Pan-Arctic Trends in Terrestrial Dissolved Organic Matter from Optical Measurements

Climate change is causing extensive warming across arctic regions resulting in permafrost degradation, alterations to regional hydrology, and shifting amounts and composition of dissolved organic matter (DOM) transported by streams and rivers. Here, we characterize the DOM composition and optical properties of the six largest arctic rivers draining into the Arctic Ocean to examine the ability of optical measurements to provide meaningful insights into terrigenous carbon export patterns and biogeochemical cycling. The chemical composition of aquatic DOM varied with season, spring months were typified by highest lignin phenol and dissolved organic carbon (DOC) concentrations with greater hydrophobic acid content, and lower proportions of hydrophilic compounds, relative to summer and winter months. Chromophoric DOM (CDOM) spectral slope (S275-295) tracked seasonal shifts in DOM composition across river basins. Fluorescence and parallel factor analysis identified seven components across the six Arctic rivers. The ratios of ‘terrestrial humic-like’ versus ‘marine humic-like’ fluorescent components co-varied with lignin monomer ratios over summer and winter months, suggesting fluorescence may provide information on the age and degradation state of riverine DOM. CDOM absorbance (a350) proved a sensitive proxy for lignin phenol concentrations across all six river basins and over the hydrograph, enabling for the first time the development of a single pan-arctic relationship between a350 and terrigenous DOC (R2 = 0.93). Combining this lignin proxy with high-resolution monitoring of a350, pan-arctic estimates of annual lignin flux were calculated to range from 156 to 185 Gg, resulting in shorter and more constrained estimates of terrigenous DOM residence times in the Arctic Ocean (spanning 7 months to 2½ years). Furthermore, multiple linear regression models incorporating both absorbance and fluorescence variables proved capable of explaining much of the variability in lignin composition across rivers and seasons. Our findings suggest that synoptic, high-resolution optical measurements can provide improved understanding of northern high-latitude organic matter cycling and flux, and prove an important technique for capturing future climate-driven changes.


explaining much of the variability in lignin composition across rivers and seasons. 26
Our findings suggest that synoptic, high-resolution optical measurements can provide 27 improved understanding of northern high-latitude organic matter cycling and flux, and 28 prove an important technique for capturing future climate-driven changes. 29

30
Northern high-latitude regions contain substantial quantities of organic carbon in 31 perennially and seasonally frozen soils, comprising more than half the entire global 32 carbon soil stock (Tarnocai et al., 2009). Large arctic rivers play an increasingly 33 recognized role in regional carbon cycling by transporting a proportion of this 34 terrigenous material from land to the ocean, whilst also acting as sites for active 35 carbon metabolism and transformation (Holmes et al., 2011;Mann et al., 2015;36 Spencer et al., 2015;Striegl et al., 2005). Arctic riverine export is substantial enough 37 (~ 10 % of the global freshwater discharge) that it imparts estuarine-like water quality 38 characteristics throughout the Arctic Ocean, influencing coastal salinity structure on a 39 localized basis (Aagaard and Carmack, 1989;McClelland et al., 2011;Serreze et al., 40 2006). Furthermore, significant quantities of dissolved organic matter (DOM) 41 accompany this freshwater flux causing higher than average dissolved organic carbon 42 (DOC) concentrations in the Arctic Ocean relative to other ocean basins (Hernes and 43 Benner, 2006;Mathis et al., 2005;Opsahl et al., 1999). Six major arctic rivers account 44 for the majority of freshwater flux, each draining vast watersheds on the Eurasian 45 (Kolyma, Ob', Lena, Yenisey) or North American (Mackenzie, Yukon) continents, 46 combined delivering ~ 64 % of the total freshwater supplied to the Arctic Ocean 47 (Holmes et al., 2011). 48 Arctic rivers are characterized by their strong seasonality and large intra-49 annual variability in runoff, driven by extreme fluctuations in snow cover and air 50 temperatures. Discharge rapidly peaks with the onset of snow melt and ice-breakup, 51 resulting in dramatic spring freshet events and rapid transport of terrigenous DOM 52 offshore (Amon et al., 2012;Mann et al., 2012;Stedmon et al., 2011a). By contrast, 53 winter months are distinguished by low discharge and DOC concentrations, with 54

