Bromine accumulation in acidic black colluvial soils

Elsevier Editorial System(tm) for Geochimica et Cosmochimica Acta Manuscript Draft Manuscript Number: GCA-D-15-00092R2 Title: Bromine accumulation in acidic black colluvial soils Article Type: Article Corresponding Author: Dr. Antonio Martinez Cortizas, PhD Corresponding Author's Institution: Universidade de Santiago de Compostela First Author: Antonio Martinez Cortizas, PhD Order of Authors: Antonio Martinez Cortizas, PhD; Cruz Ferro Vázquez, PhD; Joeri Kaal, PhD; Harald Biester, PhD; Manuela Costa Casais, PhD; Teresa Taboada Rodríguez, PhD; Luis Rodríguez Lado, PhD Abstract: Recent investigations showed that bromine is incorporated to soil organic matter (SOM), its content increasing with humification. But few research was done on its long-term accumulation and the role played by pedogenetic processes, as those involved in organic matter stabilization. We investigated bromine content and distribution in four deep, acidic, organic-rich, Holocene soils from an oceanic area of Western Europe. Bromine concentrations (93-778 μg g-1) in the silt+clay (< 50 μm) fraction were on average 3-times higher than those (17-250 μg g-1) in the fine earth (< 2 mm), the former containing almost all bromine (90±5%). Inventories were between 148 and 314 g m-2, indicating a rather large variability in a small area, and total estimated retention was low (6-16%). The degree of SOM bromination, expressed as the Br/C molar ratio, varied between 0.03 and 1.20 mmol Br/mol C. The ratio was highly correlated (n= 23, r2 0.88, p < 0.01) with the age of the SOM for the last ~12 ka. Partial least squares modelling indicates that bromine concentration depends on the amount of organic matter stabilized as aluminium-OM associations, and to a lesser extent on soil acidity (pH) and iron-OM associations. Thus, at scales of thousands of years, bromine accumulation in acidic soils is linked to the pool of metal-clay- stabilized organic matter. Bromine accumulation in acidic black colluvial 1 soils 2 3 4 Antonio Martínez Cortizas1*, Cruz Ferro Vázquez1, Joeri Kaal2, Harald Biester3, Manuela 5 Costa Casais4, Teresa Taboada Rodríguez1, Luis Rodríguez Lado1 6 7 8 1 Departamento de Edafoloxía e Química Agrícola, Facultade de Bioloxía, Campus Vida s/n, 9 15782 Santiago de Compostela, Spain 10 2 Pyrolyscience, 15706 Santiago de Compostela, Spain 11 3 Insitut für Geoökölogie, AG Umweltgeochimie, Technische Universität Braunschweig, 12 Germany 13 4 Departamento de Xeografía, Facultade de Xeografía e Historia, Praza de Universidade s/n, 14 15782 Santiago de Compostela, Spain. 15 16 17 18 19 20 21 22 23 24 25 * corresponding author: [email protected] 26 27 28 29 30 31 32 33 34 35 36 37 38 *Manuscript Abstract 39 Recent investigations showed that bromine is incorporated to soil organic matter (SOM), its 40 content increasing with humification. But few research was done on its long-term 41 accumulation and the role played by pedogenetic processes, as those involved in organic 42 matter stabilization. We investigated bromine content and distribution in four deep, acidic, 43 organic-rich, Holocene soils from an oceanic area of Western Europe. Bromine 44 concentrations (93-778 g g-1) in the silt+clay (< 50 m) fraction were on average 3-times 45 higher than those (17-250 g g-1) in the fine earth (< 2 mm), the former containing almost all 46 bromine (905%). Inventories were between 148 and 314 g m-2, indicating a rather large 47 variability in a small area, and total estimated retention was low (6-16%). The degree of SOM 48 bromination, expressed as the Br/C molar ratio, varied between 0.03 and 1.20 mmol Br/mol 49 C. The ratio was highly correlated (n= 23, r2 0.88, p < 0.01) with the age of the SOM for the 50 last ~12 ka. Partial least squares modelling indicates that bromine concentration depends on 51 the amount of organic matter stabilized as aluminium-OM associations, and to a lesser extent 52 on soil acidity (pH) and iron-OM associations. Thus, at scales of thousands of years, bromine 53 accumulation in acidic soils is linked to the pool of metal-clay-stabilized organic matter. 54 55 Keywords: acid soils, bromine, organo-bromine compounds, stabilized soil organic matter 56 57 58 59 60 61 62 63 64 65 Introdution 66 Until relatively recently, halogens were considered to be unreactive elements in the soil 67 environment, to the point that chlorine, for example, was assumed to be a conservative 68 element with regard to water and used in mass balance biogeochemical modelling. Despite 69 this, several investigations have described the halogenation of organic matter by the activity 70 of haloperoxidases (HPO) (Van Pée & Unverschut, 2003) and more than 4400 organo-71 halogens compounds, mostly with chlorine but also with bromine and a few with iodine, have 72 been identified (Gribble, 2003, 2010). Keppler et al. (2000) also proposed an abiotic 73 mechanism leading to the formation of alkyl-halides during the degradataion of the organic 74 matter by means of an electron acceptor such as Fe3+. Experiments by Leri et al. (2014) also 75 provided evidence of solar irradiation-stimulated abiotic bromination of marine algal detritus. 