Linking landscape variables, hydrology and weathering regime in Taiga and Tundra

ecoregions of Northern Sweden

Erik Smedberg

Doctoral Thesis Department of Applied Environmental Science

Stockholm University Stockholm 2008

Front page: Torneträsk, photo by C. Humborg

© Erik Smedberg, Stockholm 2008

ISBN 978-91-7155-652-3 pp. 1-31

Printed in Sweden by Universitetsservice US-AB, Stockholm, 2008

Abstract

This thesis focuses on two environmental issues; carbon and dissolved constituent fluxes in high-latitude watersheds as affected by global warming, and carbon and dissolved constituent fluxes as affected by river regulation. Historically, high-latitude watersheds have been regarded as a carbon sink with soil carbon accumulating due to slow decomposition of organic matter at low temperature and also with organic matter incorporated into permafrost layers and rendered inactive. This carbon sink is now believed to turn into a carbon source, and a potential release of soil derived carbon would act as positive feedback to climate warming. However, higher temperatures thawing permafrost soils and altering water flow paths would allow more water to percolate down to deeper soil layers. Here, some of the released carbon could be “consumed” in enhanced weathering and exported as bicarbonate to the sea. This was proved by a hydrological mixing model showing that altered water flow paths in high latitude watersheds could significantly counterbalance the predicted positive feedback resulting from thawing permafrost soils. Recent studies have shown that vegetation plays an important role in the global Si cycle and the riverine dissolved silicate (DSi) export fluxes. Vegetation-covered riparian zones in headwater areas appear to have a significant role for the dissolved constituent fluxes. This thesis shows that vegetation cover, carbon and weathering fluxes in taiga and tundra watersheds are intimately linked, and that higher concentrations of weathering products are found in taiga and tundra rivers with larger areas of coniferous forest and peat cover in their watersheds. These landscape elements can thus be regarded as “hot spots” of river loading with dissolved constituents. River regulation affects ca. 30% of the river runoff to the global ocean and has adverse effects for the ecosystems in the receiving coastal water bodies. We tested the hypothesis that damming leads to a depletion of major elements also in oligotrophic river systems as a consequence of changes in landscape elements by comparing an unregulated river with a regulated river. A loss of upper soils and vegetation through inundation, loss of soil through littoral erosion, underground channelling through tunnels, as well as reduction of water level fluctuations in reaches between the dams, prevents the contact of surface waters with vegetated soil, and consequently reduces weathering fluxes. The hypothesis that the lower fluxes of DSi in the regulated river could also be explained by biological uptake was then tested using the RIVERSTRAHLER model. Budget calculations based on the model results indicate a significant reduction of phosphorus (over 30%) and Si (close to 25%) fluxes in the Luleälven watershed as a result of regulation. The budget indicates that about 10% of this reduction can be attributed to the flooding of the fluvial corridor. The remaining 15% was attributed to diatom blooms in the reservoirs. A further more detailed study of landscape elements affected by river regulation for the headwater areas of the river Luleälven showed that only 3% of the surface area has been inundated by reservoirs. However, some 10% of the soil volumes and aggregated forest and wetland areas have been lost due to damming. Moreover, our estimates indicate that ca. 37% of the deciduous forest has been inundated by the four major reservoirs built in the Luleälven headwaters. This land cover class “deciduous forest” is almost synonymous with the riparian zone and occurrences of organic rich soil layers in these headwaters acting as hot spots for DSi and TOC loading. Such a significant loss of hot spots for river loading may indeed explain the observed lower DSi fluxes in the regulated watersheds of northern Sweden.

List of Papers

This thesis is based on the following papers:

Paper I.

Humborg, C., Smedberg, E., Blomqvist, S., Mörth, C-M., Brink, J., Rahm, L., Danielsson, Å., Sahlberg, J. (2004) Nutrient variations in boreal and subarctic

Swedish Rivers: Landscape control of land-sea fluxes. Limnology and Oceanography 49(5), 1871

Paper II.

Smedberg, E., Mörth, C-M., Swaney, D. P., Humborg, C. (2006) Modelling hydrology

and silicon-carbon interactions in taiga and tundra biomes from a landscape

perspective: Implications for global warming feedbacks. Global Biogeochemical Cycles, 20, GB2014

Paper III.

Humborg, C., Blomqvist, S., Avsan, E., Bergensund, Y., Smedberg, E., Brink, J., Mörth, C-M. (2002) Hydrological alterations with river damming in

northern Sweden: Implications for weathering and river biogeochemistry. Global Biogeochemical Cycles 16(3), 1039

Paper IV.

Sferratore, A., Billen, G., Garnier, J., Smedberg, E., Humborg, C., Rahm, L. (2008) Modelling nutrient fluxes from sub-arctic basin: comparison of pristine vs. dammed

rivers. Journal of Marine Systems. In press.

Paper V.

Smedberg, E., Humborg, C., Jakobsson, M., Mörth, C-M. Landscape elements and river chemistry as affected by river regulation – a 3D

perspective. Manuscript

Paper I is reproduced by kind permission of American Society of Limnology and Oceanography, Paper II and III is reproduced by kind permission of American Geophysical Union (AGU) and Paper IV is reproduced by kind permission of Elsevier. In Paper I, I assembled GIS data on vegetation cover, soil type and bedrock type, which could provide a uniform classification for all the watersheds. I was also involved in the interpretations of data and their discussions.

In Paper II, I was involved in all the work, from setting up the model to writing the paper.

In Paper III, I calculated the changes in landscape, i.e., the changes of the vegetation cover, lake/reservoir areas and the changes along the shorelines, resulting from the building of the Akka reservoir. I evaluated historical maps (100+ years), showing pre-damming conditions, and modern maps, by means of GIS compilations. I also took part in the discussions of the results.

In Paper IV, I was involved in compiling GIS data and information on river network, river regulation, stream flow data and landscape elements.

In Paper V, I have been involved in all the work including formulating the hypothesis, GIS work and writing.

Contents

Background ................................................................................9

High-latitude watersheds of the Northern Hemisphere – natural settings and major threats .................................................................................9

Weathering regimes of high-latitude watersheds................................11

Linkages between hydrological flow paths and river chemistry in high-latitude watersheds ............................................................................13

Summary of the papers ............................................................16

Linkages between landscape elements and weathering regimes in high-latitude watersheds (Paper I and Paper II) .................................16

Linkages between landscape elements, hydrological alterations and weathering regimes in high-latitude watersheds (Paper III, Paper IV, Paper V).............................................................................................20

Conclusions and Future Perspectives......................................24

Acknowledgements ..................................................................25

References ...............................................................................26

9

Background

High-latitude watersheds of the Northern Hemisphere – natural settings and major threats

The Arctic tundra is located in the northern hemisphere, encircling the North Pole and extending south to the coniferous forests of the taiga. Approximately two thirds of the taiga is found in Siberia and the rest is situated in Alaska, Canada and the northern parts of Scandinavia (Figure 1). Large areas of the taiga and tundra ecosystems are permafrost areas with a permanently frozen soil layer under an annually thawing top layer, the so-called active layer. Historically these high-latitude areas have been looked upon as a carbon sink. Soil carbon has been accumulating since the last glaciations, due to the relatively slow decomposition of organic matter at low temperature under anaerobic soil conditions in peatlands (Houghton et al. 2001). In addition, with the gradual build-up of the soil layers, organic matter has been incorporated into the permafrost layer and rendered inactive. This carbon sink is now believed to be a carbon source with global warming that most Global Climate Models (GCM) foresee to be most significant at the poles (Cai 2005, Kaplan & New 2006).

