thus, taking into account our climatic conditions and the high pH in the samples, the results of the present study suggest that an even higher re-volatilisation than that mentioned in the literature can be expected, and further experiments are being developed to assess the problem for rain collectors and long collection periods. Among the methods of surface analysis, dry deposition has usually been estimated on the basis of foliagerinsing (Bytnerowicz et al., 1987; Ross and Lindberg, 1994). It is important to note that branch washing results are difficult to compare with the results from throughfall studies because, in the latter, many layers of leaves collect the pollutants. Still, the branch-washing technique seems to be a simple and reliable method for evaluating atmospheric dry deposition to plant surfaces (Bytnerowicz et al., 1987; Bytnerowicz, 1994; Lindberg et al., 1984), despite the fact that this method is affected by: extraction of anions from foliage during washing, uptake from foliage during collection periods (up to 39%, Stachurski and Zimka, 2000), and re-volatilisation and re-suspension of deposition particles (Hosker and Lindberg, 1982). All of this may explain the even lower values for NH+ 4 found in the branch washing samples compared with the bulk collector. Nevertheless, branches continue to be more efficient than bulk collectors and surrogate Teflon surfaces (data not shown) for collecting nitrates (Tables 2, 4 and 9). Other authors also + reported higher values of NO 3 than NH4 (about 2 times more) for Ceanothus leaf rinse data in California (Riggan et al., 1985; Bytnerowicz et al., 1987). The complex structure of the collecting device—natural branch, branch coated with Teflon, artificial tree or artificial branch—seems to condition efficiency even more than the material itself. In fact, the Teflon-coated branches showed an identical amount of deposition ions to the uncoated branches (Bytnerowicz et al., 1991), demonstrating the usefulness of the method. When the composition of the rain samples is considered, relationships between ion concentrations are good indicators of their source (Losno et al., 1991; 269 Durand et al., 1992). The major ions are bicarbonate and calcium, followed, depending on their marine character, by chloride or sulfate and nitrate, and sodium or ammonium in the case of cations. Percentages found for MS are in agreement with those found in other coastal areas (Fujita and Kawaratani, 1988; Underwood et al., 1988) and with previous studies in the area (Carratalá, 1993). They also follow a clear gradient from seashore high values to inland low values, as was expected. Frequently, the anthropogenic origin of sulfates and nitrates is associated with the acidity of the rain samples (Zeng, 1989; Durand et al., 1992). However, the large amount of calcium and carbonates in this study, the strong positive correlation between them and with NO 3 and SO2 4 , and their negative correlation with precipitation volume which is in agreement with the type of sampling (bulk), the calcareous nature of the soils (Gatz et al., 1986; Khemani et al., 1987), and the high frequency of Saharan dust episodes (Avila, 1996; Carratalá et al., 1996; Loye-Pilot and Martin, 1996) suggest the existence of neutralisation processes occurring in the atmosphere or in the collector (Casado et al., 1992; Carratalá and Bellot, 1998). Acidity is thus unlikely to occur. Another possible source of sulfates in bulk samples is gypsum-rich soils, but none of our sampling stations show this type of soils. On the other hand, the strong correlation between nitrates and sulfates (r2=0.8 for bulk samples and 0.9 for branch washing samples) indicates their common anthropogenic origin. The difference between the two samples is that nitrates are