76 One of the most important contributions to understand the halogenation of soil organic matter 77 is that of Myneni (2002). He demonstrated that as the plant remains decay chlorine is 78 transformed from inorganic into organic forms, and that organic-Cl is the dominant form in 79 humic soil substances (mainly as phenols and aliphatic compounds with one or two Cl atoms). 80 He also found that the content in organo-chlorinated compounds increased with the degree of 81 humification, phenolic compounds increasing as the aliphatic ones decreased. Other 82 investigations also concluded that halogenated organic compounds are abundant in soil 83 organic matter (Maw & Kempton, 1982; Asplund & Grimvall, 1982; Hjelm et al., 1995; 84 Öberg et al., 1996a,b; Johanson et al., 2001) as well as in peat (Biester et al., 2004). 85 Bromine content in the continental crust is around 1.6 g g-1 (Wedepohl, 1995), varying 86 between 0.2 and 10 g g-1 with the highest values in clay sediments (Kabata-Pendias & 87 Pendias, 1992). In soils, the typical range is 5-40 g g-1 (Kabata-Pendias & Pendias, 1992), 88 but values in excess of 400 g g-1 have also been determined in epipedons of soils from 89 coastal areas (Yuita, 1983). Yamada (1968) found a significant enrichment of Br in the soil in 90 comparison to the plant remains, and Maw & Kempton (1982) determined that only a small 91 fraction (1-8%) of the total Br in the soils and peats analyzed was inorganic Br. Hjelm et al. 92 (1995) proposed that the formation of organo-halogens and the degradation of plant remains 93 are simultaneous processes. In the same way, Biester et al. (2004) indicated that 94 atmospherically deposited Br is preferentially retained in humic compounds as a result of its 95 incorporation through microbial activity or by abiotic mechanisms and that Br absorption by 96 vegetation is not a relevant process in the formation of organo-bromine compounds in peat. 97 Several investigations have described the halogenation of SOM by the activity of 98 haloperoxidases (HPO) (Van Pée & Unversucht, 2003), particularly during the degradation of 99 lignin (Schöler et al., 2003). Both chloro- and bromo-HPO are able to incorporate Br to SOM 100 (Ballschmitter, 2003). 101 Leri & Myneni (Leri & Myneni, 2012) demonstrated that bromine is subject to similar 102 biogeochemical processes as chlorine. They found that all bromine in humified plant material, 103 O-horizons and humic substances was covalently bound to organic carbon, increasing in 104 content with decay time in plant remains and suggesting bromine incorporation to organic 105 matter during decay. The soil humic substances displayed an aromatic organo-Br speciation. 106 Therefore, they concluded that the presence of organo-bromine in soil organic matter is linked 107 to decay of plant remains, and its formation may be catalyzed by HPO or HPO-like enzymes 108 in the soil environment. They also suggested, based on the micro-speciation of bromine, that 109 organo-bromine compounds may be dehalogenated as the organic matter is further degraded 110 in the soil column. Recently, Leri & Ravel (2015) found that in laboratory experiments plant 111 material and soil humus can undergo abiotic bromination under the presence of a range of 112 inorganic oxidants, producing aliphatic and aromatic organobrominated forms although, as in 113 the case of the abiotic mechanism proposed by Kepler et al. (2000), the overall importance of 114 this process in natural soils is still unknown. These studies concerned only samples of 115 vegetation and the upper centimeters of the soil and it remains to be determined whether the 116 brominated compounds are depleted or accumulated at the time scales of hundreds to 117 thousands of years of soil formation. It is also equally important to ascertain the role of the 118 main soil pedogenetical processes on the accumulation of halogens. These aspects have only 119 began to be addressed for peat soils (Keppler & Biester, 2003; Biester et al., 2004; Martínez 120 Cortizas et al., 2007). 121 In this study, we selected four deep (145 to 235 cm), acidic, organic matter-rich colluvial soils 122 (also known as Atlantic rankers; Duchaufour, 1982) formed during the Holocene (last 123 ~12,000 years). Their properties, i.e. the high organic matter content, high proportions of 124 stabilized organic matter and age (Kaal et al., 2008a), make them ideal to investigate the long-125 term accumulation of halogens. These soils have been the objective of previous 126 investigations: a study of their pedogenetic origin (Kaal et al., 2008a), the molecular 127 composition of the SOM (Kaal et al., 2008b) and charcoal (Kaal et al., 2009), and 128 palynological and anthracological studies to reconstruct past vegetation changes (Carrión et 129 al., 2010; López-Merino et al., 2012; Kaal et al., 2013). Therefore, the present study builds 130 upon detailed knowledge of their nature and the processes involved in their formation. 