P r o v i s i o n a l
DOM exhibiting lower average aromaticity and molecular weight (O'Donnell et al., 55 2012;Spencer et al., 2008). Future changes in the fluxes and composition of 56 terrigenous DOM released to and exported from arctic rivers are likely. River 57 discharge across much of the pan-arctic watershed is increasing, particularly during 58 winter months (Déry et al., 2009;McClelland et al., 2006;Peterson, 2002;Rawlins et 59 al., 2010;Smith et al., 2007). Deepening of the seasonally thawed active layer will 60 also result in leaching of deeper soil and permafrost horizons altering the amount and 61 type of DOM liberated to inland waters (Romanovsky et al., 2010). Changes in the 62 quality of DOM affect the reactivity and fate of terrigenous DOM, influencing carbon 63 turnover rates and regional carbon budgets Mann et al., 2012;64 2014;Wickland et al., 2012). Tracing future alterations in the composition as well as 65 concentration of riverine DOM is therefore crucial for understanding the effects of 66 climate change. 67 Lignin phenols are unique biomarkers of vascular plant material and therefore 68 act as sensitive indicators for the terrigenous component of aquatic DOM. As well as 69 providing pertinent information on DOM source, lignin phenols also have the capacity 70 to capture degradative processing and source information (Hernes et al., 2007;Opsahl 71 and Benner, 1995;Spencer et al., 2010a). DOM source and composition have also 72 been assessed via separation of the DOM pool using XAD fractionation techniques. 73 DOC fractionation has been used to differentiate between high molecular weight, 74 aromatic dominated carbon DOM fractions, primarily sourced from allochthonous 75 materials, and those dominated by microbially-derived or photodegraded DOM (e.g. 76 Aiken et al., 1992;Spencer et al., 2012). Despite providing critical information, both 77 lignin phenol and XAD fractionation techniques are costly and extremely time 78 consuming to conduct, limiting their applicability for high-resolution monitoring. The 79 remote nature of arctic watersheds and the rapid shifts in hydrology make effective 80 sampling and observation of these regions incredibly challenging. Despite far greater 81 understanding of constituent fluxes and biogeochemical cycles across Arctic river 82 systems, much garnered from international sampling campaigns (e.g. PARTNERS, 83 Arctic-GRO; www.arcticgreatrivers.org), insufficient temporal and spatial resolution 84 in measurements still limits our ability to capture changes in terrigenous DOM supply 85 and examine how it may alter under future scenarios. For example, the Arctic Great 86 Rivers Observatory (Arctic-GRO) captures the major seasonal patterns in river 87 chemistry and freshwater discharge across the six major arctic rivers, ensuring 88 identical sampling and analytical protocols, yet is limited with respect to the number 89 of samples that can be feasibly collected. The use of optical measurements, which can 90 be rapidly collected and measured, remotely derived or determined in-situ, is one 91 pathway that can help to address these problems. 92 A number of studies have investigated the ability of optical measurements to 93 capture changes in DOM composition occurring across rivers or over the hydrograph, 94 or to relate optical and lignin-based proxies to improve estimates of terrigenous DOM 95 residence times in the Arctic Ocean Stedmon et al., 2011a;96 Walker et al., 2013). Recently, chromophoric DOM (CDOM) absorbance 97 measurements from 30 unique US watersheds were shown to correlate to DOM 98 composition, as derived via XAD fractionation, highlighting the potential of optical 99 measurements to improve our understanding of DOM dynamics in fluvial systems 100 . Additionally, CDOM absorbance-lignin relationships have 101 been developed for the Yukon River and then scaled to the pan-arctic, assuming 102 similar relative loads of lignin in freshwater fluxes across all arctic rivers (Spencer et 103 al., 2009). Using this approach, Spencer et al., (2009) found that terrigenous DOM 104 P r o v i s i o n a l export to the Arctic Ocean was higher than previously thought, and thus concluded 105 that a greater proportion must either be modified during transit through estuaries, or 106 removal processes in the Arctic Ocean are greater than previously thought. CDOM 107 fluorescence measurements have also been shown to be potentially useful proxies for 108 lignin phenol concentration and composition in freshwaters 109 Walker et al., 2013). Successful relationships have been reported between CDOM 110 fluorescence, collected as excitation-emission matrices (EEMs) and decomposed 111 using parallel factor analysis (PARAFAC), and lignin measurements across individual 112 arctic rivers, yet pan-arctic relationships remain elusive (Walker et al., 2013). In 113 particular, no studies have attempted to develop relationships between DOM optical 114 properties from across all six arctic rivers and DOC fractionation measurements 115 (XAD), or with vascular plant biomarkers (lignin phenols) as rapid proxies for 116 terrestrial DOC export and composition across the Arctic. Additionally, no studies 117 have examined the utility of combining absorbance and fluorescence techniques to 118 develop arctic proxies for terrigenous DOM export. 119 Here, we characterize the DOM optical properties and composition (XAD and 120 lignin phenol) of the six largest arctic rivers to examine the ability of optical 121 measurements to provide meaningful insights into terrigenous carbon export patterns 122 and biogeochemical cycling across broad spatial scales in the Arctic. Specifically, we 123 attempt to identify common optical indices that trace DOC and lignin phenol 124 concentration and compositional information across all six arctic rivers. Further, we 125 examine the utility of using a combination of absorbance and fluorescence 126 measurements to predict trends in DOC and lignin phenol biomarkers. Finally, we 127 develop, for the first time, a pan-arctic optical proxy for estimating terrestrial OC flux 128 syringic acid), and two cinnamyl phenols (p-coumaric acid, ferulic acid). One blank 179 was run for every ten-sample oxidations and all samples were blank corrected. Blank 180 concentrations of lignin phenols were low (30 -40 ng) and consequently never 181 exceeded 5 % of the total lignin phenols in a sample. Lignin phenol concentrations are 182 reported as the sum of the cinnamyl, syringyl and vanillyl phenols (∑ 8 (Hu et al., 2002). The slope (S) of the absorbance spectra was 192 calculated from wavelength ranges spanning 275-295, 290-350, and 350-400 nm and 193 the slope ratio (S R ) determined as S 275-295/ S 350-400 (Helms et al., 2008). Slope 194 coefficients can provide information pertaining to CDOM composition and source,195 with steeper values and increasing S R indicative of lower molecular weight and 196 decreasing DOM aromaticity (Blough and Del Vecchio, 2002;Blough and Green, 197 1995;Helms et al., 2008). Specific UV absorbance (SUVA 254 ) was calculated by 198 dividing the decadal UV absorbance at 254 nm by the DOC concentration (Weishaar 199 et al., 2003) . The specific UV absorbance (SUVA 254 ) has been shown to be positively 200 correlated to percent aromaticity within DOM (Weishaar et al., 2003). 201 Fluorescence was analyzed using a Horiba Fluoromax-4 spectrofluorometer 202 (Jobin-Yvon). Excitation-emission matrices (EEMs) were collected at 20 º C in ratio 203 P r o v i s i o n a l (S/R) mode over excitation and emission wavelengths of 250 -450 and 320 -550 nm, 204 in 5 and 2 nm increments respectively. Measurements were performed with 0.1s 205 integration times and 5 nm slit widths on the excitation and emission 206 monochromators. Instrument specific correction files were applied before further 207 analyses. Fluorescence EEMs were blank corrected from at least daily Milli-Q blanks 208 collected identically to samples. Daily water Raman scans were collected at Ex=350 209 nm (e.g. Lawaetz and Stedmon, 2009). Raman and Rayleigh-Tyndall scatter were 210 removed and interpolated using the smootheem function and absorbance 211 measurements were used to correct EEMs for inner filter effects according to the 212 method of Lakowicz, (2013) within the drEEM toolbox (Murphy et al., 2013). The 213 fluorescence index (FI) was also calculated as the ratio of emission at 470 nm to 520 214 nm, at an excitation wavelength of 370 nm (Cory et al., 2010;McKnight et al., 2001). 215 216