Figure 1. Permafrost distribution in the Arctic

10

Taiga and tundra areas of the northern hemisphere are believed to hold 450 GT, or an equivalent of 70 annual anthropogenic emissions, approximately one third of the world’s total soil organic carbon stored within the top meter of the soil profile (Gorham 1991). New estimates of soil carbon stored in permafrost areas in thick mineral soil deposits (loess) point to a further 500 GT carbon storage (Zimov et al. 2006). For the northern hemisphere, temperature increases above the global average have been predicted at high latitudes (Bengtsson 1997), and increases in water discharge (Peterson et al. 2002) have already been recorded in the largest Eurasian rivers. A possible global warming in taiga and tundra biomes is believed to result in a massive release of dissolved organic carbon (Freeman et al. 2001, Dutta et al. 2006) to coastal waters. It has been suggested that such a potential release would act as a positive feedback to climate warming. However, it has also been postulated that increased DOC trends are a result of decreased atmospherically deposited anthropogenic sulphur and sea salt (Monteith et al. 2007). The second environmental issue dealt with in this thesis is river regulation that affect ca. 30% of the river runoff to the global ocean (Milliman 1997, Vörösmarty & Sahagian 2000) with adverse effects for the ecosystems in the receiving coastal water bodies (Ittekkot et al. 2000). We focused our studies on damming effects in high-latitude watersheds. Large areas covered by permafrost also imply that high-latitude watersheds generally have less vegetation cover compared to lower-latitude watersheds. Typical vegetation found in the taiga belt is coniferous and mixed forest, whereas in the tundra belt ice cover and bare areas, herbaceous vegetation, low deciduous forest and shrubs (mainly birch and willow) and peatlands are dominating (Snow 2005a, 2005b). In watersheds with less vegetation cover, changes in landscape elements such as wetlands and forests along riparian zones, which were described as hot spots of river loading with dissolved constituents (Laudon et al. 2004), could be more significant as compared to vegetation covered watersheds in more temperate regions. A significant change in these landscape elements has occurred with river regulation, especially in the Scandinavian and Canadian watersheds (Nilsson et al. 2005), whereas the high-latitude Siberian large watersheds are less regulated. This thesis focuses on general large-scale linkages of landscape elements, hydrology and weathering in taiga and tundra ecoregions in northern Sweden with potential bearing for Arctic taiga and tundra watersheds worldwide. The aim is to identify and describe links between various landscape elements, hydrology, weathering and carbon fluxes. The relationships found are then used to infer future changes in river chemistry as the result of permafrost melting and river regulation.

11

We examined carbon in its dissolved inorganic forms (DIC) and organic forms, e.g. dissolved organic carbon (DOC) and total organic carbon (TOC). The other element we focused on was silicon, since it is an important constituent regulating diatom production in aquatic systems (Dugdale & Wilkerson 1998, Ittekkot et al. 2000). With river discharge being the main process for redistributing dissolved silicate (DSi) from land to sea (Billen et al. 1991), and with approximately 80% of the global ocean Si supply coming from riverine export (Treguer et al. 1995), hydrology is an important component. Because of its vital function for the biological pump, any changes in riverine DSi export may alter phytoplankton species composition (Officer & Ryther 1980, Turner et al. 1998) but also, and more importantly, the amount of carbon sequestered in aquatic systems (Conley et al. 1993). Moreover, silicon dioxide is one of the most abundant components of the Earth’s crust and occurs in silicate minerals in association with igneous, metamorphic and sedimentary rocks that are being continuously subjected to physical and chemical weathering. The processes associated with weathering and the products released, form the basis of silicon biogeochemistry and its interactions with other elements. A fundamental mechanism controlling the biogeochemical Si cycle is thought to involve CO2 in the ocean-atmosphere system through continental weathering.

Weathering regimes of high-latitude watersheds

Weathering regimes in high-latitude areas are heavily affected by climatological and hydrological conditions and can be regarded as weathering limited, i.e. the transport capacity within the watersheds exceeds the weathering capacity (Drever 1997). Physical weathering in high-latitude areas is to a large extent caused by ice and freezing, and the long winters with sub-zero conditions effectively restrict water flow and thus, also the chemical weathering processes. The weathering rate varies greatly and is affected by a number of factors, for example, mineral structure of the solid phase, temperature, vegetation and pH in the soil (Berner & Berner 1996). For chemical weathering water is a crucial component in the actual weathering reactions, but also for flushing within the system and transporting the weathered products away from the weathering site. This transport not only increases the concentration levels in the receiving rivers, but also enhances the weathering rate. Transport of weathering enhancing elements from the organic rich upper soil layers is facilitated by water infiltrating and percolating down to deeper mineral soil layers. TOC with its functional carboxylic groups acts as a weak organic acid (Hruska et al. 2003) and its effect of lowering the pH as such will increase weathering. There are additional mechanisms by which soil organic carbon and/or vascular plants may enhance the weathering rate, and there are three main processes discussed in the literature; root exudation of organic acids (Grayston et al. 1997), the activity of

12

ectomycorrhizal fungi (van Breemen et al. 2000, Landeweert et al. 2001), and associated mineral dissolution by bacteria (Bennett et al. 2001). Continental weathering is regarded as a fundamental mechanism for control of CO2 in the ocean-atmosphere system, because the net effect of silicate mineral weathering is the consumption of soil carbon derived from the atmosphere through photosynthesis into dissolved HCO3

-. In general, chemical weathering of silicate and carbonate rocks can be represented as follows:

2CO2 + 3H2O + CaSiO3 → Ca++ + 2HCO3- + H4SiO4

2CO2 + 3H2O + MgSiO3 → Mg++ + 2HCO3- + H4SiO4

and for carbonates CO2 + H2O + CaCO3 → Ca++ + 2HCO3

- 2CO2 + 2H2O + CaMg(CO3)2 → Ca++ + Mg++ + 4HCO3

- Whether weathering is a carbon sink or source depends on the temporal scale: For a time frame of less than 100,000 years, rock weathering is important in the removal of CO2 from the atmosphere; for a time frame of a million years, carbon supplied from the land by carbonate weathering is returned to the atmosphere by calcite precipitation (Gaillardet et al. 1999, Harrison 2000). In addition, weathered Na and K silicate minerals can undergo reverse weathering in the sea, where formation of Na and K silicates convert HCO3

