Understanding DO Mobilization Dynamics through High ... · Understanding DO Mobilization Dynamics...
Transcript of Understanding DO Mobilization Dynamics through High ... · Understanding DO Mobilization Dynamics...
30
20
10
0
-10
Understanding DOC Mobilization Dynamics
through High Frequency Measurements in a Headwater Catchment Benedikt Werner , Andreas Musolff, Oliver Lechtenfeld, Gerrit de Rooij, Marieke Oosterwoud, Toralf Keller & Jan Fleckenstein Contact: [email protected] Helmholtz Centre for Environmental Research - UFZ, Leipzig
T h e R a p p b o d e F i e l d S i t e
Harz Mountains (540 m a.s.l.)
Temperate climate
Mean temperature: 6.9 °C
Annual precipitation: 1200 mm
Drainage area: 2.58 km²
Forest: 77 %
Grasslands: 11.2 %
Groundwater-influenced soils: 25 %
Drainage density: 2.91 km/km²
Processing of in situ UV-vis Absorption Spectra (220 nm - 720 nm)
DOC concentration (from PLS Regression)
Spectral Slope (275 nm - 295 nm)
Specific UV Absorption: SUVA254 =
Modeling DOC concentration, SUVA and Spectral Slope
Explaining time series variance
Linear Regression model selection and validation
Akaike‘s Information Criterion
Variance Inflation Factor
Nash Sutcliffe Eficiency (NSE)
Event-based patterns in DOC mobilization
Seasonal trends for the Slope, Intercept and R² of the Regression lines, Hysteresis Index and Spectral Slope (not shown) with maxima in winter and summer
Best explanation by 15 days mean temperature before event (r = > 0.52)
Indicate shift in mobilization mechanisms over the seasons
DOC quality variation in the stream
Linear Regression Modeling
C o n c l u s i o n s The variations in DOC concentration and
quality can be explained by four
variables: discharge (Q), temperature
(Temp) antecedent wetness (AW), and
the antecedent discharge-temperature
ratio ()
These variables can be linked to event-
based DOC source activation
(discharge) and more seasonal controls
(AW, ) of DOC production
O u t l o o k
Link stream-DOC concentration and composition with riparian groundwater flow and discharge generation
Fingerprint DOC compounds in the stream and riparian zone
Derive a mechanistic model of catchment-scale DOC export which includes the better understanding of DOC mobilization and transport
I n t r o d u c t i o n
Increasing DOC exports from catchments impact the ecology and quality of downstream waters
Riparian zones are the most dominant source zones for in-stream DOC in a temperate climate
High-flow events (snowmelt, heavy rain) drive DOC deliverance to streams
Hydrological and biogeochemical controls of mobilization and transport in riparian zones are complex and still elusive
High frequency, in situ sensing techniques provide new insights into dominant mobilization mechanisms and source zones of DOC
SAC254nm
DOC conc.
DO
C c
on
cen
tra
tio
n
Stream Discharge
Grab sampling
DO
C c
on
cen
tra
tio
n
Stream Discharge
High Frequency
Leipzig
Field Site
Berlin
G e
r m
a n
y
August 2013 January 2014 August 2014
August 2013 January 2014 August 2014
August 2013 January 2014 August 2014
Precipitation [mm/d]
1.5 Year High-Frequency Data Set (15 min)
Discharge derived from water level*
In-situ DOC concentration derived from UV-vis spectra
Weather Data
0.4
0.2
0
15
10
5
0
Air
tem
pera
ture
[C°]
0
40
80
[m³
s-1]
[mg
L-1]
0.75
0.5
0.25
0
1
0.9
0.8
0.7
5
4
3
2
0.4
0.2
0
-0.2
-0.4
Oct 13 Jan 14 Apr 14 Jul 14 Oct 14
Oct 13 Jan 14 Apr 14 Jul 14 Oct 14 Oct 13 Jan 14 Apr 14 Jul 14 Oct 14
Oct 13 Jan 14 Apr 14 Jul 14 Oct 14
Spectral Slope vs. Log Discharge SUVA vs. Log Discharge Comparison of the (scaled) DOC concentration -,
SUVA - and Spectral Slope Models
.0
.3
.6
Q
Q:AW
AW Q:
*calculated after Lloyd et al. 2016
O b j e c t i v e s
i) Investigate the hydrologic and biogeochemical controls for the mobilization and transport of DOC from riparian soils to the streams
ii) Evaluate how concentration and composition of DOC change during events and varying seasonal conditions
iii) Extract the most decisive factors which are needed to explain the variance of DOC concentration and composition in streams
5
4
3
2
1
-1
-2
-3
-6 -4 -2 Log(discharge)
-6 -4 -2 Log(discharge)
SUV
A [
L m
g-C
−1 m
−1 ]
Spe
ctra
l Slo
pe
DOC
wet
DOC
DOC
dry
All models use the same parameters
NSE of DOC: 0.62, SUVA: 0.52, Spectral Slope: 0.37
Different parameter loadings allow for mechanistic interpretations
* Each point indicates one discharge event. Slope, Intercept and R² were calculated from linear regressions of log DOC
vs. log discharge. Hysteresis analysis was conducted in non-log space.
DOC
SUVA
Spectral Slope
Slope of the Regression Lines*
Intercept of the Regression Lines*
R² *
15 days mean
temperature
before event
Hysteresis Indices** of Events
Contact: [email protected]
*Note: Discharge peaks cut above 0.38 m³ s-1 based on maximum observed discharge.
**HI calculated after Lloyd et al. 2016
In-stream DOC quality varies with discharge and antecedent temperature
Systematical change of DOC quality with increasing discharge:
Change in DOC-compound size distribution and increasing aromaticity
Less variability: Convergence of SUVA values and Spectral Slope
Indicates a shift in the activated source zones with rising discharge
Change of size distribution
Incr
easing aro
maticit
y