A time series of chlorophyll in response to irradiance in Antarctic coastal waters Maria Vernet 1* , Susana Diaz 2 and Wendy Kozlowski 1 1 Scripps Institution of Oceanography; La Jolla, CA USA 92093-0218; 2 *SCAR Attendee Contact: [email protected] More information at: http://polarphytoplankton.ucsd.edu From the 13-year time series we calculated: 1. Annual cycle based on natural log-transformed monthly averages and fitted with the first four harmonic components (Wilks 1995) 2. Relative Monthly Anomalies = (Monthly average – Annual cycle)/ Annual cycle (Vernet et al. 2009) 3. Cross Correlations of the Relative Monthly Anomalies with environmental variables with lag 0-12 months.Variables tested are meteorological (Temp- erature High, Temperature Low, Temperature Average, Daily Pressure High, Pressure Low and Pressure Average, Wind Speed, Average Melted Precip- itation, Average Solid Precipitation), Total Column Ozone, Irradiance and Planetary Indexes (El Niño 1-2, El Niño 3, El Niño 4 and El Niño 3-4). Irradiance weighted by Biological Weighting Function (BWF) from Neale and Kieber (2000), UVB (303-307nm), UVA (337.5-342.5nm), UVB/UVA. Phytoplankton growth off the Western Antarctic Peninsula (WAP) begins in October, after sea ice ablation and continues through April, closely related to day length. Little documentation exists on seasonal and inter-annual variability on biomass accumulation and demise. This study attempts to address the question of the magnitude of temporal variability in phytoplankton biomass, measured as chlorophyll a (chla) in scales of weeks to years with emphasis on its relationship to irradiance, ultraviolet and visible (290 nm - 700 nm). Est- imates of chla were collected bi-weekly as part of the Palmer LTER at Palmer Station, Anvers Island, Antarctica, and we summarize here thirteen years of data from 1995-2008. References Neale, P. J. and D. J. Kieber. Assessing biological and chemical effects of UV in the marine environment: spectral weighting functions in Causes and Environmental Implications of Increased UV-B Radiation, ed. R. E. Hester and R. M. Harrison, Royal Society of Chemistry, Cambridge, 2000, pp. 61-84. Wilks, S. Time Series, in Statistical Methods in the Atmospheric Sciences: An Introduction, ed. R. Dmowska and J. R. Holton, Academic Press, San Diego, 1995, pp. 284-358. Vernet, M., Díaz, S., Fuenzalida, H., Camilión, C., Booth, C.R., Cabrera, S., Casicca, C., Defarrari, G., Lovengreen, C., Paladini, A., Pedroni, J., Rosales, A., and H., Zagarese. Quality of UVR exposure for different biological systems along a latitudinal gradient. Photochemical and Photobiological Sciences, 8:1329-1345, 2009. Acknowledgements The authors gratefully acknowledge K. Baker for data, the help of the lab managers at Palmer Station for sampling, J. Jones and the many volunteers who have assisted in the data analysis. We thank K. Sines for data management and L.Yarmey for graphical representation. We also thank Antarctic Support Associated and Raytheon Polar Services Company for their logistical assistance at station. This project was funded by NSF Awards OPP90-11927, OPP96-32763, OPP02-17282 and ANT05-28728. 0 0.5 1 1.5 2 2.5 3 3.5 4 Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Chla off Palmer Station shows: a. Changes in abundance in the last decade, with higher summer concentrations and larger overall variability. b. A seasonal signal showing chla accumulation occurs during summer, at the same time that UVR irradiance peaks. c. A positive correlation between irradiance and chla accumulation would seem in opposition to the well-established process of UVB inhibition. Further analysis showed that high UVB/UVA occurred 2-3 weeks previous to the chla peak. Thus, establishing the irradiance “history” is of importance in the analysis of this time series. d. A negative correlation with El Ni ño, presumably through winds and sea ice concentration. Summer chla during El Niña years is consistently higher. e. Further analysis on weekly and daily anomalies will explain the short-term relationship between chla and irradiance, in a scale not available with monthly averages. Aug Sep Oct Nov Dec Jan Introduction Methods Results Preliminary Conclusions The annual cycle in phytoplankton abundance between 1995 and 2008 at Arthur Harbor off Palmer Station, Anvers Island, Antarctica (64.8º S, 64.1º W) was estimated from bi-weekly chlorophyll a concentration, an index of phytoplankton abundance, obtained from the seawater intake at 6 m depth, within the surface mixed layer during the growth season (October to April). 