131 We analyzed bromine concentrations in high resolution sampled (continuous samples of 5 cm 132 in thickness) soil monoliths and relate them to soil properties with the aim of i) determining 133 whether bromine concentrations change with depth, ii) which soil properties and processes 134 account for bromine distribution, and iii) how the pool of brominated organic matter changes 135 with age. 136 137 Material and methods 138 Study site and sampling 139 The four soils discussed in this paper (PRDI, PRDII, PRDIV and PRDV) are located in NW 140 Spain, in a small isolated hill (42.54º W 8.53 W) between 270 and 320 m a.s.l., 25 km east of 141 the Atlantic coast (Fig. 1). Reference soils for radiocarbon dating were collected within the 142 same area and are cited as PRDIII and PNB in the paper. The rock substratum in the area is a 143 migmatitic granite, composed mainly of quartz, K-feldspars, plagioclase and biotite. The 144 granitic material is overlaid by two pedo-sedimentary layers: a) a 0.5-1 m thick layer of 145 inorganic colluviated saprolitic material, probably of Pleistocene-early Holocene age, and b) a 146 0.8-2.35 m thick strongly humic, sandy, colluvial soil formed during the Holocene (Kaal et 147 al., 2008a). The soils classify as Haplic Umbrisol (Humic, Alumic, Pachic) (IUSS, 2007). 148 Present vegetation is represented by shrubs and herbaceous communities, consisting of 149 Ericaceae, Fabaceae and Poaceae species. A few isolated oak trees, remnants of the climax 150 woodland vegetation, and some planted oak, pine and eucalyptus trees are also present. 151 Palynological and anthracological studies have shown that this vegetation pattern had formed 152 as a result of a stepwise decline of pedunculate oak-dominated woodland and its replacement 153 by shrubland and grassland, with wildfires being among the major agents controlling this 154 process (Carrión et al., 2010; Kaal et al., 2011; López-Merino et al., 2012). 155 Climate can be characterized as mild and humid, with average annual temperatures of 15 ºC 156 (9 ºC in winter and 21 ºC in summer) and annual precipitation of 1200 mm, falling mainly in 157 autumn and winter months (Martínez Cortizas & Pérez Alberti, 1999). 158 Soil monoliths were taken from recently opened ditches, and were sliced into continuous 159 samples of 5 cm in thickness. The total depth sampled was 145 cm, 210 cm, 235 cm, and 220 160 cm for PRDI, PRDII, PRDIV, and PRDV, respectively. 161 162 Sample preparation and physico-chemical analyses 163 The soil material was air-dried at 35 ºC and sieved through a 2 mm mesh to separate the "fine 164 earth" fraction (< 2 mm) from gravel (> 2 mm). Bulk density was obtained with the aid of 165 undisturbed soil cylindres (100 cm3), drying the soil at 105 ºC until constant weight was 166 reached. All other physico-chemical analyses were performed on the fine earth fraction. 167 The methodology to determine soil acidity (pH in water, pHw, and KCl 0.1 M, pHk), total C, 168 N, Si, Al (tAl), Fe (tFe), Ti and Zr, exchange capacity (in 1 M NH4Cl), as well as extractable 169 Al (in 1M KCl, eAl), is described in detail elsewhere (Kaal et al., 2008a). Total C and N 170 contents were determined by combustion using a LECO-CNH 1000 equipment; total Si, Al, 171 Fe, Ti and Zr were determined by XRF; and extractable forms of Fe and Al were determined 172 by AAS. 173 The assessment of metal-SOM complexes is of major importance in acidic soils, and usually 174 achieved through selective dissolution techniques. Aluminium and Fe extracted in 0.1 M Na-175 pyrophosphate (pH 10) (Bascomb, 1968; Buurman et al., 1996) are interpreted as total 176 organically bound Fe and Al (obFe, obAl) (McKeague & Schupli, 1982; Kaiser & Zech, 177 1996; Drabek et al., 2003). Extractions in 0.33 M LaCl3 (pH 4.0, AlL) and 0.5 M CuCl2 (pH 178 2.8, AlCu) (Juo & Kamprath, 1979; Hargrove & Thomas, 1981) are used to evaluate weakly 179 (wbAl: AlL-eAl), moderately (mbAl: AlCu-AlL) and strongly (sbAl: obAl-AlCu) bound 180 organically associated Al fractions, respectively (Urrutia et al., 1988; García-Rodeja et al., 181 2004). To evaluate the proportion of the total Al and Fe bound to the organic matter the 182 obAl/tAl (which is also used as an index for pedogenetical evolution in acidic soils) and 183 obFe/tFe ratios were also calculated. Despite the apparently simple and straightforward 184 interpretation of the different metal fractions, these techniques are subject to debate. For 185 example, Na-pyrophosphate may sometimes dissolve Al hydroxides (mostly short-range 186 ordered, but also gibbsite) not associated to SOM (Kaiser & Zech, 1996); CuCl2, although 187 assumed to be more specific to assess Al-SOM associations, seems to have limited efficiency 188 (Juo & Kamprath, 1979; Urrutia et al., 1995). But, at least for Al, in many investigations and 189 in the soils studied here (García-Rodeja et al., 2004; Kaal et al., 2008a), the extractant 190 efficiencies follow a similar pattern: eAl < AlL < AlCu < obAl. 191 The vertical distribution of the main physico-chemical properties of the soils used in this 192 study can be found in the supporting information (EA_Fig. 1, 2, 3). A summary table with 193 selected soil properties is also provided (Table 1). In synthesis, the soils have low bulk 194 densities, are acidic, have high C content, low to very low effective exchange capacity with 195 Al as the dominant cation, most of the secondary Al is as obAl, and most of the secondary Fe 196 is in the form of crystalline compounds (on average, only a third of the Fe is obFe). 