Yukon River in 2004
Abstract
Arctic river basin has been suffering a great impact from climate change. However, dynamics and mechanism of biogeochemical processes in boreal rivers in cold season are still partly out of knowledge. Our study focused on seasonality of the Yukon River, one of the regularly ice-covered Arctic rivers in North America. Water samples were collected monthly from downstream Yukon River (known as Pilot Station) from July 2004 to September 2005. Absorption and fluorescence of dissolved organic matter in two phases: colloids and low molecular weight were measured to examine optical characteristics and size distribution of organic matter. Biological index (BIX) and humification index (HIX) are EEM-derived indicators for bio-activities and humification level in aquatic environments. The trend of BIX implied stronger biological activities in the stable fluvial ecosystem under ice. HIX followed the discharge; large river volume with predominant terrestrial input drained in summer, while low-DOC groundwater mainly contributed in winter. Three fluorescence components were decomposed from Excitation-Emission Matrix (EEMs), including two humic-like components and one protein-like component. The ratio between protein-like component and humic-like components indicates the water chemistry conditions in the Yukon River. Relatively high ratio value in the ice-covered season suggested thriving algae community activity under the ice.
Introduction
Arctic river basin has been suffering a great impact from the climate change. Our study focused on the Yukon River, one of the regularly ice-covered Arctic river basins. Tributaries of Yukon River are commonly categorized into three classes (Figure 1) according to their major contributors: ‘blackwater’, ‘whitewater’, and ‘meltwater’(Striegl et al., 2007). Blackwater tributaries, such as the Porcupine River, exhibit highly-colored river water with high DOC concentration and relatively lower DIC loading because they are draining through permafrost wetlands and peatland. River water in whitewater class are mainly derived from groundwater. High DOC but low DIC loadings are their characteristic features in river water. Finally, the tributaries in meltwater class, such as Tanana River, are dominated by meltwater from glaciers, perennial ice and snowfields. DOC concentration in these river water is relatively low, but the inorganic carbon loadings (including PIC and DIC) are high.
Figure 1. Watershed of Yukon River. The Yukon River basin covers a large exiguously-populated area of permafrost (generally underlain by continuous permafrost: 16%; generally underlain by discontinuous permafrost: 40%; generally underlain by moderately thick to thin permafrost (15-150 m): 24%; generally underlain by discontinuous permafrost: 6%; generally underlain by numerous isolated masses of permafrost: 5%; and sporadic masses of permafrost: 9%) (Brabets et al., 2000). The inserted figure shows where Yukon River basin locates in the northwest corner of North America. Sampling location, denoted by a blue dot, in the downstream is known as Pilot Station. Water samples collected in the site integrate all characteristics in the whole Yukon River watershed.
Water samples were monthly collected from upstream and downstream Yukon River between July 2004 and September 2005. Strong seasonality was observed during the one-year monitoring. First flood in 2005 occurred in the middle of April. DOC concentration was consistent with discharge rate. Water from glacier and snowfield composites lower δ18O and δ2H. When temperature raised, large meltwater input negatived δ18O in the river water from average -17.5 ‰ in the winter to average 20.1 ‰.
Figure 2. Daily discharge (upper left panel), DOC concentration (bottom left panel), δ18O (upper right panel), and δ2H (bottom right panel) of sampling seasons between 2004 and 2005.
Figure 3. SUVA254 (left) and s275-295 (right) in low-molecular weight phase (upper) and colloidal organic matter (bottom).
Figure 4. biological index (BIX) and humification index (HIX) in low-molecular weight phase (upper two plots) and colloidal organic matter (bottom two plots).
Three fluorescence components were decomposed from Excitation-Emission Matrix (EEMs), including two humic-like components and one protein-like component (Figure 5). The ratio between protein-like component and humic-like components indicates the water chemistry conditions in the Yukon River (Figure 6). Relatively high ratio value in the ice-covered season suggested thriving algae community activity under the ice.
Figure 5. Three components decomposed from EEMs-PARAFAC. Component 1 (left) and Component 2 (middle) are humic-like, and Component 3 (right) are protein-like fluorophore.
Almost previous studies about Yukon River concerned about flood season in summer (Guo & Macdonald, 2006; Spencer et al., 2008; Striegl et al., 2007). Few researchers focused on slowly-flowing Yukon river water under ice partly due to extreme sampling conditions in winter (Guo et al., 2012). However, it is still important to know the mechanisms of Yukon River in winter. With respect to the water source, major contributor in winter is ‘whitewater’, which derives from relatively-eutrophic groundwater containing exceeding DIC and DIN (Guo et al., 2012). Additionally, hydrological dynamic under ice is inactive. Here we hypothesis that the stable ice-covered river water ecosystem capacitates stronger bio-activities compared to the highly-disturbed spring and summer.
In both low molecular weight phase and high molecular weight phase, biological index (BIX) gradually escalated from warm to cold, while humification index (HIX) were lower in the winter (Figure 4). It implies more intense biological activities in the ice-covered aqua-environment. The concave HIX patterns showed that the dominant terrestrial input in spring and summer, and autochthonous DOM contributed more in the winter. Furthermore, PARAFAC-decomposed components ratio supports this hypothesis as well (Figure 6). The cliff declines of ratio between C3 and sum of C1 and C2 appeared in both colloids and molecules smaller than 1 kDa. The increasing ratio values are probably related to the boost of phytoplankton community.
Figure 6. Ratio of Component 3 to summary of Component 1 and Component 2 in low-molecular weight phase (left panel) and colloidal organic matter (right panel).
Supplementary:
Figure 7. Excitation-emission matrix (EEMs) of low molecular weight phase (left column) and colloids (right column) in the winter of 2014 (upper), first flood of 2015 (middle) and winter of 2015 (bottom).
Samples:
Yukon River downstream
Sampling time:
2004-2005
Sampling location: Pilot Station, Alaska
Basic knowledge of Yukon River 2004-2005
Sample categories:
colloidal organic matter (>1kDa), low molecular weight organic matter (<1kDa), dissolved organic matter (<0.45e-6 m)
UV-Visable spectrum results
SUVA
LMW
DOM
COM
S275-295
LMW
DOM
COM
Results from EEMs
HIX
LMW
DOM
COM
BIX
LMW
DOM
COM
FIX
LMW
DOM
COM
BIX/HIX
LMW
DOM
COM
EEM parallel factor analysis results
3 components decomposed by PARAFAC program:
4 components from PARAFAC:
Comparison
If I combine samples from 2002 and 2004 together to do parallel factor analysis
I’ve got three Split-Half-Analysis-Validated components.
Components ratios
LMW
![](\image\Yukon_2004\Component ratio-LMW.jpg)
DOM
![](\image\Yukon_2004\Component ratio-DOM.jpg)
COM
![](\image\Yukon_2004\Component ratio-COM.jpg)
Normalized Components
LMW
DOM
COM
C3/(C2+C1)
LMW
DOM
COM