Study sites (a-u) at Lompolojägänoja

Pallas research site location

• Lompolojängänoja, 2nd order spring water fed stream, 5.1 km2 catchment area

• Mire catchment with interconnected open, nutrient-rich sedge fens

• Adjacent hills mainly covered by Norway spruce forests

• Sampling 3 times during 2019 (June, July, September)

• Hydrology (isotopes d18O/d2H, current speed)

• Chemistry (pH, O2, EC, temp., Abs254, DIC,

DOC)

• CO2/CH4 fluxes and concentrations• Mikrobial community (16S rRNA)

• Combined with: • Continuous monitoring of hydrology

and chemistry of the stream (isotopes, discharge, TOC)

• Continuous monitoring of terrestrial GHG fluxes (forest and fen areas)

• Comprehensive weather data

Understanding interacting dynamics of hydrology, carbon cycle and

greenhouse gas fluxes in Arctic watershedsKaisa-Riikka Mustonen1, Hannu Marttila2, Kaisa Lehosmaa1, Iina Koivunen1, Jeffrey Welker1,3,6, Annalea Lohila4,5, Jussi Jyväsjärvi1

1Ecology and Genetics Research Unit, University of Oulu, Finland; 2Water Resources and Environmental Engineering Research Unit, University of

Oulu, Finland; 3Department of Biological Sciences, University of Alaska Anchorage, USA; 4INAR Institute for Atmospheric and Earth System Research,

University of Helsinki, Finland; 5Finnish Meteorological Institute, Climate System Research, Helsinki, Finland; 6The University of the Arctic (UArctic).

• First sampling not that far from snowy days, and right after few days of heavy rainfall and high discharge. Precipitation isotopes before June sampling are also showing relatively depleted values snow/sleet events mixed GW and snow signal in water source

• Second sampling during baseflow conditions. GW% high, no more snowmelt signal, pure baseflow conditions. Too minor rain events to have an effect on water source in the stream.

• Third sampling right after a heavy rain event. Also heavy rain events one month prior to sampling.

50

60

70

80

90

100

a b c d e f g h i j k l m n o p q r s t u

Groundwater %

June July Sept

Head waters close to inlet lake

-110

-105

-100

-95

-90

-85

-15.0 -14.0 -13.0 -12.0

d2 H

d18O

d18O / d2H relationship

Stream June Stream July

Stream Sept Lake June

Lake July Lake Sept

• September more enriched isotope values + lower GW% indicate water origin from precipitation Terrestrial areas of catchment saturated with water high surface runoff flow into the stream.

• September conditions reveal the most groundwater influenced sites (a, g, p, q, r). High GW% despite high surface runoff.

• Sites i-m have lowest GW% in September. These sites are also influenced by additional runoff from upper reach ditches more surface runoff into the stream.

• Sites p, q, r are close to inlet with no springs nearby, yet heavily influenced by GW GW seepage from surrounding wetland areas. Also sites n and o (wetland sites) indicate relatively high GW influence and hence seepage through the wetland.

-20

-15

-10

-5

0

5

10

0

15

30

45

60

75

TOC

(m

g/l)

run

off

(l/

s/km

2)

daily_runoff

daily_TOC

0

10

20

30

40

-30

-10

10

30

Pre

c. a

mo

un

t (m

m)

Air

tem

per

atu

re (

C)

0

20

40

60

10.

5.2

019

12.

5.2

019

14.

5.2

019

16.

5.2

019

18.

5.2

019

20.

5.2

019

22.

5.2

019

24.

5.2

019

26.

5.2

019

28.

5.2

019

30.

5.2

019

1.6

.20

193

.6.2

019

5.6

.20

197

.6.2

019

9.6

.20

191

1.6.

201

91

3.6.

201

91

5.6.

201

91

7.6.

201

91

9.6.

201

92

1.6.

201

92

3.6.

201

92

5.6.

201

92

7.6.

201

92

9.6.

201

91

.7.2

019

3.7

.20

195

.7.2

019

7.7

.20

199

.7.2

019

11.

7.2

019

13.

7.2

019

15.

7.2

019

17.

7.2

019

19.

7.2

019

21.

7.2

019

23.

7.2

019

25.

7.2

019

27.

7.2

019

29.

7.2

019

31.

7.2

019

2.8

.20

194

.8.2

019

6.8

.20

198

.8.2

019

10.

8.2

019

12.

8.2

019

14.

8.2

019

16.

8.2

019

18.

8.2

019

20.

8.2

019

22.

8.2

019

24.

8.2

019

26.

8.2

019

28.

8.2

019

30.

8.2

019

1.9

.20

193

.9.2

019

5.9

.20

197

.9.2

019sn

ow

dep

th (

cm)

Sampling occasionsd

18O

pre

cip

itat

ion

• Stream isotopes in June and July do not differ much, however larger variance and a bit more depleted values in June also support the still remaining snowmelt influence. More enriched isotope values in September indicate water origin from precipitation.

• Consistent enrichment of lake isotope values over summer indicate clear evaporation effect in the lake.

• GW% calculations based on amixing model approach using long term average isotope values for GW (measured regularly from groundwater pipes in the catchment) as a GW signal, and volume weighted summer precipitation isotope value for the month prior to each sampling time as the precipitation signal.