197 Bromine analysis 198 Total Br contents were determined by X-ray fluorescence at the RIAIDT XRF facility of the 199 University of Santiago de Compostela. Standard reference materials were used for the 200 calibration and the samples were measured in duplicate or triplicate. Certified reference 201 materials (NIST 1568b, 1632-A, 1633-A, 1635, 1648a and 1649a, BCR 611 and 612), 202 covering a range of concentrations from 1.4 ± 0.4 to 1190 ± 10 μg g-1, were used to calibrate 203 the equipment. The detection limit was 1 μg g-1 and replicates agreed within 5%. 204 There are only a few Br-bearing minerals and polyhalides, which are unstable in hypergenic 205 environments (Kabata-Pendias & Pendias, 1992), none of which are present in the soils 206 developed from igneous siliceous materials such as the granites of the study area. 207 Nevertheless, the presence of inorganic bromine forms cannot be discarded since 208 dehalogenation reactions, in subsuperficial soil layers, can release bromide. To determine the 209 presence of inorganic, soluble forms, we selected 10 samples from several depths of each of 210 the four soils (40 in total; PRDI: 2.5, 7.5, 22.5, 42.5, 52.5, 72.5, 87.5, 102.5, 127.5, and 132.5 211 cm; PRDII: 2.5, 7.5, 37.5, 62.5, 92.5, 122.5, 142.5, 172.5, 192.5, and 207.5 cm; PRDIV: 2.5, 212 7.5, 27.5, 67.5, 97.5, 122.5, 157.5, 192.5, 217.5, and 232.5 cm; PRDV: 2.5, 7.5, 37.5, 57.5, 213 92.5, 132.5, 152.5, 177.5, 197.5, and 217.5 cm) and analysed them for Br before and after 214 thorough rinsing with deionised, ultrapure distilled water. Data on Br concentration in 215 rainwater, measured in nearby stations (6 stations, n=20), was obtained from Rede de Control 216 da Calidade do Aire (2002). 217 218 Radiocarbon dating of SOM 219 Selected samples from each soil (18 samples in total) plus five samples from another two 220 colluvial soils of the area (PRDIII and PNB; Fig. 1) were sent for radiocarbon dating. Before 221 As for Br, our data show a wide variation in concentration and accumulation in the studied 376 soils, despite being confined to a rather small area. Bromine accumulation is higher in 377 intermediate sections of the complex colluvial sequences and in the smaller grain sizes (~90% 378 on average in the silt+clay fraction), which have concentrations more than 3-fold higher than 379 those of the fine earth. The results of the PLS model (Fig. 6, EA_Fig. 5) suggest that the 380 amount of SOM stabilized as Al-OM associations (Fig. 4B) and acidity are the most 381 important properties related to the accumulation of Br in the studied acidic soils; while Fe-382 OM associations have a smaller effect. 383 Bromine, thus, should behave in soils in a similar way to chlorine (Myneni, 2002), as recently 384 demonstrated by Leri & Myneni (2012). Although this and previous investigations (Yamada, 385 1968; Maw & Kempton, 1982; Hjelm et al., 1995) show that OM is halogenated primarily 386 during decomposition, they do not explain why organo-brominated forms are accumulated in 387 the soil. Apart from halogenation, biotic (Van Pée & Unverschut, 2003; Borch et al., 2003; 388 Fava et al., 2003; Palekar et al., 2003) and abiotic (Karlsson & Öberg, 2003) dehalogenation 389 processes have also been described. Several field studies suggest that in soils both formation 390 and mineralization of organo-halogens occur simultaneously (Öberg et al., 1996a,b; Johanson 391 et al., 2001). Yuita (1983) noted that Br was released when previous aerobic soils (forest, 392 grassland, orchard and upland fields) were converted to paddy fields (anaerobic dominated 393 environment). In agreement with this, the depth profiles of Br residuals of the soils studied by 394 us (EA_Fig. 6) show increasing negative values (lower Br concentrations than expected) at 395 the bottom of soil PRDI (and to some extent in soil PRDII), the only one for which 396 hydromorphic conditions were observed in the field. 397 Leri & Myneni (2012) indicated that the increase in organo-bromine concentration with decay 398 time may be a sign of enhanced resistance to biodegradation. In the studied soils, we have 399 found that Br is correlated to the content of Al-SOM associations (i.e. Cp1 scores, reflecting 400 the abudance of organically-bound Al; Fig. 4B). Organo-Al complexes play a crucial role in 401 the SOM dynamics in acidic soils (Hernández-Soriano, 2012). Several early studies 402 developed on similar soils to those studied here have shown the importance of these 403 components in the acidic soils of NW Spain (Carballas et al., 1967; García-Rodeja et al., 404 1987; Álvarez et al., 1992, 2002) and their relevance in the stabilization of SOM (Caraballas 405 et al., 1978, 1980; González Prieto et al., 1992). More recent experimental studies (Kalbitz et 406 al., 2003; Marschner et al., 2003; Schwesig et al., 2003) have shown that complexation with 407 Al decreases SOM degradability. An increase of the Al/C ratio to values of 0.1-0.2 reduces by 408 a half the mineralization of SOM, in particular of aromatic compounds whose average life 409 increases up to 4 fold. The specific Al/C ratio at which the maximum capacity of SOM to 410 bind metals is reached is pH-dependent (Buurman, 1985; Nierop et al., 2002), what may 411 account for the incorporation of soil reaction in the statistical model for our soils. Johansson 412 et al. (2003) also found a secondary influence of pH on the content of organic-chlorine in O-413 horizons of forest soils. 