PARAFAC and statistical analyses 217
Exploratory analysis of fluorescence EEM data was conducted using parallel factor 218 analysis (PARAFAC) to decompose the number, shape, and amounts of underlying 219 spectral components among samples. PARAFAC was conducted using the drEEM 220 (version 2.0) and N-way (version 3.20) toolbox (Murphy et al., 2013) within the 221

MATLAB R2013a environment. 222
To aid decomposition and provide greater variance within the dataset, 223 additional EEMs (total n = 645) collected across a wide range of stream and river 224 environments from the Kolyma River Basin were included and analyzed alongside the 225 Arctic-GRO fluorescence dataset. PARAFAC modeling was performed after 226 normalizing each EEM to its total signal (to unit norm) by dividing by the sum of the 227 squared values of all variables in the sample, and imposing non-negativity constraints, 228 thus negating problems caused by large concentration gradients apparent in seasonal 229 samples (Murphy et al., 2013). The number of components within the model was 230 validated using all of the techniques recommended in Murphy et al., (2013), including 231 examination of systematic variation in the dataset, visualization of spectral loadings, 232 and split half analysis. The final model was successfully validated using four splits of 233 the data and three validation tests across six different dataset halves (S 4 C 6 T 3 ) 234 (Harshman and Lundy, 1994;Murphy et al., 2013). Fluorescence loadings were 235 calculated after normalizing the dataset ensuring unscaled model scores were 236 recovered.

Constituent Flux Calculations 246
Constituent fluxes were estimated using the USGS LoadEstimator software 247 (LOADEST) within the LoadRunner software interface (Booth et al. 2007;Runkel et 248 al., 2004). LOADEST calculates daily constituent flux estimates by generating 249 relationships between measured discharge and element concentrations and was run as 250 in Holmes et al., (2012 between the discharge measurement station and sample location as has been 257 previously described (Holmes et al., 2012). Any gaps in the discharge data were filled 258 by interpolation, however there were no gaps during peak flow on any river. 259 260

Spatial and temporal patterns in chemical fractions of DOC 263
Total DOC concentrations ranged over the study period from 2.6 to 17.5 mg L -1 .