enriched with respect to sulfates in branch washing (Fig. 3), indicating a higher collection efficiency. Ammonium shows a certain correlation with nitrates in bulk samples only (r2=0.5), and no correlation with the remaining ions from bulk and branch washing. This points again to the possible existence of a higher revolatisation than for bulk samples, or to an uptake and/ or re-volatilization of ammonium in branch washing samples. When deposition is calculated from the measured bulk values, S-SO2 4 deposition varies between 7.22 and 3.1 kg ha1 year1, showing a clear decrease when compared with data from the same sites published in previous studies (Carratalá, 1993). In the case of NNO 3 , bulk deposition varies between 3.5 and 1.8 kg ha1 year1. The contributions are quite homogeneous by zones and are linked to the precipitation in each zone, i.e. to the atmospheric washing. It is interesting to note that the Burbáguena station (control), while showing precipitation levels on the order of the El Saler and La Peira stations, presents a lower deposition because it is not influenced by heavily polluted air masses from the coastal areas (Millán et al., 1997). With respect to the literature and taking into consideration stations with precipitation levels similar to those of this study, similar deposition values are also observed. Only 270 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 stations in areas with elevated pollution levels and high precipitation volumes show significantly higher bulk deposition values (Birkenes, 1861 mm and 8.9 kg ha1 year1, Table 5). With respect to N-NH+ 4 bulk deposition, the behavior is similar to that of N-NO 3 , excepting the fact that the Burbáguena station does not react to ammonium like a background station. With this exception the levels oscillate around 3 kg N-NH+ ha1 year1 in the 4 1 northern transect and from 2 to 1 kg N-NH+ 4 ha 1 year in the other transects. Comparing this case with the literature, we can find similar results in Spanish and European Mediterranean stations with similar environments and precipitation volumes (Pedersen et al., 1992; Carratalá, 1993; Camarero and Catalán, 1993; Rodà et al., 1993; Avila, 1996). The stations with the highest ammonium deposition are those located in areas with high ammonium emissions and high annual precipitation, as is the case of the stations in the Netherlands (Table 5). To prepare the dry-deposition estimations within a 14-day period, a rain event was considered to wash the pine branches completely and it was assumed that the gentle washing of the leaf surface would cause minimal leaching from the leaf interior (Ondo et al., 1984). In this way, dry deposition accounted for only a few 14-day dry periods and several periods with 1–14 days of dry deposition. Thus, it must be kept in mind that this is an underestimation of the annual dry deposition. To avoid this, the dry deposition data were recalculated by applying daily deposition rates to all the rainless days in the collection period (Table 5). It should be noticed that after applying this procedure, the total N deposition was then overestimated because of the overestimation of the wet deposition when assumed as bulk deposition. A further correction was made by subtracting from the bulk deposition the dry deposition for rainless days. This is to say that the fraction of dry deposition estimated as branch washing and corrected for surface efficiency as per Table 9 was subtracted from the bulk deposition. However, since the bulk collector resulted in a less efficient surface for dry deposition, and there was a suspicion that NH+ 4 dry deposition had been underestimated, it is probable that even with this correction the wet and total N deposition in Table 9 are still underestimated. Dry deposition