414 The stabilization of brominated-SOM is supported by the significant correlation between the 415 Br/C molar ratio and the age of SOM. The slope of the regression equation (Fig. 4) indicates 416 that the degree of bromination of organic matter in the PRD soils has increased by 7.6 mmol 417 Br/mol of C per century, while the intercept term - a theoretical zero time- (0.18 mmol Br/mol 418 C; significantly different from 0, p < 0.01) indicates that bromination of SOM may be a fast 419 process that already occurs at the soil surface. The later is in agreement with Leri & Myneni 420 (2012), who consider that the relatively rapid oxidation of part of the deposited bromide 421 seems to lead to a rapid transformation to organic forms. Our results also indicate that metal-422 bromine-SOM compounds become enriched at millennial time scales, likely due to the more 423 intensive degradation of non-stabilized SOM fractions. This is in agreement with the studies 424 on SOM stabilization by aluminium (Kalbitz et al., 2003; Marschner et al., 2003; Schwesig et 425 al., 2003), in which it was found that the mineralization of labile carbon forms was not 426 affected by the incorporation of Al. Albeit that some authors have suggested that Fe oxides 427 may be the most important agents in the formation of organo-mineral associations in subsoils 428 (Guggenberger & Kaiser, 2003; Kögel-Knabner et al., 2008), they seem to play a minor role 429 in the studied soils. This is probably due to the relatively low Fe content of the parent 430 materials. 431 The NaOH-extractable SOM in soils PRDI and PRDII, which accounts for 80-100 % of total 432 SOM, is composed primarily of microbial/degraded and pyrogenic SOM (from palaeofires) 433 (Kaal et al., 2008b). Large proportions of the pyrolysis products 5-methyl-2-furaldehyde and 434 phenol, which originate predominantly from microbial/degraded SOM, coincide with 435 relatively high Br/C ratios, while low Br/C coincides with pyrolysis products that originate 436 from fire residues. These results suggest that the bromination reaction is related to microbial 437 degradation and therefore that it is mainly a biobromination process. The microbial/degraded 438 biomass occurs mostly in microaggregates, where it is stabilised by Al-binding in 439 combination with physical occlusion (Kaal, 2011). Furthermore, there is a negative 440 relationship between peak contents in macroscopic charcoal fragments (> 2 mm) and minima 441 in the Br/C ratio (not shown), suggesting that fire residues, and large charcoal particles in 442 particular, seem relatively recalcitrant to bromination. It is noteworthy that the frequent 443 combustion of SOM in the topsoil during fires may have resulted in the partial depletion of 444 organobrominated compounds at the soil surface. It should be emphasized that selective 445 bromination of different organic matter constituents, selective incorporation into aggregates, 446 potentially selective debromination by fires, and selective degradation/preservation during 447 long-term storage in soil, are complex processes that are hardly understood, and therefore that 448 the apparent link between Br/C ratios and microbial vs. fire residues in these soils is only 449 hypothetical. 450 Our results indicate that the accumulation of bromine in acidic soils is coupled to the 451 chemistry of Al (and Fe) and the stabilization of organic matter, ultimately controlled by 452 weathering and pedogenetical processes (responsible for the release of Al and Fe). The role of 453 SOM stabilization mechanisms may also be important in other, different soil environments, 454 since Gerzabek et al. (1999) found that Br content was higher in calcareous than in carbonate 455 free soils, and in the former the Br concentration was correlated to clay and SOM content and 456 exchangeable Ca. It is known that in calcareous soils the formation of Ca-humates is the most 457 efficient mechanism of SOM stabilization (Duchaufour, 1982). 458 459 Concluding remarks 460 Our results have an interesting implication for studies of organic matter in marine sediments. 