P r o v i s i o n a l
The composition of DOC was further characterized by calculating specific 278 UV-visible absorbance (SUVA 254 ) of DOC and its major chemical fractions (Table 1). 279 Mean SUVA 254 values of total DOC varied considerably among rivers with lowest 280 values measured in the Mackenzie River (2.5 L mgC -1 m -1 ) and highest in the Yenisey 281 River (3.4 L mgC -1 m -1 ). Mean SUVA 254 values of the HPOA fraction also varied 282 among rivers with lowest values observed in the Kolyma River (3.6 L mgC -1 m -1 ) and 283 highest in the Yenisey (4.3 L mgC -1 m -1 ), and were consistently higher than bulk DOC 284 highlighting the greater number of highly aromatic compounds represented by this 285 fraction. The SUVA 254 values of the HPI (1.2 to 2.3 L mgC -1 m -1 ) and TPIA (2.0 to 286 3.1 L mgC -1 m -1 ) fractions were less variable across rivers, and lower than bulk DOC, 287 indicating the presence of a lower relative number of aromatic moieties (Table 1). 288 The contribution of HPOA to the total DOC pool was generally highest during 289 spring months with maximum contributions varying considerably between rivers (47 290 to 60 %; Table 1). Percent contributions of the HPOA fraction were typically lowest 291 during winter flow periods. No clear seasonal differences in the fraction of HPOA 292 present were observed in the Ob' River (55 ± 3 to 57 ±1 %). Mean SUVA 254 values of 293 total DOC were consistently highest across all rivers during spring months, 294 intermediate during summer months (2.3 to 3.1 L mgC -1 m -1 ), and lowest in winter 295 across all rivers (1.5 to 3.0 L mgC -1 m -1 ; Table 1). SUVA 254 values of the HPOA 296 fraction followed similar seasonal trends as total DOC. TPIA and HPI SUVA 254 297 values displayed less clear seasonal patterns (Table 1). 298 299

Spatial and temporal patterns in lignin phenols 300
Mean lignin phenol concentrations (∑ 8 ) varied significantly among the six rivers, with 301 lowest concentrations observed in the Mackenzie River (9.5 µg L -1 ) and highest in the 302 Lena River (70.0 µg L -1 ; Table 2). Highest carbon normalized lignin yields (Λ 8 ) were 303 observed in the Yenisey River, mostly due to lower mean DOC concentrations 304 relative to the Lena River (Table 2). Lowest mean Λ 8 values were measured in the 305 Mackenzie River (0.19 (mg(100 mg OC)) -1 ). Lignin values measured in this study 306 were consistent with prior measurements in the Yukon and Russian arctic rivers 307 (Lobbes et al., 2000;Spencer et al., 2009) but notably lower than lignin measurements 308 from the earlier PARTNERS project (Amon et al., 2012). 309 The Lena and Yenisey rivers displayed lowest mean cinnamyl (C) to vanillyl 310 (V) phenol ratios (C/V), indicative of greater contributions of woody versus non-311 woody sources to bulk DOM of these rivers (Hedges and Mann, 1979). Highest C/V 312 ratios were measured in the Mackenzie and Ob' Rivers (Table 2). Spatial variability in 313 syringyl (S) to vanillyl ratios (S/V) mainly mirrored those of C/V, except for higher 314 S/V values in the Yukon River as compared to the Mackenzie River (Table 2). Higher 315 S/V ratios are indicative of greater proportions of angiosperm versus gymnosperm 316 sources to DOM (Hedges and Mann, 1979). 317 Acid to aldehyde ratios (Ad/Al) have been suggested to provide evidence of 318 the relative degree of DOM degradation, with higher ratios indicating greater 319 degradation of plant tissues (Hedges et al., 1988;Hernes and Benner, 2003;Opsahl 320 and Benner, 1995). Mean ratios of vanillic acid to vanillin (Ad/Al)v ranged from 1.07 321 in the Mackenzie River to 1.48 in the Yukon River ( Table 2). Ratios of syringic acid 322 to syringaldehyde (Ad/Al)s varied from 0.84 in the Yenisey to 1.06 in the Mackenzie 323

