View
5
Download
0
Category
Preview:
Citation preview
The influence of in-stream processes on stream nutrient chemistry: insights from low- and high-resolution data
Susana Bernal1 Anna Lupon2
Juan Lluís Riera2
Francesc Sabater2
Eugènia Martí1
Photo: A. Lupon 1Centre for Advanced Studies (CEAB-CSIC)
2University of Barcelona (UB) All in Catalonia, Spain
FONT DEL REGÀS IN AUTUMN
WORKSHOP ON HIGH RESOLUTION TIME SERIES HELMHOLTZ UFZ
MAGDEBURG JULY 2014
Based on Frissell et al. 1986
Hierarchical organization of stream ecosystems
The “kidney” The “nephron”
catchment
section
reach
subreach
microhabitat
Forest & Landscape Ecology
Stream Ecology
Characterization of stream nutrient dynamics
9/98 3/99 9/99 3/00 9/00 3/01 9/01 3/02
mg
NO
3-N
L-1
0.0
0.5
1.0
1.5
2.0
SOURCES HYDROLOGIC TRANSPORT
IN-STREAM CYCLING BIOTIC PROCESSING
Stream nutrient dynamics result from the combination of sources and in-
stream nutrient transformations.
The relevant time-scales for understanding biogeochemical transformation at each hierarchical level differ.
Conceptual model of in-stream nutrient cycling
von Schiller D., Bernal S., Sabater F., Marti E. 2014. Freshwater science (in press)
I O = I - UnetUnet = Ugross - R
water column
Ugross
assimilationsorption
co-precipitationnitrification
active streambed
R
mineralizationdesorptiondissolution
Unet < 0 Unet > 0Unet ~ 0
~100 m
Mass balance
In-stream net areal uptake rate (Unet, in mg/m2/s). Empirical approaches
von Schiller, D., Bernal, S. Martí, E. 2011. Biogeosciences
𝑈𝑈𝑈𝑈 =𝑄𝑇𝑇𝑇 ∙ 𝐶𝑇𝑇𝑇 + 𝑄𝐺𝐺 ∙ 𝐶𝐺𝐺 − 𝑄𝐵𝑇𝑇 ∙ 𝐶𝐵𝑇𝑇
𝑤𝑤 ∙ 𝑥
Q discharge (L/s) C concentration (mg/l) ww wet width (m) x lenght (m)
~100 m
~15 m
Spiraling method
Downstream distance (m)
Conc
entr
atio
n
Cl- (dilution)
Nutrients (dilution + bgc)
𝑈𝑈𝑈𝑈 =𝑄 ∙ 𝐶
1 𝑘⁄ ∙ 𝑤𝑤
Q discharge (L/s) C concentration (mg/l) ww wet width (m) k slope (1/m)
A decade+ of stream ecology at La Tordera catchment (NE Spain)
①
③
The Fuirosos stream (since 1999) ① 14.3 km2 m a.s.l. = 115 P = 470 mm PET = 1390 mm RC = 7% Temporary Semiarid Med ②
② The Santa Fe stream (since 2002)
2.5 km2 m a.s.l. = 1140 P = 992 mm PET = 989 mm RC = 17% Perennial Subhumid Med
③ The Font del Regàs stream (since 2009) 13 km2
m a.s.l. = 470 P = 857 PET = 1064 mm RC = 28 % Perennial Subhumid Med
Temporal variation of in-stream net areal uptake rate (Unet)
Unet > 0 Uptake > Release Unet < 0 Uptake < Release Unet ≈ 0 Uptake ≈ Release
Unet can be highly variable over time within the same stream. The stream can alternatively act as a net source, net sink, or be at bgc steady-state.
Sta Fe Fui
-150
-100
-50
0
50
100
150
NO3
Sta Fe Fui-15
-10
-5
0
5
10
15
Sta Fe Fui-15
-10
-5
0
5
10
15U
net (
µg/m
2 /m
in)
2.4 15.9
NH4 1.8 8.1
PO4 10.9 4.6 IQR/|med|
2004-2006. 2 YEARS OF MONTHLY SAMPLING at Sta. Fe and Fuirosos
Uptake
Release
Bernal, S., von Schiller, D., Sabater, F., Martí, E. 2012. J. Geophysical Res. von Schiller, D., Bernal,S., Sabater, F., Martí, E. 2014. Freshwater Science (in press)
Unet > 0 Uptake > Release Unet < 0 Uptake < Release Unet ≈ 0 Uptake ≈ Release
The temporal variation of Unet is not random, but can follow a seasonal pattern that differs among nutrients. Nutrient processing by stream biota responds to the seasonality of environmental variables.