461 The Br/C ratio has been used as a source estimator (Mayer et al., 2007) for the sedimentary 462 OM in the marine environment, assuming that terrestrial OM has a low ratio when compared 463 to marine OM. A Br/C ratio of 0.43 (mg Br/g Corg) has been applied as representative end 464 member of terrestrial OM and of 7.7 mg Br/g Corg as end member for marine OM (Mayer et 465 al., 2007). In the soils studied by us, Br/C molar ratios equivalent to 0.2-8.0 mg Br/g Corg 466 have been determined, increasing with the age of the SOM. Thus, the average age of the 467 terrestial SOM exported to the marine environment may influence these calculations: the 468 younger the age the larger the underestimation and the older the age the larger the 469 overestimation of the marine OM contribution. On the other hand, since Br content in 470 rainwater usually decreases rapidly from the coast inland, Br accumulation and bromination 471 of SOM may decrease proportionately. Hence, the SOM of sediments delivered from small 472 watersheds adjacent to the ocean would be expected to have higher Br/C ratios relative to 473 large rivers whose suspended sediment load derives from farther inland soils. Once in the 474 ocean, any process affecting the fractionation between brominated and non-brominated 475 terrestrial OM (preferential degradation, longer transport of finer fractions, to name only a 476 few) is likely to have an effect on the estimations. 477 478 Acknowledgements 479 This work was done under the framework of projects CGL2010-20672 and CSD2007-00058 480 (Plan Nacional I+D+i, Spanish Ministerio de Economía y Competitividad) and 481 10PXIB200182PR, PGIDT02CCP60601, and EM2012/60 (Dirección Xeral I+D, Xunta de 482 Galicia). Luis Rodríguez Lado was supported by the Isidro Parga Pondal Research 483 Programme (Xunta de Galicia, 2011). Thanks are also extended to Ward Chesworth (LRS, 484 University of Guelph, Canada) for his comments and suggestions to an earlier version of this 485 manuscript and to four anonymous reviewers whose comments helped to improve the 486 manuscript. 487 488 Supporting information available 489 Radiocarbon ages of dated soil samples (EA_Table 1); depth records of soil properties 490 (EA_Fig. 1); depth records of elements (Si, Al, Fe, Ti, Zr) concentratrions (EA_Fig. 2); depth 491 records of extractable Al and Fe (EA_Fig. 3); depth records of the scores of the four principal 492 components extracted on the soil properties and elements concentrations (EA_Fig. 4); 493 summary statistics of the PLS model (EA_Fig. 5); depth records of the statistical residues of 494 bromine concentration (EA_Fig. 6). 495 496 References 497 Abdi H. (2010) Partial least squares regression and projection on latent structure 498 regression (PLS Regression). WIREs Comput Stat, Focus Article, DOI: 499 10.1002/wics.051. 500 Álvarez E., Martínez Cortizas A., Calvo R. (1992) Geochemical aspects of aluminum in 501 forest soils in Galicia (NW Spain). Biogeochemistry 16, 167-180. 502 Álvarez E., Monterroso C., Fernández Marcos M.L. (2002) Aluminum fractionation in 503 Galician (NW Spain) forest soils as related to vegetation and parent material. For 504 Ecol Manag 166, 193-206. 505 Asplund G., Grimvall A. (1991) Organohalogens in nature, more widespread than 506 previously assumed. Environ Sci Technol 25, 1346-1350. 507 Ballschmitter K. (2003) Pattern and sources of naturally produced organohalogens in 508 the marine environment: biogenic formation of organohalogens. Chemosphere 52, 509 313-324. 510 Bascomb C.L. (1968) Distribution of pyrophosphate extractable iron and organic carbon 511 in soils of various groups. J Soil Sci 19, 251–268 512 Biester H., Keppler F., Putschew A., Martínez Cortizas A., Petri M. (2004) Halogenes, 513 organohalogens and the role of organic matter decomposition on halogen 514 enrichment in two peat bogs from Tierra del Fuego (Chile). Envir Sci Technol 38, 515 1984-1991. 516 Borch T., Ambus P., Laturnus F., Svensmark B., Gron C. (2003) Biodegradation of 517 chlorinated solvents in a water unsaturated topsoil. Chemosphere 51, 143-152. 518 Buurman P. (1985) Carbon/sexquioxide ratios in organic complexes and the transition 519 albic-spodic horizon. J Soil Sci 36, 255-260. 520 Buurman P., van Lagen B., Velthorst, E. J. (eds.) (1996) Manual for Soil and Water 521 Analysis. Backhuys Publishers, Leiden, The Netherlands. 314pp. 522 Carballas Fernández MA. (1982) Estudio de la génesis del Ranquer Atlántico. PHD 523 Thesis, UNiversidad de Santiago de Compostela, Santiago de Compostela, 163 pp. 524 Carballas T., Carballas M., Jacquin F. (1978) Biodegradation et humification de la matière 525 organique des sols humiferes atlantiques. Anal Edaf Agrobiol 37, 205-212. 526 Carballas T., Duchaufour Ph., Jacquin F. (1967) Évolution de la matière organicque des 527 rankers. Bull l’École Nat Sup Agron Nancy 9, 20-28. 528 Carballas T., Guitián F., Cabaneiro A. (1980) Organo-metallic complexes in Atlantic 529 humiferous soils. Anal Edaf Agrobiol 39, 1033-1043. 530 Carrión Y., Kaal J., López-Sáez JA., López-Merino L., Martínez Cortizas A. (2010) Holocene 531 vegetation changes in NW Iberia revealed by anthracological and palynological 532 records from a colluvial soil. Holocene 20, 53-66. 533 Drabek O., Boruvka L., Mladkova L., Kocarek M. (2003) Possible method of aluminium 534 speciation in forest soils. J Inorg Biochem 97, 8–15. 535 Duchaufour, P. (1982) Pedology: pedogeneis and classification. George Allen & Unwin, 536 London. 