- to CO2, and thus, returning CO2 to the atmosphere (Berner 1992). Hence, in a long-term geological perspective it is only the weathering of Ca and Mg silicates, mainly plagioclase as the most important mineral in northern Sweden, that leads to a long term removal of CO2 from the atmosphere and a deposition of carbon in the sea (Gaillardet et al. 1999). However, previously stored inactive organic soil-carbon released due to global warming could be either “consumed” in chemical weathering processes and exported as bicarbonate to the sea, or respired along its way as TOC or DOC (Freeman et al. 2001). Whereas the latter process is a potential positive feedback to global warming, the former can be regarded as a negative feedback to global warming. Therefore, in an assessment of the potential impact of climate change on carbon fluxes and global warming feedbacks, both organic, as well as inorganic carbon forms, have to be considered because carbon plays a role both as a weathering agent (soil CO2 from degradation of organic matter) and as a weathering product (alkalinity or HCO3

-). In other words, HCO3- and CO3

2- can be regarded as residuals from soil respiration (Cole et al. 2007), since gaseous

13

CO2 from soil respiration of organic material is consumed in weathering reactions and converted to dissolved constituents locked in the aqueous phase. Therefore, C fluxes both in its inorganic form (HCO3

- and CO32-) and in its

organic form (total organic carbon; TOC) are addressed in this thesis.

Linkages between hydrological flow paths and river chemistry in high-latitude watersheds

The hydrological regime is an essential factor for the overall land-sea fluxes of dissolved constituents and for the seasonal variations in the chemical composition of river water (Zakharova et al. 2007). Water circulating through soils enhances weathering rates. Specific discharge is a significant component in weathering yields (kg Si or C km-2 yr-1). Jennerjahn et al. (2006) and Gaillardet et al. (1999) found that runoff coupled with physical weathering is a governing factor for chemical silicate weathering on a global scale. The hydrological regime of high-latitude watersheds is characterized by a sharp peak flow during spring as exemplified here by the Råneälven in northern Sweden (Figure 2, Paper II). Generally, ca. 50% or more of the annual water discharge occurs during April-July after snow and ice melt (Shiklomanov et al. 2000, Lammers et al. 2001).

Water routing

0

2

4

6

8

10

12

Apr-94 Apr-95 Apr-96 Apr-97 Apr-98 Apr-99

Cm

wate

r

RSW

SW

OGW

Figure 2. Water routing through the River Råneälven watershed. Old Groundwater (OGW) has a steady contribution with a longer residence time compared to soil water (SW) that can be regarded as “old water” but with a shorter residence time. The shortest residence time has water from rain and snow (RSW), which also reacts first to precipitation.

In Sweden overland flow is negligible (Rohde 1981), meaning that essentially all water will be routed in various flow paths through the soil. Some of the water will drain as near-surface flow occurring when inflow exceeds the percolation

14

rate and the soils are saturated or have a deeper frozen layer. This water draining as surface or near-surface flow (conceptually addressed in Figure 2 as water from rain and snow) has a short residence time and has more of a flushing character, whereas water percolating down through the soil layers (conceptually addressed in Figure 2 as soil water) and eventually becoming groundwater (conceptually addressed in Figure 2 as old groundwater) will have a much longer residence time and thus, more time to interact with the soil. This means that the water flowing through a watershed is not a single homogenous flow on a two-dimensional scale, but rather a three-dimensional flow with a number of flow paths. Different soil types will generate different flow paths contributing to groundwater flow in various degrees and, thus, the water in the separated paths will have varying residence times (Soulsby et al. 2006) and different chemical characteristics. Water draining mainly through organic rich topsoil layers in a situation with a shallow active layer (flow path A in Figure 3) will have higher concentrations of DOC while water draining deeper through mineral soil layers in a situation with a deepened active layer (flow path B in Figure 3) will have higher concentrations of DIC and weathered elements.

Figure 3. Water following flow path A will have a shorter residence time and deliver a different nutrient load than water following flow path B. Changes in hydrology may thus alter the total nutrient loads. As the flows in the different layers are separated temporally the flow may even be considered four-dimensional. For example, peaks in DOC concentrations emanating from near-surface layers are found to be separated in time from concentration peaks of solutes coming from deeper soil layers with different

15

flow-paths (Inamdar et al. 2004). Consequently, any fluctuations and changes in hydrology and flow paths through either global warming or river regulation will affect erosion, weathering and transport of DSi, DIC and TOC in high-latitude watersheds.

In permafrost areas long winters interrupted only by short summer periods allow only a shallow active layer to develop when the topsoils thaw, while the permafrost layer remains frozen. Thus, loading of river water with soil C and weathering products is constrained to this annually thawing and freezing layer. A vertical expansion of the active layers has been noted in Siberia (Zhang et al. 2005), and with rising temperatures this active layer could expand and deepen further, allowing previously bound soil carbon to be flushed out acting as a positive feedback to climate warming. Changes in the hydrological regime such as earlier snowmelt, lower daily maximum discharge during snowmelt and significant increases in streamflow during the cold season have been observed for the Lena River and are believed to be caused by changed permafrost conditions (Yang et al. 2002). However, comparisons between permafrost areas and areas with seasonal frost show that the permafrost areas are a greater source of DOC (Carey 2003) than non-permafrost areas in taiga and tundra regions. In permafrost areas, meltwater is forced through organic rich active layers as near-surface flow above the impervious permafrost layer, as opposed to areas with seasonal frost where the water is able to infiltrate down to deeper lying mineral soil layers where sorption can take place (Figure 4).

Figure 4. Conceptual model for flow paths with and without a permafrost layer, affecting DOC export and weathering rate. O/A indicates organic rich A-horizon, P indicates permafrost layer, arrows describe flow paths. Redrawn after MacLean 1999.

16

MacLean et al. (1999) suggests that a thawing of the underlying permafrost ice sheet would result in decreased hydrologic losses of DOC and dissolved organic nitrogen (DON) and increased losses of weathering products such as Ca and Mg. For the Yukon River a decrease in the quantity of DOC exported relative to water discharge during summer through autumn has been noted and it is believed to be a result of increased flow to deeper soil layers, residence time, and microbial materialization of DOC in the soil active layer (Striegl et al. 2005). Moreover, a general upward trend in groundwater contribution has been recorded for the Yukon River as a result of changed hydrological flow paths due to permafrost thawing, and this alteration of flow paths is believed to further cause decreased DOC and DON exports and increased dissolved inorganic carbon (DIC) and dissolved inorganic nitrogen (DIN) exports (Walvoord & Striegl 2007). Accordingly, Hinzman et al. (2005) showed increasing alkalinity concentrations in Toolik Lake, Alaska, suggesting deeper infiltration of runoff water and increased weathering. Both studies indicate that climate warming resulting in reduced permafrost areas would not necessary act entirely as a positive feedback to global warming, since increased mobilization of soil organic carbon could be counterbalanced to some extent by increased weathering and DIC formation. It has also been suggested that carbon from thawing permafrost soils would be exported primarily as particulate organic carbon (POC) and that most of the observed increase in DOC is a result of an increased terrestrial primary production (Guo et al. 2007). Thus, the potential hydrological changes in permafrost areas, allowing more water to percolate down becoming groundwater and forming more DIC, in combination with a potential increase in terrestrial primary production, could significantly dampen the foreseen massive release of old soil carbon and its enhancing effect on atmospheric CO2 concentrations.

