spatial analysis of abyssal temperature from the aloha ... · 2/16/2016 · microsoft powerpoint -...
TRANSCRIPT
Spatial Analysis of Abyssal Temperature Variations Observed From the ALOHA
Cabled Observatory and WHOTS Moorings
Roger Lukas*, Fernando Santiago‐Mandujano*, Bruce Howe*, Albert Plueddemann#, Robert Weller#, R. Walter Deppe*, David Murphy+, Richard Guenther+, and Nordeen Larson+
* SOEST/University of Hawaii# Woods Hole Oceanographic Institution+ Sea‐Bird Electronics 2016 Ocean Sciences Meeting
Outline of Talk (not for display)• Introduction/Motivation
– Background on abyssal circulation– Cold events and oscillations– Topographic constraints (x‐ref to Deppe et al. poster)
• Station ALOHA (HOT/WHOTS/ACO)– Use Station ALOHA bathymetry zoom map– Instruments and locations
• Temperature calibrations and accuracy– Sea‐Bird calibration laboratory and sensor history– in situ comparison of co‐located instruments
• Temporal character of temperature differences• Discussion
– large temperature differences over days and O(10 km)
• Conclusions– large differences over days and O(10 km)– signal/noise ~ 10
• Introduction/Motivation– Background on abyssal circulation– Cold events and oscillations. Hawaii Ocean Time‐series (HOT)
• Temperature calibrations and accuracy• Temporal character of temperature differences. ALOHA Cabled Observatory (ACO)
• Spatial character of temperature differences. WHOTS mooring data
• Discussion/Summary
Introduction (not for display)
– Brief background on abyssal circulation– Cold events and oscillations– Topographic constraints (x‐ref to Deppe et al. poster)
bathymetry from Hawaii Mapping Research Group/SOEST/UH
Central North Pacific AABW Circulation and Topographic Constraints
Maui Deep
Kauai Deep
25N
20N
155W160W
Station ALOHA
Nu`uanu Seamounts
Antarctic Bottom Water via Samoan Passage
100 km
Rapid Cold Event Onset and Slow Recovery
HOT‐98 two deep casts, ∆t ~ 2 days∆T ~ 0.03°C
~40 m
~70 m
HOT cruise # indicated below each profile
Hawaii Ocean Time‐series deep Sea‐Bird 911+CTD profiles, dual SBE‐3F temperature sensors calibrated to better than 1 mK after every cruise @ Sea‐Bird
Accuracy and Stability of Thermometers used in SBE‐37 Microcats, SBE‐16 Seacats
1 mDegree Full Scale
Calibration standard used for deep ocean temperatures; shows the stability of the calibration
Histogram of the upper right plot.Average error is calculated for each calibration across the calibration range of ‐1.0 – 32.5 CHigh accuracy and stability
Primary Std at Triple Point of Water Errors evaluated by post deployment calibration relative to calibration at time of manufacture
bottom water overflow pathways
(see Deppe et al. poster 90302)SBE‐37 on outrigger
from Observatorymodule
(see Howe et al. OD11A‐04)
History of Cold Bottom Water Events at Station ALOHAHawaii Ocean Time‐series (Lukas et al., 2001) and ALOHA Cabled Observatory (Alford et al., 2011)
SBE‐37 Microcat 1.8 m above bottom (4728 m) on the ALOHA Cabled Observatory
SBE‐911+ CTD with SBE‐3F T‐sensors
HOT‐98 cold event
Narrative (not for reading)
Not long after the ACO was deployed, a cold event began (Alford et al., 2011). Over several months, potential temperature dropped from 1.110°C to 1.09°C (0.02°C = 20 mK). The cooling was not steady, but was superimposed with rapid changes of ~±5 mK. Temperature increased nearly linearly by about 15 mK to mid‐2012, plateauing through the remainder of the year before increasing again by 5 mK through mid‐2013. The recovery phase of this cold event was characterized by small variations around distinct trends, consistent with turbulent diffusion of heat downward as modeled by Lukas et al. (2001; Deep‐Sea Research).
Narrative (not for reading)
A subsequent rapid cooling was observed in January 2014, followed by large oscillations, warming, and then another rapid cooling in November 2014. A slower cooling similar to 2011 occurred from August‐October of 2015.