may go up to 75% for N at the driest location (Agost). The real values of total N deposition can be found between the N-total deposition values of Table 5 and those of Table 9. Thus, the maximum values of N total deposition are obtained at Agost: 11.0 and 9.78 kg N ha1 year1, respectively. Annual deposition rates in the present study are similar to the annual N deposition inputs (6–11 kg ha1 year1) found in southwestern Sierra Nevada (Chorover et al., 1994), and they are similar to N storage (4–10 kg ha 1 year 1) in the vegetation increment of western forests (Johnson and Lindberg, 1992), suggesting that current N deposition rates may be near the assimilation capacity of the overstory vegetation. Higher values for yearly inorganic N-input have been reported in central Switzerland, with a dry deposition of 16.8 kg N ha1 year1 and a bulk deposition of 12.3 kg N ha1 year1(Schleppi et al., 1998), and in the Alpine area of Austria, with 24–30 kg N ha1 year 1, including dry and occult deposition, and the wet component being approximately 14 kg N ha1 year 1 (Kalina et al., 2000). Acknowledgements This work has been financed by the Generalitat Valenciana (Conselleria de Agricultura, Pesca y Alimentación) and Bancaixa. Thanks are given to Dr. Bea, F. Sanz, C. Coll, P. Bornay, P. Vela, E. Calvo, S. Cosin and J.J. Dieguez for helping us in the field and in the laboratory. We are also indebted to Dr. Bytnerowicz and Dr. Kruppa for their suggestions. J. Scheiding’s corrections to the English text are appreciated. References Avila, A., 1996. Time trends in the precipitation chemistry at a mountain site in north-eastern Spain for the period 1983–1994. Atmospheric Environment 30, 1363–1373. Bytnerowicz, A., 1994. Studies on nitrogen deposition to forests in California. In: Climate and Atmospheric Deposition Studies in Forests. Conference Papers. IGSO PAS, Warszawa, pp. 165–176. Bytnerowicz, A., Fenn, M.E., 1996. Nitrogen deposition in California forests: a review. Environmental Pollution 92, 127–146. Bytnerowicz, A., Miller, P.R., Olszyk, D.M., 1987. Dry deposition of nitrate, ammonium and sulfate to Ceanothus crassifolius canopy and surrogate surfaces. Atmospheric Environment 21, 1749–1757. Bytnerowicz, A., Dawson, P.J., Morrison, C.L., Poe, M.P., 1991. Deposition of atmospheric ions to pine branches and surrogate surfaces in the vicinity of Emerald Lake watershed, Sequoia national park. Atmospheric Environment 25, 2203–2210. Camarero, L., Catalan, J., 1993. Chemistry of bulk precipitation in the Central and Eastern Pyrenees, Northeast Spain. Atmospheric Environment 27A (1), 83–94. Carratalá, A., 1993. Caracterización de la precipitación en la Comunidad Valenciana: distribución espacial y temporal. Tesis doctoral, Universitat de Alacant. Carratalá, A., Bellot, J., 1998. Neutralisation of sulfate and nitrate in precipitation on the Eastern Mediterranean Coast of Spain. Implications for acidification risk. Water, Air and Soil Pollution 104 (3– 4), 237–257. Carratalá, A., Bellot, J., Gomez, A., Millán, M.M., 1996. African dust influence on rainwater on the eastern coast of Spain. In: Guerzoni, S., Chester, R. (Eds.), Desert Dust Across the Mediterranean. Kluwer Academic Publisher, London, pp. 323–332. Casado, H., Encinas, D., Lacaux, M., 1992. The moderating effect of calcium on the acidity in precipitation. Atmospheric Environment 26A (6), 1175. Chorover, J.P.M., Vitousek, D.A., Everson, A.M., Esperanza, A.M., Turner, D., 1994. Solution chemistry profiles of mixed-conifer forest before and after fire. Biogeochemistry 26, 115–144. M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 De Pena, R.G., Walker, W.C., Lebowitz, L., Micka, J.G., 1984. Wet deposition monitoring-effect of sampling period. Atmospheric Environment 19 (1), 151–156. Dignon, J., Hameed, S., 