Summary of the papers

Linkages between landscape elements and weathering regimes in high-latitude watersheds (Paper I and Paper II)

Globally, land–sea fluxes of DSi vary significantly between different regions and rivers with a general decreasing trend towards the poles. Rivers in the Arctic show concentrations of approximately 110 µM DSi and approximately 710 µM are found in the tropics (Meybeck 1979). Classically, DSi river concentrations and loads have been described as a function of runoff, temperature and bedrock type (Meybeck 1979) or runoff, temperature and physical denudation (Gaillardet et al. 1999). More recently, accumulating evidence shows that terrestrial

17

vegetation plays a significant role in the global Si cycle (Conley 2002) and control the riverine DSi export fluxes (Derry et al. 2005). TOC in arctic and boreal rivers, consisting to more than 80 % of dissolved organic carbon (DOC) (Wetzel 2001), may be used as a proxy for vegetation and weathering enhancing products such as organic acids emanating from vegetation-covered soils in high-latitude rivers. Using different landscape elements, e.g. land cover types, soil types and bedrock types, from 19 boreal and subarctic watersheds in northern Sweden we examined the significance of these landscape variables on carbon and dissolved constituent fluxes (Paper I). We tested the main hypothesis whether the relative amount of vegetation-covered areas in a watershed would determine the riverine export fluxes of carbon and dissolved constituents for those watersheds. The study contained watersheds ranging from small (34 km2) to large (40,000 km2) with characteristic vegetation types for boreal and subarctic watersheds. Fluxes of total organic carbon (TOC) and dissolved silica (DSi) but also, dissolved inorganic nitrogen (DIN) and phosphorus (DIP), were described for these 19 watersheds. The fluxes of TOC, DIP, and DSi increased by an order of magnitude with an increasing proportion of forest and wetland areas, whereas DIN did not follow this pattern, but remained constantly low. A combination of principal component analysis using the landscape variables and a further cluster analysis showed that the watersheds could be separated into two groups. The first group could be characterized as mountainous headwaters with a low percentage of forest and wetlands and the second group included those watersheds dominated by forests and wetlands. These two clusters were also valid in relation to river chemistry (TOC, DIP, and DSi) and this was confirmed with a redundancy analysis, including river chemistry and principal components as environmental variables. The mountainous watersheds showed significantly lower TOC, DIP and DSi concentrations compared to the watersheds dominated by forests and wetlands. Thus, this study showed that vegetation cover, carbon and weathering fluxes in taiga and tundra rivers are intimately linked, and that higher concentrations of weathering products are found in taiga and tundra rivers with larger areas of coniferous forest and peat cover in their watersheds. These landscape elements can thus be regarded as “hot spots” of river loading with dissolved constituents. Moreover, this study demonstrates nicely that these high-latitude watersheds can also be regarded as a potential sink for atmospheric carbon, since increasing temperature is related to more vegetation cover and, thus, higher weathering fluxes. A positive relationship between DOC and DSi has been also found for the MacKenzie watershed (Millot et al. 2003).

18

With a simple watershed model, CSIM (Mörth et al. 2007), we tested the hypothesis whether the export of total organic carbon (TOC) from topsoils in tundra and taiga watersheds as a potential positive feedback to global warming might be compensated by the export of alkalinity from groundwater (Paper II). First, we produced a representative description of the hydrology on an annual basis by means of a simple hydrological mixing model. Water was routed in three different flow paths, each running through different soil layers and each water volume getting a representative loading of TOC and DSi. During snowmelt, TOC, previously stored as soil carbon, is exported when topsoils are flushed (Figure 5), and can potentially be released to the atmosphere via respiration during riverine transport to the sea. The TOC yields of the watersheds investigated increased with vegetation and peat cover and ranged between 0.5 and 2.8 tons km-2 yr-1. During frozen periods, stream flow is dominated by groundwater. This water has percolated through the soils and is rich in DSi and alkalinity (Figure 5), i.e. atmospheric carbon that has been “consumed” in chemical weathering processes.

Figure 5. Seasonal concentration patterns of TOC, DSi and alkalinity in the Råneälven. TOC have the highest concentrations during spring flow when upper soil layers are flushed. The weathering products alkalinity and DSi peak during winter when the contribution from groundwater is at maximum. The bicarbonate export of the wetland and coniferous forest watersheds investigated was between 0.4 and 1.2 tons C km-2 yr-1 corresponding to 15-73% of the TOC export (Paper II). In the investigated rivers, alkalinity can be defined as Alk = [HCO3

-] – [H+], since CO3

2- ions can be neglected in waters below pH 8. In the investigated

1-A

pr-

96

1-A

ug

-96

1-D

ec

-96

1-A

pr-

97

1-A

ug

-97

1-D

ec

-97

1-A

pr-

98

1-A

ug

-98

1-D

ec

-98

1-A

pr-

99

1-A

ug

-99

1-D

ec

-99

1-A

pr-

00

1-A

ug

-00

1-D

ec

-00

1-A

pr-

01

1-A

ug

-01

1-D

ec

-01

0

2

4

6

8

10

12

14

TO

C,

HC

O3

-C,

DS

i [m

g L

-1]

TOC

DSi

HCO3-

19

waters, pH was 6.0 to 7.5 and therefore also H+ ions can also be neglected. Thus, by far most of the alkalinity in the watersheds of northern Sweden is HCO3

-. The watersheds investigated exported between 0.4 and 1.2 tons C km-2 yr-1 (exported as HCO3

-) corresponding to 15-73% of the TOC export. Our study suggests that global warming, which might affect water flow through the soils in taiga and tundra ecoregions, will also have an effect on carbon budgets and that alkalinity export may compensate for significant amounts of the exported TOC, thereby reducing the positive feedback to atmospheric CO2. The simple mechanistic understanding of water flow and related dissolved constituent transport from various soil layers that we have developed for these ecoregions in northern Sweden should have significant implications for the further understanding of taiga and tundra biomes acting as carbon sinks or sources. This is especially true since HCO3