0.01 0.1 1 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Original Time Series Weekly Average Annual Cycle 10 100 Figure 1 - Map showing Palmer Station, Anvers Island, west of Antarctic Peninsula where the time series sampling was performed. Figure 3 - The seasonal cycle is characterized by a maximum at the end of January (3.14 μg/l) and a minimum in July (0.08 μg/l). There is a seasonal asymmetry with gradual increase from July to January and a faster, more pronounced decrease in the fall. The accumulation period lasts, on the average, 4 months and the decline, 3 months. Figure 2 - Time series of chlorophyll-a concentration for the period 1995 - 2008 in Arthur Harbor, Anvers Island, in units of μg/l. A high inter-annual variability is observed, both for winters and summers. In summer, maximum concentrations vary by more than an order of magnitude, a minimum in 1998 - 1999 (~ 2 μg/l) and maximum in 2000 - 2001 (~49 μg/l). Other years of high accumulation were 1995-1996, 2001-2002, 2004-2005, 2005-2006 and 2006-2007. Figure 4 - Monthly Relative Anomalies showing that: a. Inter-annual variability is high. Both positive and negative anomalies can be observed at any time of the year. b. The frequency of summers with high phytoplankton accumulation is ~40% (5 out of 13), as indicated by the monthly anomalies. c. The 1995-2000 period had most of the positive anomalies in the winter and negatives in the summer. In contrast, the period after 2000 had a higher frequency of positive anomalies in the summer, i.e., increased summer biomass. Date chla (μg/l) Table describes cross correlations for the monthly chla anomalies, indicating that: a. Monthly chla correlates positively with temp- erature, UVB/UVA in the same month as well as atmospheric pressure, UVB and UVB/UVA in the previous month. Similarly, positive anomalies in the spring (or 1-2 months of canomalous high chla oncentration) correlate with UVB irradiance 2-3 weeks before (Fig 5) b. Anomalies are negatively correlated with El Niño 3 4 and El Niño 4 indeces, with a 6-8 month lag. Date Monthly Relative Anomaly Chla chla (μg/l) Month Figure 5 - Chla Relative anomaly for October 2000 was unusually high, with daily anomalies indicating high chla on 26 October (red line in Fig. 4). Plotting Ozone (black line) and UVB (303-307 nm in blue), the red dots indicate sampling dates during spring and the red square the day of the large chla anomaly. We can see that sampling follows 3 weeks of high UVB irradiance, supporting the positive correlation of UVB/ UVA and Chla monthly anomalies (Table 1). For every spring in the time series, the higher chla anomalies correlate with cumulative anomalies on UVB and UVB/UVA 15 to 25 days previous to the chla peak. This positive correlation between UVB/UVA and phytoplankton biomass in the same month is unexpected. Further analyses are being carried out to understand the processes involved. Total Ozone Column (Du) Month irradiance UVB (μW/cm 2 ) Ozone UVB/UVA UVB (303-307 nm) Temperature High Atmospheric Pressure High Atmospheric Pressure Average Atmospheric Pressure Low El Ninio 4 El Ninio 3 4 Lag 0 0.18 0.17 0.19 Lag 1 0.21 0.21 0.2 0.18 0.17 0.17 Lag 2 Lag 3 Lag 4 Lag 5 Lag 6 0.18 Lag 7 0.19 Lag 8 0.18 0.18 Lag 9 Lag 10 Lag 11 Lag 12 Table 1 - Green indicates positive correlation and red, negative correlation. Correlations significant at alpha < 0.05. -2 -1 0 1 2 3 4 5 6/1995 6/1996 6/1997 6/1998 6/1999 6/2000 6/2001 6/2002 6/2003 6/2004 6/2005 6/2006 6/2007 - 65˚S - 60˚S 70˚W 60˚W Weddell Sea Bellingshausen Sea Antarctic Peninsula Palmer Station N Drake Passage 65˚S 60˚S 55˚S 50˚S 80˚W 70˚W 60˚W 50˚W Anvers Is.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A time series of chlorophyll in response to irradiance in Antarctic coastal waters
Maria Vernet1*
, Susana Diaz2
and Wendy Kozlowski1
1Scripps Institution of Oceanography; La Jolla, CA USA 92093-0218; 2
*SCAR Attendee Contact: [email protected] More information at: http://polarphytoplankton.ucsd.edu
From the 13-year time series we calculated:1. Annual cycle based on natural log-transformed monthly averages and fitted with the first four harmonic components (Wilks 1995)
2. Relative Monthly Anomalies = (Monthly average – Annual cycle)/ Annual cycle (Vernet et al. 2009)
3. Cross Correlations of the Relative Monthly Anomalies with environmental variables with lag 0-12 months. Variables tested are meteorological (Temp- erature High, Temperature Low, Temperature Average, Daily Pressure High, Pressure Low and Pressure Average, Wind Speed, Average Melted Precip- itation, Average Solid Precipitation), Total Column Ozone, Irradiance and Planetary Indexes (El Niño 1-2, El Niño 3, El Niño 4 and El Niño 3-4). Irradiance weighted by Biological Weighting Function (BWF) from Neale and Kieber (2000), UVB (303-307nm), UVA (337.5-342.5nm), UVB/UVA.