AU WT SP SM0
20
40
60
80
100
AU WT SP SM0
20
40
60
80
100
AU WT SP SM0
20
40
60
80
100NO3 NH4 PO4
Freq
uenc
y (%
) Temporal variation of in-stream net areal uptake rate (Unet)
Adapted from Bernal, S., von Schiller, D., Sabater, F., Martí, E. 2012. J. Geophysical Res.
Spatio-temporal variation of stream nutrient concentration 2010-2011. 2 years of bimonthly sampling at Font del Regàs
SW Rip GW
1.5 m Piezometer
(a) Longitudinal pattern of stream nutrient concentrations
(b) Riparian groundwater nutrient chemistry along the reach
Chl
orid
e(m
g l-1
)
0
10
20
Nitr
ate
( g
N l-1
)
0
200
400
600
Am
mon
ium
( g
N l-1
)
0
15
30
45
8/10
10/10
12/10
2/1
1 4/1
1 6/1
1 8/1
1
10/11
12/11
Ph
ospa
te(
g P
l-1)
0
15
30
45
Bernal, S., Lupon, A., Ribot, M., Sabater, F., Martí, E. 2014. Biogeosciences (in review)
Spatio-temporal variation of in-stream net uptake rate (Unet)
Streamwater
Stream Biota
IN OUT
Groundwater
Unet 𝑈𝑈𝑈𝑈 =
𝑄𝑇𝑇𝑇 ∙ 𝐶𝑇𝑇𝑇 + 𝑄𝑇𝑅𝑅𝐵 ∙ 𝐶𝑇𝑅𝑅𝐵 + 𝑄𝐺𝐺 ∙ 𝐶𝐺𝐺 − 𝑄𝐵𝑇𝑇 ∙ 𝐶𝐵𝑇𝑇𝑠𝑈𝑠𝑠𝑈𝑈𝑈 𝑙𝑈𝑈𝑠𝑙𝑈
NO3 334.0 59.9 224.0 40.1
NH4 4.8 51.6 4.5 48.4
PO4 12.1 52.2 11.1 47.8
σ2 Among sites
% σ2 Within sites
%
-4
-2
0
2
4
230440
6401400
16401840
20502255
24652680
3000-10
-5
0
5
10
-40
-20
0
20
40 NO3
NH4
PO4 Une
t (µg
N o
r P /m
/s)
The longitudinal variation of Unet within a single date can be as high as within the same stream segment over time.
Distance from headwaters (m)
Adapted from Bernal, S., Lupon, A., Ribot, M., Sabater, F., Martí, E. 2014. Biogeosciences (in review)
Spatio-temporal variation of in-stream net uptake rate (Unet)
Unet > 0 Uptake > Release Unet < 0 Uptake < Release Unet ≈ 0 Uptake ≈ Release
# eve
nts
020406080
100
# eve
nts
020406080
100
Distance from headwaters (m)
230
440
640
400
640
840
2050
2255
2465
2680
3000
3200
3430
3690
# eve
nts0
20406080
100
NO
3 N
H4
PO4
In-stream bgc transformations can either follow a spatial pattern (NO3), or be consistent along the reach (NH4). Nutrient processing by stream biota responds to the spatial pattern of environmental variables.
Bernal, S., Lupon, A., Ribot, M., Sabater, F., Martí, E. 2014. Biogeosciences (in review)
Streamwater
Stream Biota
IN OUT
Groundwater
Unet 𝑈𝑈𝑈𝑈 =
𝑄𝑇𝑇𝑇 ∙ 𝐶𝑇𝑇𝑇 + 𝑄𝑇𝑅𝑅𝐵 ∙ 𝐶𝑇𝑅𝑅𝐵 + 𝑄𝐺𝐺 ∙ 𝐶𝐺𝐺 − 𝑄𝐵𝑇𝑇 ∙ 𝐶𝐵𝑇𝑇𝑠𝑈𝑠𝑠𝑈𝑈𝑈 𝑙𝑈𝑈𝑠𝑙𝑈
Frequency of events (%)
High resolution spatio-temporal variation of stream nutrient concentration
2010-2011. 1 year of sampling at 12-h intervals
Chlo
ride
(mg/
l)
0
5
10
15
NItr
ate
(mg/
l)
0.0
0.3
0.6
0.9
9/10 11/10 1/11 3/11 5/11 7/11 9/11 Am
mon
ium
(g/
l)
0
20
40
60
9/10 11/10 1/11 3/11 5/11 7/11 9/11
Phos
phat
e (
g/l)
0
20
40
Can high resolution data provide some mechanistic explanation on the spatio-temporal variation of in-stream nutrient processing?