- in the large Siberian rivers is significantly higher than in the low-buffered Swedish rivers, whereas TOC concentrations are similar (Gordeev 2000). The strong relation between vegetation cover, high organic carbon and weathering fluxes (Paper I) is somewhat counter-intuitive, since many studies have shown a negative correlation between TOC and DSi on a seasonal scale (Guo et al. 2004). Although TOC and DSi have a strong annual relationship, the seasonal concentrations in stream waters show the opposite pattern (Paper II; Figure 5). TOC, with its source in the upper soil layers, reaches its concentration peak during spring when topsoils are flushed at snow melt, while the weathering products DSi and alkalinity peak during winter when the groundwater relative contribution to stream flow is at its maximum. Studies of dissolved constituent concentrations in Arctic Russian rivers show that groundwater discharge is relatively constant over the year and the dilution with meltwater during peak flow lowers the concentrations of weathered elements (Gordeev 2000). Temperature not only affects chemical dissolution in a direct way, but also indirectly through vegetation cover and microbial activity in the soils. The weathering rate for plagioclase increases by ca. 8% for each ºC (by using the Arrhenius equation and an activation energy of 15 kcal mol-1; see (Lasaga 1998)). Thus, an increase of 5 ºC in high latitudes as foreseen by many GCMs would theoretically enhance weathering fluxes by a factor of 1.6. On the other hand (Drever 1994), suggested that weathering rates of silicate rocks increase by a factor of 2, but locally by a factor of 10, as an effect of terrestrial plants. Other experiments on plant induced weathering of a basaltic rock suggested a 5-fold increase with vegetation (Hinsinger et al. 2001). Therefore, in taiga and tundra biomes where vegetation growth is limited by short summers, cold winters and frozen soil conditions, temperature may have a more important effect on DSi, alkalinity and TOC fluxes through its effect on vegetation patterns and growing conditions. A longer growing season leading to increasing biomass in the

20

vegetated areas and also vegetation expanding into areas previously unvegetated will increase the weathering fluxes probably much more than the physico-chemical effect accelerating chemical reactions.

Linkages between landscape elements, hydrological alterations and weathering regimes in high-latitude watersheds (Paper III, Paper IV, Paper V)

Today, more than 30% of the global river flow is dammed or diverted (Milliman 1997, Vörösmarty et al. 1997, Vörösmarty & Sahagian 2000). Around 40 000 large dams (defined as more than 15 m in height) and more than 800 000 smaller ones are in operation, and more are still being constructed at an accelerating rate (Nilsson & Berggren 2000, Vörösmarty & Sahagian 2000). A series of studies in major global rivers have demonstrated that river regulation and damming leads to decreased dissolved constituent transport to coastal seas, although the mechanisms are multi-fold (Friedl & Wuest 2002, Humborg et al. 2002). Initially, the primary mechanism responsible for the decrease in DSi loads was thought to be solely due to the trapping and subsequent sediment burial of biogenic silica (BSi) in the form of diatoms that were autochthonously produced in the reservoirs due to decreasing water currents and improved light conditions and termed the artificial lake effect (Van Bennekom & Salomons 1979). However, it is not fully understood how damming affects the Si cycle, since damming effects have been observed in eutrophic (Humborg et al. 1997) as well as ultra-oligotrophic Swedish rivers (Paper III). Thus, there must be factors other than flourishing diatoms in the reservoirs causing the low DSi concentrations in these rivers. Building of dams and creating large water reservoirs for hydroelectric power plants can also be regarded as large-scale experiments on vegetation effects on dissolved constituent fluxes, since vegetation cover along the riparian zone of the riverbed is drastically changed with river regulation (Nilsson & Berggren 2000). With the newly formed reservoirs inundating vegetation and thus altering the ratio of vegetated soils versus lakes and streams in a watershed, a decrease in large scale production of organic acids through root exudations, microbial degradation of plant litter or secretions by mycorrhyzal symbionts can be expected, processes regarded as weathering enhancing (Berner 1992, Berner & Berner 1996). The hypothesis that damming leads to a depletion of major elements in river systems also in oligotrophic rivers as a consequence of geomorphological changes of the watersheds (Paper III) was tested by comparing the unregulated and relatively unperturbed river Kalixälven with one of the heaviest regulated rivers in Eurasia, the Luleälven (Dynesius & Nilsson 1994). In the unperturbed

21

river, concentrations of dissolved constituents increased gradually all through the course from the headwaters down to the river mouth while in the regulated river only small increases were recorded. Since the two rivers were similar with regard to climate, vegetation types and lithology, the differences in river chemistry may be related to changes caused by river regulation affecting landscape variables controlling the dissolved constituent levels in the river. One of the most visible effects of river regulation is the inundation and loss of vegetation-clad shorelines, and the creation of large bodies of water with dried out riverbeds in-between, this would definitely have an impact on dissolved constituent fluxes. The forest area to lentic area ratio in the headwater was reduced dramatically with damming, from 2.65 to 0.84. In reaches between the reservoirs, underground channeling of water between hydroelectric power plants and reservoirs, and reductions of normal water level fluctuations have resulted in further decrease in soil/water contact. Moreover, due to the oligotrophic status of the investigated rivers it seems unlikely that the decrease would result only from diatom blooms in the reservoirs, but as a result from a reduced weathering flux from the surrounding soils. In a following study we tested the hypothesis that the lower fluxes of biogenic elements could be explained by biological uptake. The RIVERSTRAHLER model (Billen et al. 1994) offers a general framework for the study of the biogeochemical functioning of river systems and was used in Paper IV to compare DSi behavior in river networks that differ by their hydrological regime and climate, as well as by the human activities in their watershed. We used the RIVERSTRAHLER model results from the coupling of a unique model of biogeochemical processes (RIVE) and a hydrological model (HYDROSTRAHLER), describing in an idealized way, the water fluxes in the drainage network represented by a regular scheme of confluence of tributaries of increasing stream-order with mean characteristics (see Strahler’s concept of stream order (Strahler 1957)). River regulation for hydroelectric power, such as those found in northern Sweden, are causing a reversed annual discharge pattern with low discharge during spring and summer and high discharge during winter when power consumption peaks (Nilsson et al. 1997, Jansson et al. 2000). This reversed hydrology disrupts the normal interactions of groundwater and stream water in the hyporheic zone (White 1993). This zone located beneath and lateral to a streambed is the interface between the water types and the flow dynamics in this zone is regarded to be important for surface water/groundwater interactions (Hancock 2002), such as dissolved constituent fluxes from vegetation-covered active soil-layers and surface water. Due to a prolonged water residence time in reservoirs, the most obvious effect is a smoothing of the seasonal variations in dissolved constituent concentrations at the outlet; however, this in itself does not