River. 324
∑ 8 values among rivers were strongly linearly related to runoff (R 2 = 0.69), 325 suggesting terrigenous DOM export dynamics were largely controlled by hydrology. 326 Accordingly, highest ∑ 8 concentrations were recorded across all rivers during the 327 spring freshet and lowest concentrations during base flow winter conditions. Highest 328 individual ∑ 8 values were measured in the Lena River (120 µg L -1 ) and lowest in the 329 Kolyma and Yukon Rivers (3.8 µg L -1 ). During the freshet, Λ 8 yields were between 330 2.6 and 4.8 times higher than winter Λ 8 values across sites; the Yenisey River 331 displayed the least variability and the Yukon the greatest (Table 2). 332 C/V and S/V ratios generally declined with increasing runoff across all rivers 333 (Table 2). Acid aldehyde ratios (Al/Ad) were highly variable among rivers, generally 334 increasing with greater runoff, yet in some cases (e.g. Lena River) demonstrated 335 opposing patterns. Highest Ad/Al ratios during the spring freshet may represent the 336 export of greater quantities of largely 'fresh' microbially unprocessed DOM relative 337 to later in the year (Hernes et al., 2007;Spencer et al., 2008;Amon et al., 2012). 338 339

Chromophoric and fluorescence DOM of arctic rivers 340
The absorbance coefficient of CDOM at 350 nm (a 350 ) ranged from 2.3 to 42.6 m -1 341 among rivers and seasons, and similar to DOC and ∑ 8 concentrations, generally 342 increased with greater freshwater runoff (R 2 = 0.57, p < 0.001, n = 60; Supplemental 343 Table 1). Spectral slope values (S 275-295 , S 290-350 , S 350-400 ) steepened with decreasing 344 runoff, in good agreement with previous studies Stedmon et al., 345 2011a), indicating the export of lower molecular weight material, or DOM with 346 decreasing aromaticity as discharge rates decline (Blough and Del Vecchio, 2002;347 Blough and Green, 1995). The slope ratio (S R ) showed an opposing pattern to spectral 348 slopes, declining at higher runoff rates thus confirming an increase in DOM molecular 349 weight during the spring freshet and reduction during winter base flow months 350 (Helms et al., 2008;Spencer et al., 2010a;  Tucker, 1951) the excitation and emission spectra of previously identified 368 components from 37 independent studies (Table 3). 369 Four of the seven AG components were very closely related (TCC ≥ 0.97) to 370 those reported in the five-component Horsens catchment model (Murphy et al., 2014). 371 For each of the seven components identified here, at least three independent studies 372 had previously identified statistically similar spectra, except for component AG3 for 373 which there was only a single match. Table 3 provides information and a description 374 of the seven components identified. 375

P r o v i s i o n a l
AG3 was closely related (TCC = 0.97) to C1 in Murphy et al., (2014) where it 376 was identified as displaying an emission spectrum identical to syringaldehyde, a 377 product of lignin breakdown. The AG model shared two components with a five-378 component PARAFAC model explaining florescence DOM collected from five of the 379 major Arctic rivers sampled here over 2004(Walker et al., 2013. AG6 was 380 similar (TCC >0.95) to C1 from this model (Walker et al., 2013), while AG7 was 381 identical to C5, which can be described as tryptophan-like and has been commonly 382 associated with biological production in surface waters (Determann et al., 1994). 383 Two AG model components were also highly related to components 384 previously reported in sea ice. AG1 was comparable to the terrestrially-derived 385 component C2 found within Baltic sea ice (Stedmon et al., 2007). AG1 was also 386 identical to C3 in coastal Canadian Arctic waters, which proved to be highly 387 positively correlated with ∑ 8 (Walker et al., 2009). AG5 was identical to C6 in a study 388 of Antarctic sea ice brines (Stedmon et al., 2011b), and is similar to the commonly 389 described 'M' peak across a wide range of environments (Coble, 1996;2007;Fellman 390 et al., 2010). 391 Component AG4 contributed the greatest (23.4% ±0.6%) and AG7 the lowest 392 percentage (5.1% ±0.3%) towards total DOM fluorescence across all rivers and 393 seasons. In contrast to previous studies, no consistent pan-arctic seasonal or spatial 394 patterns were apparent in the fluorescence loadings or percent contribution of any of 395 the seven components (Walker et al., 2009). Individual patterns in fluorescence were, 396 however, observed across rivers and seasons. Component AG1 contributed a 397 substantially higher proportion of total fluorescence during the summer months in the 398 Kolyma (20.5% ± 0.1%) and Lena Rivers (20.6% ± 0.6%) relative to each of the other 399 rivers (16.8 to 17.7%), yet comprised similar amounts during the rest of the year. 400 Similarly, the proportion of AG3 was significantly higher in the Kolyma (23.2% ± 401 0.8%) and Lena (19.1% ± 1.4%) Rivers relative to the others (11.3% to 17.6%) during 402 the summer months alone. Opposing patterns were observed in AG6, with 403 significantly lower proportions in the Kolyma (4.4% ± 0.1%) and Lena Rivers (5.3% 404 ± 0.6%) relative to the others, in particular the Yenisey (9.7% ± 1.2%) and Ob' Rivers 405 (8.4% ± 1.8%). The Mackenzie River contained high proportions of AG7 during the 406 summer months (6.7% ± 1.1%) relative to all other rivers (3.9 to 5.2%).   (Table 4). This indicates that the relative amount of non-420 chromophoric DOM varies across Arctic rivers, and suggests that the proportion of 421 DOC per unit CDOM within individual river basins should in the future be separately 422 determined (Table 4). Interestingly, we also find that the variability in river-specific 423 slope and intercepts were well-explained by total annual river discharge, with 424 increasing discharge resulting in higher DOC:a 350 intercepts (R 2 = 0.58 ; p <0.05, not 425 shown) and shallower slopes (R 2 = 0.72, p <0.05, not shown). The relationship 426 between annual discharge and DOC:a 350 intercepts improved significantly (R 2 = 0.98 ; 427 p <0.01) with the exclusion of the Mackenzie River. Thus, greater dilution of DOM 428 and export of non-chromophoric organics occurs with increasing total discharge. The 429 different relationship observed in the Mackenzie may be due to its relatively low 430 DOM yield, high abundance of suspended sediments, as well as high proportion of 431 lakes relative to other watersheds (Stedmon et al. 2011a). 432 HPOA fraction was closely related to S 275-295 across rivers, with the relative 433 proportion of HPOA decreasing with steepening slope (R 2 = 0.65; p < 0.001, n = 58), 434 as previously reported across five of these rivers in 2004(Walker et al., 2013. 435 This suggests that average DOM molecular weight and aromaticity decreases as the 436 proportion of HPOA declines, in good agreement with a number of previous studies 437 (Neff et al., 2006;O'Donnell et al., 2012;Spencer et al., 2012;Striegl et al., 2007).