0

50

100

150

200

250

300

350

400

a b c d e f g h i j k l m n o p q r s t u

(um

ol/

L)

pCO2

Head waters close to inlet lake

0

5

10

15

a b c d e f g h i j k l m n o p q r s t u

(um

ol s

ec-1

m-2

)

Flux CO2

June July Sept

lake

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

a b c d e f g h i j k l m n o p q r s t u(u

mo

l/L)

pCH4

0

20

40

60

80

100

a b c d e f g h i j k l m n o p q r s t u(nm

ol s

ec-1

m-2

)

Flux CH4

June July Sept

Head waters close to inlet

• All stream and lake sites were constant sources of CO2 and CH4 to the atmosphere as they were always supersaturated with respect to the atmosphere (pCO2/pCH4 values positive).

• Lowest CO2 flux at most groundwater influenced sites, but highest CO2

concentrations. Lower current speed at most GW influenced sites CO2 input from GW or other processes Emitted to the atmosphere when moving downstream to more turbulent locations.

• CH4 fluxes behave differently. No significant relationship with current speed. E.g. Groundwater hotspot site g: very high CH4 flux.

anaerobic/hypoxic microbial activity?

• Overall stream CO2 flux staying constant throughout the summer, but the concentrations do increase water sources and microbial activity playing a role here.

• Stream and lake CH4 fluxes increasing consistently over summer, but concentrations showing a slight inverse relationship.

• Stream sites seem to be contributing more to CO2 emissions overall than lakes. Stream turbulence plays a role here, however also concentration higher. Difference between stream and lake site CH4 values depend on the sampling occasion.

um

ol s

ec-1

m-2

um

ol /

L

Stream Lake

Flux CO2 pCO2

Stream Lake

Flux CH4 pCH4

nm

ol s

ec-1

m-2

um

ol /

L

• CO2 and CH4 fluxes were measured at site using a Licor (Li-7810) CH4/CO2/H2O Trace Gas Analyzer attached to a floating aluminum chamber build for the purpose.

• Flux measurements were based on the concentration change within the sealed chamber over time (5 minutes).

• CO2 and CH4 concentrations were measured from 50ml water samples using a headspace equilibrium technique, where 50ml of reference air is shaken for 2mins with the 50ml water sample. pCO2/pCH4 in the headspace were then analyzed using an infrared gas analyzer and normalized to the reference air sample by using Henry’s law.

• pCO2/pCH4 values = Cw-Ca where Cw is concentration in water and Ca is concentration in water assuming the concentration of water is in equilibrium with the atmosphere.

0

1

2

3

4

5

0

2

4

6

8

10

12

14

16

18 June

DOC (mg/l) DIC (mg/l)

0

1

2

3

4

5

0

2

4

6

8

10

12

14

16

18

July

0

1

2

3

4

5

a b c d e f g h i j k l m n o p q r s t u

0

2

4

6

8

10

12

14

16

18 September

lakeHead waters close to inlet

DO

C (

mg/

L)D

OC

(m

g/L)

DO

C (

mg/

L)

DIC

(m

g/L)

DIC

(m

g/L)

DIC

(m

g/L)

• High DOC amounts in June (except GW site g). DOC has been delivered to the stream via still snowmelt influenced surface runoff (based on depleted isotope values and heavy prior rain events). DOC relatively easily released from surrounding terrestrial areas with not much vegetation Microbes not consuming DOC that efficiently yet?

• In July: Baseflow conditions. DOC is being efficiently mineralized by mirobes. GW input highest, no new DOC sources DIC values increase throughout the stream.

• In september: Lots of rain before and discharge higher. Groundwater percentage low compared to other times = surface runoff from surrounding terrestrial landscape high DOC values again, but now very efficiently mineralized which can be seen from the inverse DOC/DIC relationship.

• SUVA254 –index (Specific UV Absorbance) correlates positively with the DOC aromaticity and molecular mass. Heavier and more aromatic combounds are thought to be unfavorable for microbes. Some of the GW hotspots seem to be producing more favourable C for microbes at least in June and July.

lake

0.4

0.6

0.8

1

1.2

1.4

a b c d e f g h i j k l m n o p q r s t u

SUVA254

June

July

Sept

Head waters close to inlet

• Water sources in the stream change spatially (along stream continuum) and temporally (throughout the open water season) influencing the CO2 and CH4 fluxes and concentrations.

• Water sources, hydrological conditions and connected terrestrial landscape are controlling the spatial and temporal variability of DOC amount and quality within the stream.

• Still missing microbial community data will give us answers to many previously mentioned microbe related questions. Which goups are presents and active during which sampling occasion? Is community structure dependent on water source spatially or temporally or both? What role do microbes play in C-cycling and resulted GHG emissions?

AcknowledgementsFunding: University of Oulu startup research grant for Hannu Marttila and Jussi Jyväsjärvi, Kaisa-Riikka Mustonen; University of Oulu UArctic Research Chairship Postdoctoral Fellowship and Jeffrey Welker’s Academy of Finland research project, Annalea Lohila; Finnish Meteorological Institute. Collaborators: Metsähallitus, Finnish Environment Institute