22

involve any dissolved constituent retention. Budget calculations based on the RIVERSTRAHLER model results indicate a significant reduction of phosphorus (over 30%) and Si (close to 25%) fluxes in the Luleälven watershed as a result of regulation. A part of this reduction (10%) can be attributed to the flooding of the fluvial corridor of the 5th stream order (and a part of the 4th stream order) changing the land cover from the formerly forested valleys to bare rocky shores and large reservoirs. In the model, this land cover change is accounted for by simply suppressing a dissolved constituent contribution from about 2000 km² wetlands and forest area, the basin area flooded by the reservoirs. However, the retention of DSi is also significant and is associated to the low diatom growth in the reservoir during the summer period (and not only in the spring as in the unregulated Kalixälven). However, the depth, and, principally, the very low phosphorus loading limit algal development in this reservoir. The results described in papers III and IV could not satisfactorily answer the question to what extent diatom blooms in the reservoirs or changes in weathering rates induced by hydrological alterations contribute to the observed differences in DSi fluxes. Both processes are probably significant, however, the general question appeared, whether a change in a few percent of a watershed area as a result of reservoir formation and underground channeling may lead to such significant changes in the concentrations and fluxes of dissolved constituents. Moreover, as shown in papers I, II and III there are hot spots in watersheds with respect to river loading of dissolved constituents. These hot spots are probably the vegetated areas, especially the forests and wetlands (Struyf & Conley 2008) along the riparian zone. Moreover, organic rich soils like peat appear to supply much more carbon and weathering products (DIC, DSi) compared to less organic rich soil types found especially above the tree-line. Therefore, the last paper (Paper V) focused on these hot spot areas and their respective loss due to river regulations. Areas can be lost either due to inundation when water levels are raised in the newly formed reservoirs or from being bypassed when water is rerouted through headrace tunnels. In the case of inundation, the areas are truly lost when they get submerged and the normally vegetated riparian zone is gone. For the bypassed areas they are lost in the sense that they are disrupted from the runoff coming from upstream watersheds diverted by headrace tunnels. Thus, their importance for dissolved constituent loading is probably greatly diminished by decreased runoff and annual floodings. Generally, good correlations were found when TOC and DSi stream concentrations were plotted versus the calculated average soil depth of a watershed or versus the amount of forest and wetland areas. Thus, soil depth explains the variability in dissolved constituents between watersheds to the same extent as vegetation cover, since both variables co-vary. However, the average

23

soil cover affects the flow paths of waters and thereby the water residence time. McGuire et al. (2005) related water residence time using topographic index to river chemistry and showed that the increased concentrations of weathering products can be explained by an increase in water residence time. The spatial distribution of the soil thickness in a landscape varies, but follows a pattern created by basic physical laws of gravity and friction resulting in thicker soils layers on the valley floors. The locations of these hot spots is clearly seen on a draped digital elevation model (DEM) showing the limited amount of thicker soil layers and forest and wetland in mountainous headwaters, whereas areas located along the fluvial corridors on the valley floors are covered with deciduous forest and wetlands (Figure 6). Infiltration of these soil layers forming the riparian zone increases water residence time along the valley floors. In other words, an increase in water residence time implies a longer contact of water with mineral surfaces that ultimately determines the weathering rate.

Figure 6. An unregulated headwater watershed with the vegetation classes related to high TOC and DSi concentrations; wetland (purple) and forest (deciduous=light green, mixed=medium green, coniferous=dark green) located close to the fluvial corridors. These are locations that will get inundated in the event of a regulation.

Within the headwater areas of the river Luleälven, only 3% of the surface has been inundated by the creation of reservoirs. However, some 10% of the soil volumes and aggregated forest and wetland areas have been lost due to damming and further hydrological alteration such as bypassing entire sub-watersheds by headrace tunnels. Moreover, looking at individual forest classes, our estimates indicate that ca. 37% of the deciduous forests have been inundated by the four

24

major reservoirs built in the Luleälven headwaters. This land cover class “deciduous forest” is almost synonymous with the riparian zone and occurrences of organic rich soil layers in these headwaters that are hot spots for DSi and TOC loadings (Struyf & Conley 2008), especially in high latitudes. Such a significant loss of hot spots for river loading may indeed explain the observed lower DSi fluxes in the regulated watersheds of northern Sweden, though more detailed studies are needed, quantifying more precisely the relative contribution of the riparian zones and wetlands to the overall dissolved constituent yields of high-latitude watersheds.

Conclusions and Future Perspectives This thesis shows that large-scale changes in watersheds such as river regulation and global climate change will alter biogeochemical fluxes and the transport of biogenic elements from land to sea. Vegetation cover and soil thickness are clearly significant landscape variables for the dissolved constituents load in taiga and tundra watersheds. However, it is not sufficient to classify a landscape by bedrock type or soil type or vegetation type, the location and slope angle of a given land class is also vital. The riparian zone for example can be regarded as a hot spot for river loading with C and dissolved solids. Along the same line, water residence time and pathways through soil layers are key factors for carbon and weathered element fluxes. Future studies should focus further on the role of the biosphere for weathering, and export patterns of dissolved solids. This is especially important given the fact that a global warming with altered precipitation patterns and higher temperatures will move the boundaries for various vegetation types and alter species composition. Possible effects are taiga forests spreading north, expanding into tundra areas, or wetlands expanding or drying out due to changed precipitation patterns. To what extent such changes will affect the fluxes of dissolved constituents needs to be studied. However, with a decline in permafrost bound areas and altered precipitation patterns, relative distribution of water flowing through the different flow paths in a watershed is likely to change and, thus, the composition of the dissolved constituent load. The water pathways through soils are essential in governing dissolved constituent loads. A decrease in DOC and an increase in DIC loads to the Arctic Ocean might be a possible scenario as demonstrated in this thesis and first indications of such changes have been observed recently in the Yukon watershed.

25

Acknowledgements Throughout the work on this thesis I have met many generous people who have helped me with practical arrangements, inspiring discussions and support in general, some for a short while and others throughout the process. I wish to express my sincere gratitude to all of you. All the good parts in this thesis are thanks to my collaboration with you, and all mistakes are my own. I would especially like to thank my supervisor Christoph Humborg who has been around for the whole ride. From the first sampling trips up north, with me navigating a small boat on big uncharted waters, to last minute feedback on manuscript preparation. Without his support this would have been less fun. And also, Carl-Magnus Mörth, Lars Rahm and Sven Blomqvist who has been co-supervisors for various parts of the trip. They all deserve sincere thanks for their contributions at various stages of my work. I also want to thank all co-authors and in particular Dennis Swaney who, beside the academic input, gave me a comprehensive tour of Ithaca and Syracuse including an introduction to doggy bags. Finally, I want to thank those who at one time or another have shared a room with me without complaining, and also faculty staff, fellow students and administrative personnel at both ITM and Systems Ecology and also the people at BNI. You have all contributed in various ways. Thank You !!!