Optical measurements and lignin concentration and composition 476
CDOM (a 350 ) measurements were highly correlated to ∑ 8 across all six rivers basins 477 (R 2 = 0.93; p <0.001; n = 31; Figure 3b). This represents the first pan-arctic 478 relationship to be reported between a 350 and ∑ 8 across all six major rivers. The 479 observed linear relationship (∑ 8 = -8.06 ± 2.71+ (2.80 ± 0.14a 350 )) was similar to, yet 480 displayed a slightly higher slope, than reported in Spencer et al., (2008)  The differences in ∑ 8 concentrations may be due to methodological differences, as 488 suggested by Walker et al., (2013), and raises concern over future potential in 489 comparing datasets. For example, comparison of data from Spencer et al., (2008) Templier et al., 2005). This was confirmed by a 501 significant positive correlation between proportion HPOA and Λ 8 yields (R 2 = 0.71, p 502 < 0.001, n = 29; not shown). Lignin phenol biomarkers thus appear capable of 503 providing information on the biogeochemical cycling of the entire hydrophobic DOM 504 pool, which comprises up to two-thirds of aquatic DOM. 505 Lignin phenol C/V ratios increased with steepening S 275-295 values (R 2 = 0.54; 506 p < 0.001, n = 31) and declining total SUVA 254 values (R 2 = 0.48; p < 0.001, n = 31) 507 (Tables 1 and 2 and Supplementary Table 1). The overall decline in C/V ratios with 508 increasing freshwater runoff appears to represent increased contributions of lignin 509 from litter and surface soil layers alongside greater proportions of aromatic and higher 510 molecular weight DOM export (Hedges and Mann, 1979). These findings however 511 appear counterintuitive relative to what we currently understand about hydrologic 512 flow paths and sources of DOM to aquatic systems. Non-woody litter tissues 513 associated with surficial, predominantly overland flow paths are expected to impart 514 higher C/V ratios and lower degradative alteration than observed in DOM exported 515 during base flow conditions derived from deeper flow paths. Physical processes, such 516 as leaching and sorption, can also influence lignin phenol ratios (Hernes et al., 2007;517 2008) and may therefore be responsible for the observed trends. S/V and acid to 518 aldehyde ratios did not correlate closely with spectral slope, SUVA 254 , or S R values. 519 The overall trends in lignin phenol composition we report are similar to those 520 previously shown across Arctic rivers (Amon et al., 2012;Spencer et al., 2008;, 521 and demonstrate a shift from predominantly modern surface-derived and lignin-rich 522 DOM during the spring freshet to older, less lignin-rich DOM under base-flow winter 523 conditions. 524

P r o v i s i o n a l
No pan-arctic relationships were observed between fluorescent PARAFAC 525 component loadings and lignin phenol concentration or composition measures 526 (Supplemental Tables 2 & 3). A weak yet significant relationship was however found 527 between the relative proportions of components AG3 and AG4 to ∑ 8 concentration 528 (R 2 = 0.18; p < 0.02) but again was significantly stronger across the Kolyma and Lena 529 Rivers in particular (R 2 = 0.79 ; p < 0.001). FI values positively correlated with 530 increasing C/V ratios (R 2 = 0.48, p < 0.001, n = 30) confirming losses in the 531 proportion of woody tissues with increased autochthonous or less aromatic DOM 532 supply. 533 534