537 Eriksson L., Johansson E., Kettaneh-Wold N., Wold S. (1999) Introduction to multi- and 538 mega-variate data anlysis using projection methods (PCA & PLS). Umetrics AB, 539 Umea. 540 Fava F., Gentilucci S., Zanaroli G. (2003) Anaerobic biodegradation of weathered 541 polychlorinated biphenyls (PCBs) in contaminated sediments of Porto Marghera 542 (Vence Lagoon, Italy). Chemosphere 53, 101-109. 543 García-Rodeja E., Nóvoa-Muñoz J. C., Pontevedra-Pombal X., Martínez-Cortizas A., 544 Buurman, P. (2004) Aluminium fractionation of European volcanic soils by 545 selective dissolution techniques. Catena 56, 155–183. 546 García-Rodeja E., Silva B., Macías F. (1987) Andosols developed from non volcanic 547 materials in Galicia (NW Spain). J Soil Sci 38, 573–591. 548 Gerzabek M.H., Muramatsu Y., Strebl F., Yoshida S. (1999) Iodine and bromine contents 549 of some Austrian soils and relations to soil characteristics. J Plant Nutr Soil Sci 162, 550 415-419. 551 González Prieto S.J., Villar M.C., Carballas M., Carballas T. (1992) Nitrogen mineralization 552 and its controlling factors in various kinds of temperate humid-zone soils. Plant 553 Soil 144, 31-44. 554 Gribble GW. (2003) The diversity of naturally produced organohalogens. Chemosphere 555 52, 289-297. 556 Gribble GW. (2010) Naturally occurring Organohalogen Compounds - A Comprehensive 557 Update. Progress in the Chemistry of Organic Natural Products 91, 1-613. Springer. 558 Guggenberger G., Kaiser K. (2003) Dissolved organic matter in soil: challenging 559 paradigm of sorptive preservation. Goederma 113, 293-310. 560 Hargrove WL., Thomas GW. (1981) Extraction of aluminum from aluminum-organic 561 matter complexes. Soil Sci Soc Am J 45, 151–153. 562 Hjelm O., Johansson M., Öberg G. (1995) Organically bound halogens in coniferous forest 563 soil. Chemosphere 30, 2353-2364 564 IUSS Working Group WRB. (2007) World reference base for soil resources 2006, first 565 update 2007. World Resources Reports Food and Agriculture organization of the 566 United Nations. Rome. No. 103. 567 Johanson E., Ebena G., Sanden P., Svenson T., Öberg G. (2001) Organic chlorine and 568 chloride in Swedish spruce forest soil: influence of nitrogen. Geoderma 10, 1-13. 569 Johanson E., Sanden P., Öberg G. (2003) Organic chlorine in deciduous and coniferous 570 forest soils in southern Sweden. Soil Science 168, 347-355. 571 Juo AS., Kamprath EJ. (1979) Copper chloride as an extractant for estimating the 572 potentially reactive aluminium pool in acid soils. Soil Sci Soc Am J 43, 35–38. 573 Kaal J. (2011) Identification, molecular characterisation and significance of fire residues 574 in colluvial soils from Campo Lameiro (NW Spain). Ph.D. Dissertation, University of 575 Santiago de Compostela, Santiago de Compostela, Spain, http://hdl.handle.net 576 /10261/35227. 577 Kaal J., Carrión Marco Y., Asouti E., Martín Seijo M., Martínez Cortizas A., Costa Casáis M., 578 Criado Boado F. (2011) Long-term deforestation in NW Spain: Linking the 579 Holocene fire history to vegetation change and human activities. Quat Sci Rev 30, 580 161 - 175. 581 Kaal J., Costa-Casais M., Ferro-Vázquez C., Pontevedra-Pombal X., Martínez Cortizas A. 582 (2008a) Soil formation of "Atlantic Rankers" from NW Spain - A high resolution 583 aluminium and iron fractionation study. Pedosphere 18, 441-453. 584 Kaal J., Criado-Boado F., Costa-Casais M., López-Sáez JA., López-Merino L., Mighall T., 585 Carrión Y., Silva Sánchez N., Martínez Cortizas A. (2013) Prehistoric land use at an 586 archaeologicla hot-spot (the rock art park of Campo Lameiro, NW Spain) inferred 587 from charcoal, synathropic pollen and non-pollen palynomorph proxies. J Archaeol 588 Sci 40, 1518-1527. 589 Kaal J., Martínez Cortizas A., Nierop K.G.J., Buurman P. (2008b) A detailed pyrolysis-590 GC/MS analysis of a black carbon-rich acidic colluvial soil (Atlantic ranker) from 591 NW Spain. Appl Geochem 23, 2395-2405. 592 Kaal J., Martínez Cortizas A., Eckmeier E., Costa Casais M., Santos Estévez M., Criado 593 Boado F. (2008c) Holocene fire history of black colluvial soils reveealed by 594 pyrolisis-GC/MS: a case study from Campo Lameiro (NW Spain). J Archaeol Sci 35, 595 2133-2143. 596 Kaal J., Martínez Cortizas A., Nierop KGJ. (2009) Characterisation of aged charcoal using 597 a coil probe pyrolysis-GC/MS method optimsed for black carbon. J Anal Appl 598 Pyrolysis 85, 408-416. 599 Kabata-Pendias A., Pendias H. (1992) Trace elements in soils and plants. CRC Press. 600 Kaiser K., Zech W. (1996) Defects in estimation of aluminum in humus complexes of 601 podzolic soils by pyrophosphate extraction. Soil Sci 161, 452–458. 602 Kalbitz K., Schmerwitz J., Schwesig D., Matzner E. (2003) Biodegradation of soil-derived 603 dissolved organic matter as related to its properties. Geoderma 113, 273-291. 604 Karlsson S.A., Öberg G. (2003) UV-light induced mineralization of organic matter bound 605 chlorine in lake Bjän. Chemosphere 52, 463-469. 606 Keppler F., Biester H. (2003) Peatlands: a major source of naturally produced organic 607 chlorine. Chemosphere 52, 451-453. 