Phytoplankton growth off the Western Antarctic
Peninsula (WAP) begins in October, after sea ice
ablation and continues through April, closely related
to day length. Little documentation exists on seasonal
and inter-annual variability on biomass accumulation
and demise. This study attempts to address the
question of the magnitude of temporal variability in
phytoplankton biomass, measured as chlorophyll a (chla)
in scales of weeks to years with emphasis on its relationship
to irradiance, ultraviolet and visible (290 nm - 700 nm). Est-
imates of chla were collected bi-weekly as part of the Palmer
LTER at Palmer Station, Anvers Island, Antarctica, and we
summarize here thirteen years of data from 1995-2008.
References Neale, P. J. and D. J. Kieber. Assessing biological and chemical effects of UV in the marine environment: spectral weighting functions in Causes and Environmental Implications of Increased UV-B Radiation, ed. R. E. Hester and R. M. Harrison, Royal Society of Chemistry, Cambridge, 2000, pp. 61-84.Wilks, S. Time Series, in Statistical Methods in the Atmospheric Sciences: An Introduction, ed. R. Dmowska and J. R. Holton, Academic Press, San Diego, 1995, pp. 284-358.Vernet, M., Díaz, S., Fuenzalida, H., Camilión, C., Booth, C.R., Cabrera, S., Casicca, C., Defarrari, G., Lovengreen, C., Paladini, A., Pedroni, J., Rosales, A., and H., Zagarese. Quality of UVR exposure for different biological systems along a latitudinal gradient. Photochemical and Photobiological Sciences, 8:1329-1345, 2009.
AcknowledgementsThe authors gratefully acknowledge K. Baker for data, the help of the lab managers at Palmer Station for sampling, J. Jones and the many volunteers who have assisted in the data analysis. We thank K. Sines for data management and L. Yarmey for graphical representation. We also thank Antarctic Support Associated and Raytheon Polar Services Company for their logistical assistance at station. This project was funded by NSF Awards OPP90-11927, OPP96-32763, OPP02-17282 and ANT05-28728.
0
0.5
1
1.5
2
2.5
3
3.5
4
Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun
Chla off Palmer Station shows:
a. Changes in abundance in the last decade, with higher summer
concentrations and larger overall variability.
b. A seasonal signal showing chla accumulation occurs during
summer, at the same time that UVR irradiance peaks.
c. A positive correlation between irradiance and chla
accumulation would seem in opposition to the well-established
process of UVB inhibition. Further analysis showed that high
UVB/UVA occurred 2-3 weeks previous to the chla peak. Thus,
establishing the irradiance “history” is of importance in the
analysis of this time series.
d. A negative correlation with El Niño, presumably through
winds and sea ice concentration. Summer chla during El
Niña years is consistently higher.
e. Further analysis on weekly and daily anomalies will explain
the short-term relationship between chla and irradiance, in a
scale not available with monthly averages.Aug Sep Oct Nov Dec Jan
Introduction Methods
Results
Preliminary Conclusions
The annual cycle in phytoplankton abundance between 1995 and 2008 at Arthur Harbor off Palmer Station, Anvers Island, Antarctica (64.8º S, 64.1º W) was estimated from bi-weekly chlorophyll a concentration, an index of phytoplankton abundance, obtained from the seawater intake at 6 m depth, within the surface mixed layer during the growth season (October to April).