Linking optical properties to Arctic river DOM composition 535
Underlying patterns and relationships between optical DOM parameters, DOC, and 536 lignin were further explored using principle component analysis (PCA), which can 537 identify the structure of data that best explains the variance within the dataset. The 538 optical properties of DOM varied with season across all rivers, as demonstrated by 539 PCA plots containing PARAFAC fluorescence components (percent contribution) and 540 spectral slope information. The addition of FI and S R values added little additional 541 information to the PCA analyses. Furthermore, SUVA 254 followed identical patterns 542 to each of the spectral slope parameters and its inclusion led to similar PCA plots. 543 These indices were therefore omitted from the final PCA model for clarity. Three 544 principle components (PCs; eigenvalue > 1) were identified that together explained 545 80% of the total variance in the optical data (PCopt 1-3). PC1opt was related to 546 increasing fluorescence contributions from AG3, AG1, and AG5, but decreasing 547 contributions from AG6, AG4, and AG2 (Fig. 4). Components AG1, 3, and 5 548 represent DOM fluorescence signatures that have all previously been reported to be 549 susceptible to microbial processing, or to be a byproduct of vascular material 550 degradation (Table 3 and references herein). These fluorescence signatures may 551 therefore represent indicators of 'degraded' or processed humic-like components. In 552 contrast, components AG 2, 4, and 6 appear to represent more unreactive and stable 553 components, previously being described as refractory in nature and shown not to co-554 vary with bacterial production (Table 3). PC1opt may therefore reflect potential 555 reactivity or be an indicator of prior DOM processing. PC2opt appeared to be related 556 to the shifting molecular weight of DOM, as indicated by strong relationships with 557 changes in all spectral slopes (and SUVA 254 , not shown), whereas PC3opt was 558 positively related to increased protein-like or phenolic DOM (AG7) and decreasing 559 contributions from humic-like DOM (AG2 and AG1). PCA models run with only 560 PARAFAC components contained two principle components, each indistinguishable 561 from PC1opt and PC3opt, demonstrating that information on DOM potential 562 reactivity and the relative contribution of protein-like versus humic-like could be 563 obtained from fluorescence measurements alone. 564 Seasonal changes in DOM composition across all six rivers were most clearly 565 separated along the PC2opt axis, with spring waters containing higher molecular 566 weight material with shallower spectral slopes than summer and winter month waters. 567 Positive scores on PC3opt during spring and winter months, relative to summer, 568 suggest greater contributions of protein-like or phenolic material (as inferred by the 569 proportion of component AG7), potentially representing reductions in allochthonous 570 supply or increased export of fresh organics from surface layers, respectively. No 571 clear separation among the six different rivers was apparent with optical properties 572 alone across any of the PC axes. 573

P r o v i s i o n a l
To examine if the observed trends in DOM optical properties were related to 574 geochemical changes in organic matter we conducted a separate PCA incorporating 575 all lignin phenol and DOC fractionation variables. We subsequently compared the 576 identified PCs with those extracted from optical measurements alone. Two PCs 577 (PCmol 1-2) were identified, in combination explaining 67 % of the total variance in 578 geochemical composition (Fig. 5) therefore also be contained in the relative ratio of more or less reactive or degraded 600 PARAFAC components. Interestingly, (Ad/Al)v ratios have also been shown to 601 correlate with the average 14 C age of DOC within these Arctic rivers (Amon et al., 602 2012). Fluorescence measurements may therefore provide information pertaining to 603 both the age and degradation history of DOM across Arctic systems. Broad patterns in 604 the temporal variability of DOM composition over pan-arctic scales therefore appear 605 best captured using simple CDOM spectral slope and SUVA 254 measurements. 606 Information on DOM processing, source, and age may instead be contained within 607 CDOM fluorescence spectra and the relative contributions of PARAFAC components. 608 609

Modeling terrestrial biomarkers with optical measurements 610
We ran a series of multiple linear regression models with the aim of predicting Λ 8 , 611 ±0.002*a 350 )+(0.074 ±0.027*%AG7)). S/V could not be explained with absorbance 628 measurements alone. The ability to predict lignin composition as well as 629 concentration using fluorescence measurements has previously been reported using 630 partial least squares models of samples collected over a two year period on the 631 Sacremento River/ San Joaquin River Delta, California . The 632 authors demonstrated that the most significant predictive capability for lignin was 633 within the commonly referred to protein-like fluorescence region (similar to our 634 component AG7). Fluorescence of propylphenol monomers, that structurally comprise 635 lignin, can generate fluorescence signatures in a similar region to amino acids and in 636 the region known as 'protein-like', thus our results may indicate information obtained 637 from changes in phenolics rather than amino acid or proteins . 638 Therefore, it seems that rapid, inexpensive optical measurements may be capable of 639 acting as a proxy for dissolved lignin compositional parameters as well as 640 concentration across pan-arctic scales and catchments. The combination of 641 absorbance and fluorescence metrics can also add predictive power when attempting 642 to predict shifts in the composition of terrigenous DOC. 643 644