608 Keppler F., Elden R., Niedan V., Pracht J., Schöler H.F. (2000) Halocarbons produced by 609 natural oxidation processes during degradation of organic matter. Nature 403, 610 298-300. 611 Kögel-Knabner I., Guggenberger G., Kleber M., Kandeler E., Klabitz K., Scheu S., Esterhues 612 K., Leinweber P. (2008) Organo-mineral associations in temperate soils: 613 integrating biology, mineralogy, and organic matter chemistry. J Plant Nutr Soil Sci 614 171, 61-82. 615 Kristiansen S., Dalsgaard K., Holst MK., Aaby B., Heinemeier J. (2003) Dating of 616 prehistoric burial mounds by 14C analysis of SOM fractions. Radiocarbon 45, 101-617 112. 618 Leri A., Mayer L.M., Thornton K.R., Ravel, B. (2014) Bromination of marine particulate 619 organic matter through oxidative mechanisms. Geochim Cosmochim Ac 142, 53-63. 620 Leri A.C., Myneni S.C.B. (2012) Natural organobromine in terrestrial ecosystems. 621 Geochim Cosmochim Ac 77, 1-10. 622 Leri A., Ravel B. (2015) Abiotic bromination of soil organic matter. Envir Sci Technol 623 (accepted), DOI: 10.1201/acs.est.5b03937. 624 López-Merino L., Silva Sánchez N., Kaal J., López-Sáez JA., Martínez Cortizas A. (2012) 625 Post-disturbance vegetation dynamics during the Late Pleistocene and the 626 Holocene: An example from NW Iberia. Glob Planet Change 92-93, 58-70. 627 Marschner B., Kalbitz K. (2003) Controls of bioavailability and biodegradability of 628 dissolved organic matter in soils. Geoderma 113, 211-235. 629 Martínez Cortizas A., Biester H., Mighall T., Bindler R. (2007) Climate-driven enrichment 630 of pollutants in peatlands. Biogeosciences 4, 905-911. 631 Martínez-Cortizas A, Pérez-Alberti A. (1999) Atlas Climático de Galicia (in Spanish). 632 Consellería de Medio Ambiente (Xunta de Galicia), Santiago de Compostela. 250pp. 633 Maw G.A., Kempton R.J. (1982) Bromine in soils and peats. Plant Soil 65, 103-109. 634 Mayer L.M., Schick L.L., Allison M.A., Ruttenberg K.C., Bentley S.J. (2007) Marine vs 635 terrigenous organic matter in Louisiana coastal sediments: the uses of 636 bromine:organic carbon ratios. Mar Chem 107, 244-254. 637 McKeague J. A., Schuppli P. A. (1982) Changes in concentration of iron and aluminum in 638 pyrophosphate extracts of soil and composition of sediment resulting from ultra-639 centrifugation in relation to spodic horizons. Soil Sci 134, 265–270. 640 Mevik B.H. (2006) The pls R package. R News 6, 12-17. 641 Myneni S.C. (2002) Formation of stable chlorinated hydrocarbons in weathering plant 642 material. Science 295, 1039-1041. 643 Nierop KGJ., Buurman P., de Leeuw JW. (1999) Effect of vegetation on chemical 644 composition of H horizons in incipient podzols as characterized by 13C NMR and 645 pyrolysis-GC/MS. Geoderma 90, 111-129. 646 Nierop K.G.J, Jansen B., Vestraten J.M. (2002) Dissolved organic matter, aluminium and 647 iron interactions: precipitation induced by metal/carbon ratio, pH, and 648 competition. Sci Total Environ 300, 201-211. 649 Öberg G., Börjesson I., Samuelsson B. (1996a) Net change in organically bound halogens 650 during decomposition of Norway spruce litter. Water Air Soil Poll 89, 351-361. 651 Öberg G., Nordlund E., Berg B. (1996b) In situ formation of organically bound halogens 652 during decomposition of Norway sproce litter – Effects of fertilization. Can J For 653 Res 26, 1040-1048. 654 Orlova LA., Panychev VA. (1993) The reliability of radiocarbon dating buried soils. 655 Radiocarbon 35, 369-377. 656 Palekar L.D., Maruya K.A., Kostka J.E., Wiegel J. (2003) Dehalogenation of 2,6-657 dibromobiphenyl and 2,3,4,5,6-pentachlorobiphenyl in contaminated estuarine 658 sediment. Chemosphere 53, 593-600. 659 Pessenda, L.C.R., Gouveia, S.E.M., Aravena, R., 2001. Radiocarbon dating of total soil 660 organic matter and its comparison with 14C ages of fossil charcoal. Radiocarbon 661 43, 595-601. 662 Rede de Control da Calidade do Aire 1998-2001 (2002) Xunta de Galicia. 663 Reimer P. J., Baillie M.G.L., Bard E., Bayliss A., Beck J.W., Blackwell P.G., Bronk Ramsey C., 664 Buck C.E., Burr G.S., Edwards R.L., Friedrich M., Grootes P.M., Guilderson T.P., 665 Hajdas I., Heaton T.J., Hogg A.G., Hughen K.A., Kaiser K.F., Kromer B., McCormac 666 F.G., Manning S.W., Reimer R.W., Richards D.A., Southon J.R., Talamo S., Turney 667 C.S.M., van der Plicht J., Weyhenmeyer C.E. (2009) IntCal09 and Marine09 668 radiocarbon age calibration curves, 0-50,000 years cal BP. Radiocarbon 51, 1111-669 1150. 670 Schöler H.F., Keppler F., Fahimi I.J., Niedan V.W. (2003) Fluxes of trichloroacetic acid 671 between atmosphere, biota, soils, and groundwater. Chemosphere 52, 339-354. 672 Schwesig D., Kalbitz K., Matzner E. (2003) Effects of aluminum on the mineralization of 673 dissolved organic carbon derived from forest floors. Eur J Soil Sci 54, 311-322. 674 Figure 2 Click here to download high resolution image Figure 3 Click here to download high resolution image Figure 4 Click here to download high resolution image Figure 5 Click here to download high resolution image Figure 6 Click here to download high resolution image TOC Click here to download high resolution image Electronic Annex Click here to download Electronic Annex: Electronix Annex.pdf