∆Am
mon
ium
(µg/
l)
-20
0
20
9/10 11/10 1/11 3/11 5/11 7/11 9/11
Pho
spha
te (
g/l)
-20
0
20
2010-2011. 1 year of sampling at 12-h intervals
Nitr
ate
(g/
l)
-200
-100
0
100
200
d.l. ~ 5 ppb (87% of the cases)
d.l. ~ 4 ppb (92% of the cases)
Analytical resolution can still be a limiting factor.
High resolution spatio-temporal variation of stream nutrient concentration
Spatio-temporal pattern of diel cycles of stream nitrate concentration
Nitrate Chloride
∆𝐶 =𝐶𝑛𝑛𝑛𝑛𝑛 − 𝐶𝑑𝑑𝑑
𝐶𝑛𝑛𝑛𝑛𝑛× 100
ΔNO3 during base flow conditions was conspicous at the MID and DOWN sites in spring.
Upstream
Midstream
Downstream
Lupon, A., Marti E., Sabater, F., Bernal, S. 2014. Ecology (in review)
2010-2011. 1 year of sampling at 12-h intervals
C
(%)
-10
0
10
20
30
Time (day/month)19/03 02/04 16/04 30/04 14/05 28/05 11/06 25/06
∆ C
(%)
-10
0
10
20
30
ΔNO3 resulted from in-stream bgc processes.
Lupon, A., Marti E., Sabater, F., Bernal, S. 2014. Ecology (in review)
Stream
Riparian grounwater Nitrate Chloride
Spatio-temporal pattern of diel cycles of stream nitrate concentration
Rip groundwater Stream water
∆𝐶 =𝐶𝑛𝑛𝑛𝑛𝑛 − 𝐶𝑑𝑑𝑑
𝐶𝑛𝑛𝑛𝑛𝑛× 100
2012. Spring (4 months) of sampling at 6-h interval + GPP
∆ C
(%)
-10
0
10
20
30
Dai
ly P
AR
(mol
/m2 /d
ay)
0
5
10
15
20
25
GPP (g O
2 /m2/day)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
[DO] = GPP– R ± E GPP gross primary productivity R respiration E reaeration
Oximeter
single-station diel DO change method (Bott 2006) 30 min-interval
∆NO3 vs GPP: R2 = 0.74
Time (day/month)19/03 02/04 16/04 30/04 14/05 28/05 11/06 25/06
∆ C
(%)
-10
0
10
20
30
Temporal variation of ΔNO3 explained by the influence of light inputs on the assimilation of NO3 by stream photoautotrophs.
Lupon, A., Marti E., Sabater, F., Bernal, S. 2014. Ecology (in review)
Stream
Riparian grounwater Nitrate Chloride
Spatio-temporal pattern of diel cycles of stream nitrate concentration
Rip groundwater Stream water
GPP vs PAR: R2 = 0.74
2012. Spring (4 months) of sampling at 6-h interval + GPP
Spatio-temporal pattern of diel cycles of stream nitrate concentration
Lupon, A., Marti E., Sabater, F., Bernal, S. 2014. Ecology (in review)
Spatial variation of ΔNO3 explained by changes in light and temperature regimes along the stream.
UP 57 66
MID 99 73 83 25
DOWN 103 80 104 57
T (º
C)
5
10
15
20
Time (day/month) 12/03 26/03 09/04 23/04 07/05 21/05 04/06 18/06 02/07
PA
R (m
ol m
-2d-1
)
0
8
16
24
>10ºC
>4 mol/m2/d
days T>10ºC
% days ΣPAR>4 mol/m2/d
%
2012. Spring (4 months) of sampling at 6-h interval + GPP
Past, Present, Future
Long term monitoring vs. hypothesis-driven studies. Analytical resolution issues (PO4, NH4).
Powerful tool for exploring major drivers of in stream solute concentrations within and between catchments at different spatio-temporal scales.
Advances Understand environmental drivers of in-stream nutrient processing. Better assessment of the contribution of in-stream processes to stream nutrient dynamics. Perspective
Hands on work
High resolution temporal data Provide some mechanistic explanation on the spatio-temporal variability of in-stream nutrient processing shown by low resolution data.
I O = I - UnetUnet = Ugross - R
water column
Ugross
assimilationsorption
co-precipitationnitrification
active streambed
R
mineralizationdesorptiondissolution
Thanks to Daniel von Schiller, Miquel Ribot, Silvia Poblador, Eduardo Martín, Clara Romero, Roser Ventosa.
Financial Support MONTES (Consolider-Ingenio, Spanish, MICINN) MED_FORESTREAM (Spanish MEC) REFRESH and EUROLIMPACS (EU) Juan de la Cierva contracts (Spanish MEC) JAE_DOC postdoctoral contract (CSIC).
SUSANA BERNAL postdoc associate at CEAB-CSIC sbernal@ceab.csic.es
Acknowledgements
Recommended