Improving terrigenous OC export estimates 645
The absorbance coefficient at 350nm (a 350 ) has previously been shown to be a 646 sensitive and inexpensive proxy for lignin phenol concentration across a range of 647 freshwater environments within Arctic river basins 648 has led to significantly higher and better constrained DOC export estimates, 650 particularly after the inclusion of samples from across the spring freshet period (e.g. 651 Köhler et al., 2003;Striegl et al., 2005;Holmes et al., 2012). Here, we investigate if 652 the combination of a lignin proxy with high-resolution monitoring of a 350 over Arctic 653 river hydrographs may be used to develop improved estimates of pan-arctic 654 terrigenous DOC export, hereby refining land-to-ocean carbon flux estimates. 655 CDOM-derived lignin phenol concentrations (lignin 350 ) were calculated using 656 the linear regression of ∑ 8 and a 350 (Fig. 3b). Lignin 350 values were derived from a 350 657 measurements taken from waters collected over the main Arctic-GRO sampling 658 campaign and additional high-resolution samples taken over the freshet hydrographs. 659 Inclusion of near-daily absorbance measurements collected over the peak discharge 660 period alongside measurements spanning the entire year was crucial in adequately 661 constraining fluxes during the spring freshet, when the majority of annual lignin 662 export is expected (Amon et al., 2012;Spencer et al., 2008). Lignin 350 concentrations 663 calculated for samples with concurrent ∑ 8 measurements were highly correlated across 664 all Arctic rivers (r 2 = 0.92, p < 0.01, n = 31, Standard error of estimate, SE E = 8.8%) 665 demonstrating the robust nature of this approach. Gg y -1 in the Lena River (Table 5). The Lena, Yenisey, and Ob' Rivers export > 85 % 671 of the total annual lignin discharge from the six largest Arctic rivers, a proportion that 672 is very similar to that found by Amon et al., (2012). Flux estimates using lignin 673 phenol concentrations measured using identical methods and approaches compared 674 well. Our mean annual Yukon River lignin flux derived for 2001-2009 (5.4 ± 1.7 Gg 675 y -1 ; Table 5)  Gg y -1 (PA1) to 185.3 Gg y -1 (PA2; Table 5) across these two geographic regions. 690 Dissolved lignin concentrations have previously been applied as a tracer of 691 terrigenous DOM to the Arctic Ocean (Benner et al., 2005;Fichot et al., 2013;Opsahl 692 et al., 1999) and used to estimate turnover rates of terrigenous DOC in the ocean 693 (Hernes and Benner, 2006;Opsahl et al., 1999). Applying our pan-arctic flux (derived 694 using our lignin 350 proxy) and assuming Arctic Ocean lignin concentrations ranging 695 between 84 to 320 ng L -1 (Opsahl et al., 1999), we calculate the residence time of 696 terrigenous DOC in polar surface waters to be in the order of 7 months to 2.5 years. 697 This compares well, yet slightly shorter than residence time estimates of <1 to 4 years 698 calculated with comparable freshwater fluxes but scaled from the Yukon River alone 699 . Assuming the export of lignin phenol concentrations twice as 700 high, similar to those reported by Amon et al., (2012), would result in even faster 701 residence time estimates of < 4 months to 1 year. Overall, the short timeframes 702 identified by these studies indicate either rapid losses of terrigenous DOC, via 703 microbial, photochemical, or flocculation processes, or faster physical transport from 704 Arctic Ocean waters to the North Atlantic than previously thought.  Table 3. 761   Table 1. Total dissolved organic carbon concentrations (DOC) and major chemical fractions of DOC and fraction-specific ultraviolet absorbance (SUVA 254 ) across the six major arctic rivers (mean ± standard error) during Spring (May and June), Summer (July through to October) and Winter (November through to April). Hydrophobic acids (HPOA), transphilic acids (TPI) and hydrophilic organic matter (HPI) presented as percentage of total DOC concentrations and the sample mass of each fraction (HPON comprises the remaining <10 % of the DOC pool).

Site
Season   Table 3. Excitation and emission maxima (Ex max / Em max ) of the seven components identified using parallel factor analysis of DOM fluorescence (Fig. 2). Description of previously identified components displaying similar optical properties (TCC > 0.95; see text for details). *ID number refers to assigned study number in OpenFluor (http://www.openfluor.org).