1989. Global emissions of nitrogen and sulfur-oxides from 1960 to 1980. The Journal of the Air and Waste Management Association 39, 180–186. Durand, P., Colin, N., Lelong, F., 1992. Anthropogenic and natural contributions to the rain chemistry of a mountainous area in Cevennes National Park (Mont-Lozere), France. Journal of Hydrology 130, 71–85. Eaton, A.D., Clesceri, L.S., Greenberg, A.E., 1998. Standard Methods for Examination of Water and Wastewater, 20th Edition. United Book Press, Baltimore. Escarré, A., Carratalá, A., Avila, A., Millán, M.M., 1999. Precipitation chemistry and pollution environment. In: Bellot, J., Gracia, C., Retana, J., Rodà, F. (Eds.), Ecology of the Mediterranean Evergreen Oak Forest. Kluwer Academic Publishers, London. Fenn, M.E., 1991. Increased site fertility and litter decomposition rate in high pollution sites in the San Bernardino National Forest in Southern California. Forest Science 37, 11621181. Fenn, M.E., Poth, M.A., 1999. Temporal and spatial trends in streamwater nitrate concentrations in the San Bernardino Mountains, Southern California. J. Environ. Qual. 28, 822836. Fenn, M.E., Poth, M.A., Aber, J.D., Baron, J.S., Bormann, B.T., Johnson, D.W., Lemly, A.D., Mcnulty, S.G., Ryan, D.F., Stottlemyer, R., 1998. Nitrogen excess in north american ecosystems: predisposing factors, ecosystem responses, and management strategies. Ecological Applications 8, 706–733. Fujita, S., Kawaratani, R.K., Fujita, S., Kawaratani, R.K., 1988. Wet deposition of sulfate in the inland sea region of Japan. Journal of Atmospheric Chemistry 7, 59–72. Gatz, D.F., Barnard, W.R., Stensland, G.J., 1986. The role of alkaline materials in precipitation chemistry: a brief review of the issues. Water, Air and Soil Pollution 30, 245–251. Hosker, R.P., Lindberg, S.E., 1982. Review: atmospheric deposition and plant assimilation of gases and particles. Atmospheric Environment 16, 889–910. Ivens, W.P.M.F., 1999. Atmospheric deposition onto forest: an analysis of the deposition varibility by means of throughfall measurements. PhD, University of Utrech, The Netherlands. Johnson, D.W., Lindberg, S.E., 1992. Atmospheric Deposition and Forest Nutrient Cycling. Springer-Verlag, Utrech. Kalina, F.M., Zambo, E., Puxbaum, H., 2000. Assessment of wet, dry and occult deposition of sulfur and nitrogen at an alpine site (special issue). Environmental Science and Pollution Research 1, 53–58. Keene, W.C., Galloway, J.N., Hawley, M.E., 1986. Sea salt corrections and interpretations of constituent ratios in marine precipitation. Journal of Geophysical Research 91 (6), 6647– 6658. Khemani, L.T., Momin, G.A., Prakasa Rao, P.S., Sefai, P.D., Prakash, P., 1987. Influence of alkaline particulates on the chemistry of fog water at Delhi. Water, Air and Soil Pollution 34, 183–189. Krupa, S.V., 2000. Sampling and physico-chemical analysis of precipitation. In: Lodge, J.P. (Ed.), Methods of Air Sampling and Analysis. CRC/Lewis, Boca Raton, FL. Lindberg, S.E., Lovett, G.M., Bondietti, E.A., Davison, C.I., 1984. Recent field studies of dry deposition to surfaces in plant canopies. In: 77th Annual Meeting & Exhibition of the Air Pollution Control Association (No. 84-108.5). San Francisco, CA. Losno, R., Beregameti, G., Carlier, P., Mouvier, G., 1991. Major ions in marine rainwater with attention to sources of alkaline and acidic species. Atmospheric Environment 25A (3–4), 763–770. Loye-Pilot, M.D., Martin, J.M., 1996. Saharan dust input to the Western Mediterranean: an eleven year record in Corsica. In: 271 Guerzoni, S., Chester, R. (Eds.), Desert Dust Across the Mediterranean. Kluwer Academic Publisher, London, pp. 191–199. Millán, M.M., Salvador, R., Mantilla, E., Kallos, G., 1997. Photooxidant dynamics in the western Mediterranean in summer: results from European research projects. Journal of Geophysical Research 102 (D7), 8811–8823. Millán, M.M., Artiñano, B., Alonso, L.A., Castro, M., FernandezPatier, R., Goberna, J., 1992. Mesometeorological Cycles of Air Pollution in the Iberian Peninsula, (MECAPIP). ECE (Research Report 44). Millán, M.M., Mantilla, E., Salvador, R., Carratalá, A., Sanz, M.J., Alonso, L., Gangoiti, G., Navazo, M., 2000. Ozone cycles in the western Mediterranean Basin: interpretation of monitoring data in complex coastal terrain. Journal of Applied Meteorology 39 (4), 487–508. Mosello, R., Bianchi, M., Brizzio, M.C., Geiss, H., Leyendecker, W., Marchetto, A., Rembges, D., Tartari, G.A., Muntau, H., 1999. AQUACOM-MedBAs Project. Subproject 6. Acid Rain Analysis. Intercomparison 1/98. pp. 1–76. Mueller, D.K., Hamilton, P.A., Helsel, D.R., Hitt, K.H., Ruddy, B.C., 1995. Nutrients in Ground Water and Surface Water of the United States—An Analysis of Data Through 1992 (USGS Water Resour. Invest, Report No. 95-4031). Ondo, J.L., John, W., Wall, S.M., 1984. Dry acid deposition on leaves of Ligustrum and a new subrogate leaf. In: 77th Annual Meeting of the Air Pollution Control Association, San Francisco, CA. Pacyna, J.M., Steinar, L., Semb, A., 2000. European survey for NOx emissions with emphasis on eastern Europe. Atmospheric Pollution 25A (2), 425–439. Pedersen, U., Schaug, J., Sklelmoen, J.E., 1992. EMEP Data report 1990. Part 1: Annual Summaries (EMEP/CPP-Report 2/92: pp. 1–230). Ramos, C., 1996. Effect of agricultural practices on the nitrogen losses to the environment. Fertilization Research 43, 183–189. Ramundo, R.A., Seastedt, T.R., 1990. Site-specific underestimation of wetfall NH4 using NADP data. Atmospheric Environment 24A (12), 3093–3095. Riggan, P.J., Lockwood, R.N., Lopez, E.N., 1985. Deposition and processing of airborne nitrogen pollutants in Mediterranean-type ecosystems of southern California. Environmental Science Technology 19, 781–789. Rodà, F., Bellot, J., Avila, A., Escarré, A., Piñol, J., Terradas, J., 1993. Saharan dust and the atmospheric input to Mediterranean ecosystems. Water, Air and Soil Pollution 66, 277–288. Ross, H.B., Lindberg, S.E., 1994. Atmospheric chemical input to small catchments. In: Moldan, B., Cerny, J. (Eds.), Biogeochemistry of Small Catchments. Scope 51. Wiley, Chichester, pp. 55–84. Schleppi, P., Muller, N., Feyen, H., Papritz, A., Bucher, J.B., Fl, Hler., 1998. Nitrogen budgets of two small experimental forested catchments at Alptal, Switzerland. Forest Ecology and Management 101, 177–185. Seufert, G., 1997. BEMA: a European Commision project on biogenic emissions in the Mediterranean area (Special issue). Atmospheric Environment 31, 1–255. Sisterson, D.L., Wurfel, B.E., Lesht, B.M., 1985. Chemical differences between event and weekly precipitation samples in northeastern Illinois. Atmospheric Environment 19 (9), 1453–1469. Stachurski, A., Zimka, J.R., 2000. Atmospheric input of elements to forest ecosystems: a method of estimation using artificial foliage placed above rain collectors. Atmospheric Environment 110, 345–356. Underwood, J.K., Waller, D.H., Thirumutthi, D., 1988. The influence of the ocean on the chemistry of precipitation in Nova Scotia. Atmosphere-Ocean 26 (3), 467–479. Varela, M., 1991. Situación de la contaminación por nitratos en las aguas subterráneas del territorio peninsular y balear. Informaciones y Estudios No. 53. Servicio Geológico, Madrid. 272 M.J. Sanz et al. / Environmental Pollution 118 (2002) 259–272 Vesely, J., 1990. Stability of the pH and the contents of ammonium and nitrate in precipitation samples. Atmospheric Environment 24A (12), 3085–3089. Zeng, Y., Hopke, P.K., 1989. A study of the sources of acid precipitation in Ontario, Canada. Atmospheric Environment 23 (7), 1499–1509.