0.01
0.1
1
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Original Time Series Weekly Average Annual Cycle
10
100
Figure 1 - Map showing Palmer Station, Anvers Island, west of Antarctic Peninsula where the time
series sampling was performed.
Figure 3 - The seasonal cycle is characterized by a maximum at the end of January (3.14 µg/l) and a minimum in July (0.08 µg/l). There is a seasonal asymmetry with gradual increase from July to January and a faster, more pronounced decrease in the fall. The accumulation period lasts, on the average, 4 months and the decline, 3 months.
Figure 2 - Time series of chlorophyll-a concentration for the period 1995 - 2008 in Arthur Harbor, Anvers Island, in units of µg/l. A high inter-annual variability is observed, both for winters and summers. In summer, maximum concentrations vary by more than an order of magnitude, a minimum in 1998 - 1999 (~ 2 µg/l) and maximum in 2000 - 2001 (~49 µg/l). Other years of high accumulation were 1995-1996, 2001-2002, 2004-2005, 2005-2006 and 2006-2007.
Figure 4 - Monthly Relative Anomalies showing that: a. Inter-annual variability is high. Both positive and negative anomalies can be observed at any time of the year. b. The frequency of summers with high phytoplankton accumulation is ~40% (5 out of 13), as indicated by the monthly anomalies. c. The 1995-2000 period had most of the positive anomalies in the winter and negatives in the summer. In contrast, the period after 2000 had a higher frequency of positive anomalies in the summer, i.e., increased summer biomass.
Date
chla
(µg
/l)
Table describes cross correlations for the monthly chla anomalies, indicating that: a. Monthly chla correlates positively with temp- erature, UVB/UVA in the same month as well as atmospheric pressure, UVB and UVB/UVA in the previous month. Similarly, positive anomalies in the spring (or 1-2 months of canomalous high chla oncentration) correlate with UVB irradiance 2-3 weeks before (Fig 5)b. Anomalies are negatively correlated with El Niño 3 4 and El Niño 4 indeces, with a 6-8 month lag.
Date
Mon
thly
Rel
ativ
e A
nom
aly
Chl
a
chla
(µg
/l)
Month
Figure 5 - Chla Relative anomaly for October 2000 was unusually high, with daily anomalies indicating high chla on 26 October (red line in Fig. 4). Plotting Ozone (black line) and UVB (303-307 nm in blue), the red dots indicate sampling dates during spring and the red square the day of the large chla anomaly. We can see that sampling follows 3 weeks of high UVB irradiance, supporting the positive correlation of UVB/ UVA and Chla monthly anomalies (Table 1). For every spring in the time series, the higher chla anomalies correlate with cumulative anomalies on UVB and UVB/UVA 15 to 25 days previous to the chla peak. This positive correlation between UVB/UVA and phytoplankton biomass in the same month is unexpected.Further analyses are being carried out to understand the processes involved.
Tota
l Ozo
ne C
olum
n (D
u)
Month
irradiance UV
B ( µW
/cm2)
Ozone UVB/UVA
UVB (303-307
nm)Temperature
HighAtmospheric
Pressure High
Atmospheric Pressure Average
Atmospheric Pressure Low
El Ninio 4
El Ninio 3 4
Lag 0 0.18 0.17 0.19Lag 1 0.21 0.21 0.2 0.18 0.17 0.17Lag 2Lag 3Lag 4Lag 5Lag 6 0.18Lag 7 0.19Lag 8 0.18 0.18Lag 9Lag 10Lag 11Lag 12
Table 1 - Green indicates positive correlation and red, negative correlation. Correlations significant at alpha < 0.05.
-2
-1
0
1
2
3
4
5
6/1995 6/1996 6/1997 6/1998 6/1999 6/2000 6/2001 6/2002 6/2003 6/2004 6/2005 6/2006 6/2007
- 65˚S
- 60˚S
70˚W 60˚W
WeddellSea
Bellingsh
ausen Sea
AntarcticPeninsula
Palmer Station
N
Drake Passage
65˚S
60˚S
55˚S
50˚S80˚W
70˚W 60˚W
50˚W
Anvers Is.
Related Documents