26

References

Bengtsson L (1997) A numerical simulation of anthropogenic climate change. Ambio 26:58-65

Bennett PC, Rogers JR, Choi WJ (2001) Silicates, silicate weathering, and microbial ecology. Geomicrobiology Journal 18:3-19

Berner EK, Berner RA (1996) Global Environment: Water, Air and Geochemical Cycles, Vol. Prentice Hall, New York

Berner RA (1992) Weathering, plants, and the long-term carbon-cycle. Geochimica Et Cosmochimica Acta 56:3225-3231

Billen G, Garnier J, Hanset P (1994) Modelling phytoplankton development in whole drainage networks : the RIVERSTRAHLER model applied to the Seine river system. Hydrobiologia 289:119-137

Billen G, Lancelot C, Meybeck M (1991) N, P, and Si retention along the aquatic continuum from land to the ocean. In: Mantoura RFC, Martin J-M, Wollast R (eds) Ocean Margin Processes in Global Change. Wiley Sons, p 19-14

Cai M (2005) Dynamical amplification of polar warming. Geophysical Research Letters 32

Carey SK (2003) Dissolved organic carbon fluxes in a discontinuous permafrost subarctic alpine catchment. Permafrost and Periglacial Processes 14:161-171

Cole JJ, Prairie YT, Caraco NF, McDowell WH, Tranvik LJ, Striegl RG, Duarte CM, Kortelainen P, Downing JA, Middelburg JJ, Melack J (2007) Plumbing the global carbon cycle: Integrating inland waters into the terrestrial carbon budget. Ecosystems 10:171-184

Conley DJ (2002) Terrestrial ecosystems and the global biogeochemical silica cycle. Global Biogeochemical Cycles 16

Conley DJ, Schelske CL, Stoermer EF (1993) Modification of the Biogeochemical Cycle of Silica with Eutrophication. Marine Ecology-Progress Series 101:179-192

Derry LA, Kurtz AC, Ziegler K, Chadwick OA (2005) Biological control of terrestrial silica cycling and export fluxes to watersheds. Nature 433:728-731

Drever JI (1994) The Effect of Land Plants on Weathering Rates of Silicate Minerals. Geochimica Et Cosmochimica Acta 58:2325-2332

Drever JI (1997) The Geochemistry of Natural Waters, Vol. Prentice Hall Dugdale RC, Wilkerson FP (1998) Silicate regulation of new production in the

equatorial Pacific upwelling. Nature 391:270-273 Dutta K, Schuur EAG, Neff JC, Zimov SA (2006) Potential carbon release from

permafrost soils of Northeastern Siberia. Global Change Biology 12:2336-2351

27

Dynesius M, Nilsson C (1994) Fragmentation and flow regulation of river systems in the northern 3rd of the world. Science 266:753-762

Freeman C, Evans CD, Monteith DT, Reynolds B, Fenner N (2001) Export of organic carbon from peat soils. Nature 412:785-785

Friedl G, Wuest A (2002) Disrupting biogeochemical cycles - Consequences of damming. Aquatic Sciences 64:55-65

Gaillardet J, Dupre B, Louvat P, Allegre CJ (1999) Global silicate weathering and CO2 consumption rates deduced from the chemistry of large rivers. Chemical Geology 159:3-30

Gordeev VV (2000) River input of water, sediment, major ions, nutrients and trace metals from russian territory to the Arctic Ocean. In: Lewis EL, et al. (ed) The Freshwater Budget of the Artic Ocean. Kluwer Academic, Dordrecht, London, p 297-322

Gorham E (1991) Northern Peatlands - Role in the Carbon-Cycle and Probable Responses to Climatic Warming. Ecological Applications 1:182-195

Grayston SJ, Vaughan D, Jones D (1997) Rhizosphere carbon flow in trees, in comparison with annual plants: The importance of root exudation and its impact on microbial activity and nutrient availability. Applied Soil Ecology 5:29-56

Guo L, Ping CL, Macdonald RW (2007) Mobilization pathways of organic carbon from permafrost to arctic rivers in a changing climate. Geophysical Research Letters 34

Guo LD, Zhang JZ, Gueguen C (2004) Speciation and fluxes of nutrients (N, P, Si) from the upper Yukon River. Global Biogeochemical Cycles 18

Hancock PJ (2002) Human impacts on the stream-groundwater exchange zone. Environmental Management 29:763-781

Harrison KG (2000) Role of increased marine silica input on paleo-pCO(2) levels. Paleoceanography 15:292-298

Hinsinger P, Barros ONF, Benedetti MF, Noack Y, Callot G (2001) Plant-induced weathering of a basaltic rock: Experimental evidence. Geochimica Et Cosmochimica Acta 65:137-152

Hinzman LD, Bettez ND, Bolton WR, Chapin FS, Dyurgerov MB, Fastie CL, Griffith B, Hollister RD, Hope A, Huntington HP, Jensen AM, Jia GJ, Jorgenson T, Kane DL, Klein DR, Kofinas G, Lynch AH, Lloyd AH, McGuire AD, Nelson FE, Oechel WC, Osterkamp TE, Racine CH, Romanovsky VE, Stone RS, Stow DA, Sturm M, Tweedie CE, Vourlitis GL, Walker MD, Walker DA, Webber PJ, Welker JM, Winker K, Yoshikawa K (2005) Evidence and implications of recent climate change in northern Alaska and other arctic regions. Climatic Change 72:251-298

Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K, Johnson CA (eds) (2001) IPCC, Climate change 2001: The scientific basis. Contribution of working group I to the Third assessment Report of

28

the Intergovermental Panel on Climate Change., Vol. Cambridge University Press United Kingdom and New York, Cambridge

Hruska J, Köhler S, Laudon H, Bishop K (2003) Is a universal model of organic acidity possible: Comparison of the acid/base properties of dissolved organic carbon in the boreal and temperate zones. Environmental Science & Technology 37:1726-1730

Humborg C, Blomqvist S, Avsan E, Bergensund Y, Smedberg E, Brink J, Mörth C-M (2002) Hydrological alterations with river damming in northern Sweden: Implications for weathering and river biogeochemistry. Global Biogeochemical Cycles 16:art. no.-1039

Humborg C, Ittekkot V, Cociasu A, von Bodungen B (1997) Effect of Danube River dam on Black Sea biogeochemistry and ecosystem structure. Nature 386:385-388

Inamdar SP, Christopher SF, Mitchell MJ (2004) Export mechanisms for dissolved organic carbon and nitrate during summer storm events in a glaciated forested catchment in New York, USA. Hydrological Processes 18:2651-2661

Ittekkot V, Humborg C, Schäfer P (2000) Hydrological alterations and marine biogeochemistry: A silicate issue? Bioscience 50:776-782

Jansson R, Nilsson C, Dynesius M, Andersson E (2000) Effects of river regulation on river-margin vegetation: A comparison of eight boreal rivers. Ecological Applications 10:203-224

Jennerjahn TC, Knoppers BA, Souza WFL, Brunskill GJ, Silva EIL, Adi S (2006) Factors Controlling Dissolved Silica in Tropical Rivers. In: Ittekkot V, Unger D, Humborg C, Tac An N (eds) The Silicon Cycle-Human Perturbations and Impacts on Aquatic Systems, Vol SCOPE 66. Island Press, Washington D.C., p 29-51

Kaplan JO, New M (2006) Arctic climate change with a 2 degrees C global warming: Timing, climate patterns and vegetation change. Climatic Change 79:213-241

Lammers RB, Shiklomanov AI, Vörösmarty CJ, Fekete BM, Peterson BJ (2001) Assessment of contemporary Arctic river runoff based on observational discharge records. Journal of Geophysical Research-Atmospheres 106:3321-3334

Landeweert R, Hoffland E, Finlay RD, Kuyper TW, van Breemen N (2001) Linking plants to rocks: ectomycorrhizal fungi mobilize nutrients from minerals. Trends in Ecology & Evolution 16:248-254

Lasaga A (1998) Kinetic Theory in Earth Sciences, Vol. Princeton University Press

Laudon H, Köhler S, Buffam I (2004) Seasonal TOC export from seven boreal catchments in northern Sweden. Aquatic Sciences 66:223-230

29

MacLean R, Oswood MW, Irons JG, McDowell WH (1999) The effect of permafrost on stream biogeochemistry: A case study of two streams in the Alaskan (USA) taiga. Biogeochemistry 47:239-267

McGuire KJ, McDonnell JJ, Weiler M, Kendall C, McGlynn BL, Welker JM, Seibert J (2005) The role of topography on catchment-scale water residence time. Water Resources Research 41

Meybeck M (1979) Pathways of major elements from land to ocean through rivers. In: Martin J-M, Burton JD, Eisma D (eds) River Inputs to Ocean Systems. UNEP IOC SCOR United Nations, p 18-30

Milliman JD (1997) Oceanography - Blessed dams or damned dams? Nature 386:325-326

Millot R, Gaillardet J, Dupre B, Allegre CJ (2003) Northern latitude chemical weathering rates: Clues from the Mackenzie River Basin, Canada. Geochimica Et Cosmochimica Acta 67:1305-1329

Monteith DT, Stoddard JL, Evans CD, de Wit HA, Forsius M, Høgåsen T, Wilander A, Skjelkvåle BL, Jeffries DS, Vuorenmaa J, Keller B, Kopacek J, Vesely J (2007) Dissolved organic carbon trends resulting from changes in atmospheric deposition chemistry. Nature 450:537-U539

Mörth C-M, Humborg C, Eriksson E, Danielsson A, Medina R, Löfgren S, Swaney DP, Rahm L (2007) Modeling riverine nutrient transport of the Baltic Sea-A large scale approach. Ambio 36:124-133

Nilsson C, Berggren K (2000) Alterations of riparian ecosystems caused by river regulation. Bioscience 50:783-792

Nilsson C, Jansson R, Zinko U (1997) Long-term responses of river-margin vegetation to water-level regulation. Science 276:798-800

Nilsson C, Reidy CA, Dynesius M, Revenga C (2005) Fragmentation and flow regulation of the world's large river systems. Science 308:405-408

Officer CB, Ryther JH (1980) The possible importance of silicon in marine eutrophication. Mar Ecol 3:83-91

Peterson BJ, Holmes RM, McClelland JW, Vörösmarty CJ, Lammers RB, Shiklomanov AI, Shiklomanov IA, Rahmstorf S (2002) Increasing river discharge to the Arctic Ocean. Science 298:2171-2173

Rohde A (1981) Spring flood - meltwater or groundwater? Nordic Hydrology 12:21-30

Shiklomanov IA, Shiklomanov RB, Lammers BJ, Peterson BJ, Vörösmarty CJ (2000) The dynamics of river water inflow to the Arctic Ocean. In: Lewis EL (ed) The freshwater Budget of the Arctic Ocean. Kluwer Academic Publishers, p 281-296

Snow M (2005a) Taiga climate. In: Oliver JE (ed) Encyclopedia of World Climate. Springer, Dordrecht, p 705-707

Snow M (2005b) Tundra climate. In: Oliver JE (ed) Encyclopedia of World Climatology. Springer, Dordrecht, p 756-759

30

Soulsby C, Tetzlaff D, Rodgers P, Dunn S, Waldron S (2006) Runoff processes, stream water residence times and controlling landscape characteristics in a mesoscale catchment: An initial evaluation. Journal of Hydrology 325:197-221

Strahler AH (1957) Quantitative analysis of watershed geomorphology. Geophysical Union Transactions 38:1290–1299

Striegl RG, Aiken GR, Dornblaser MM, Raymond PA, Wickland KP (2005) A decrease in discharge-normalized DOC export by the Yukon River during summer through autumn. Geophysical Research Letters 32

Struyf E, Conley DJ (2008) Silica: an essential nutrient in wetland biogeochemistry. Frontiers in Ecology and the Environment preprint:0000-0000

Treguer P, Nelson DM, van Bennekom AJ, Demaster DJ, Leynaert A, Queguiner B (1995) The silica balance in the world ocean - a reestimate. Science 268:375-379

Turner RE, Qureshi N, Rabalais NN, Dortch Q, Justic D, Shaw RF, Cope J (1998) Fluctuating silicate : nitrate ratios and coastal plankton food webs. Proceedings of the National Academy of Sciences of the United States of America 95:13048-13051

Walvoord MA, Striegl RG (2007) Increased groundwater to stream discharge from permafrost thawing in the Yukon River basin: Potential impacts on lateral export of carbon and nitrogen. Geophysical Research Letters 34

Van Bennekom J, Salomons W (1979) Pathways of nutrients and organic matter from land to ocean through rivers. In: Martin J-M, Burton JD, Eisma D (eds) River Inputs to Ocean Systems. UNEP, IOC, SCOR, United Nations, New York, Rome, p 33-51

van Breemen N, Finlay R, Lundström U, Jongmans AG, Giesler R, Olsson M (2000) Mycorrhizal weathering: A true case of mineral plant nutrition ? Biogeochemistry 49:53-67

Wetzel RG (2001) Limnology. In. Academic Press, p 1006 p White DS (1993) Perspectives on Defining and Delineating Hyporheic Zones.

Journal of the North American Benthological Society 12:61-69 Vörösmarty CJ, Sahagian D (2000) Anthropogenic disturbance of the terrestrial

water cycle. Bioscience 50:753-765 Vörösmarty CJ, Sharma KP, Fekete BM, Copeland AH, Holden J, Marble J,

Lough JA (1997) The storage and aging of continental runoff in large reservoir systems of the world. Ambio 26:210-219

Yang DQ, Kane DL, Hinzman LD, Zhang XB, Zhang TJ, Ye HC (2002) Siberian Lena River hydrologic regime and recent change. Journal of Geophysical Research-Atmospheres 107

Zakharova EA, Pokrovsky OS, Dupre B, Gaillardet J, Efimova LE (2007) Chemical weathering of silicate rocks in Karelia region and Kola

31

peninsula, NW Russia: Assessing the effect of rock composition, wetlands and vegetation. Chemical Geology 242:255-277

Zhang TJ, Frauenfeld OW, Serreze MC, Etringer A, Oelke C, McCreight J, Barry RG, Gilichinsky D, Yang DQ, Ye HC, Ling F, Chudinova S (2005) Spatial and temporal variability in active layer thickness over the Russian Arctic drainage basin. Journal of Geophysical Research-Atmospheres 110

Zimov SA, Schuur EAG, Chapin FS (2006) Permafrost and the global carbon budget. Science 312:1612-1613