SBE‐37 Microcat on ACO/OBS (unpumped, blue) and SBE‐37SM (pumped, red) on ACO/BSP‐1
Separation ~ 5 m
RMS difference ~0.3 mK
Max difference = 1.4 mK
Initial drift of ~0.5 mK for BSP‐1, with stable difference ~‐0.3 mK
Positive BSP‐OBS differences attributed to some dynamic signals being removed from OBS θ when removing thermals caused by OBS power dissipation
1 mK
Data Is Useful But Requires Manipulation
• Red crosses are data from equipment lowered from ship
• Note that continuous sampling yields much more information than intermittent data from ships
4 m°C
HOT CTD
HOT CTD
• Data is de‐spiked and averaged daily
• Variations are real and not equipment induced artifact
20 m°C
RL1
Slide 13
RL1 Just a note - HOT CTD generally about 1.2 km away
CTD measurements are made over a few minutes at most
internal tides are seen by the CTD, but not by ACO daily averagesRoger Lukas, 2/16/2016
Low‐pass filtered (120‐day cutoff) ACO signals are 15 mK peak‐to‐peak
High‐passed variability 27.6 mK p‐p
WHOI‐Hawaii Ocean Time‐series Site moorings with dual SBE‐16 Seacats 37 m above bottom WHOTS‐9 WHOTS‐10 WHOTS‐11
Cold events show increased variance an order of magnitude larger than measurement uncertainty
WHOTS‐9 & ‐11
WHOTS‐10
ACOanchor locations and buoy watch circles
11.4 km
9.6 km12 km
Isobaths at 20 m intervals
Station ALOHA bathymetry and instrument locations
All instruments less than 100 m vertical separation
ACO Microcat is 1.8 m above bottom (4728 m)
All WHOTS Seacats were 37 m above the seafloor on taut mooring segment (instrument knock‐down not significant)
Three WHOTS mooring deployments each with dual SBE‐16 Seacats above anchor releases, 37 m above bottom
Potential temperature differences from ACO reach more than ±0.02°C (20 mK).
ACO current speed and direction reveal complex dynamics
3‐day mooring overlaps
CW, CCW, CW
detided daily averages0.1 m/s
ACO
ACO
WH‐10
WH‐10
WH‐9
WH‐11
10 mK
0.01°C
10 mK
0.01°C
Simultaneous θdifferences across 10‐12 km over three daysACO Microcat is 1.8 m above bottom (4728 m)
WHOTS Seacats were paired 37 m above the seafloor on taut moorings
All instruments less than 100 m vertical separation;
near‐bottom vertical gradient of potential temperature only during cold events
11 mK warming in one day at WH‐11
WHOTS‐ACO differences 0‐11 mK, WHOTS colder (9.6‐11.4 km separation ~along isobaths)
WHOTS‐WHOTS differences 2‐11 mK (12 km separation across isobaths)
Daily averages
Numerous multi‐day, very significant differences between ACO and WHOTS‐11
Indicative of strong near‐bottom baroclinic pressure gradients
WHOTS‐9 tracks ACO very closely except for one month‐long event.
WHOTS‐10 and ‐11 differences from ACO are larger and lead ACO by 1‐day and 2‐3 days respectively. 10‐12 km separation, so ~0.13 m/s propagation speed (faster than mean currents).
negative lags for WHOTS leading ACO
WHOTS‐ACO WHOTS‐9 WHOTS‐10 WHOTS‐11
Mean ∆ ‐0.22 mK ‐2.18 mK ‐1.63 mK
RMS ∆ 2.62 mK 5.07 mK 4.51 mK
~one‐year mooring deployments
Mean θ differences from ACO highly significant for WHOTS‐10 and ‐11, with WHOTS sites to east and southeast of ACO colder
Magnitude and variability of WHOTS‐ACO θ differences changed markedly between WHOTS‐9 and subsequent period
WHOTS‐ACO Potential Temperature Differences
All instruments near bottom, with vertical separation < 100 m
Discussion‐ Large dynamic abyssal temperature signals in the Kauai Deep ―cold overflow events and oscillations ―much greater than instrumental uncertainty
‐ One‐year mean θ differences are real and significant;WHOTS colder than ACO is consistent with bottom water source to east ― the Maui Deep
‐ RMS ACO‐WHOTS differences ~5 mK during 2014‐2015 (~10 km separation)
Simultaneous WHOTS‐ACO differences exceed 1 mK/km over several days
Strong near‐bottom baroclinic pressure gradients
DiscussionSignals propagate from east to west on average, faster than the daily mean near‐bottom flow @ the ALOHA Cabled Observatory
Internal Kelvin waves with periods of days to weeks?Consistent with westward propagation at ALOHA (i.e. counterclockwise propagation around Kauai Deep)
bottom water overflow pathways
(see Deppe et al. poster 90302)
Internal Kelvin wave propagation ?
DiscussionSignals propagate from east to west on average, faster than the daily mean near‐bottom flow @ the ALOHA Cabled Observatory
Internal Kelvin waves with periods of days to weeks?Consistent with westward propagation at ALOHA (i.e. counterclockwise propagation around Kauai Deep)
Forcing?: May be from above (surface weather; baroclinic eddies) or by overflow gravity currents overshooting levels of neutral buoyancy
Issues: Internal Rossby radius of deformation depends on stratification,which depends on frequency and magnitude of cold overflows
Summary• Temperature measurements accurate to better than 0.001°C
across multiple Sea‐Bird instruments, both in calibration lab and deployed in the abyss
• Large dynamic signals (cold overflow events; oscillations) in the Kauai Deep: Signal/noise ratio > 10
• Simultaneous WHOTS‐ACO differences exceed 1 mK/km over several days; one‐year mean differences are much larger than measurement uncertainty
• Signals propagate along isobaths from east to west at ALOHA
• Such non‐tidal dynamics are likely occurring at many other locations along abyssal circulation pathways
aloha.manoa.hawaii.edu
Estimate of Temperature Error Between Calibrations for HOT SBE 3 Thermometers
• Top plot shows an estimate of the error incurred in data collected between calibrations for 6 SBE 3 thermometers used in the HOT program
• Bottom plot is a histogram of top plot
mean and RMS differences < 1 mK
Seafloor Temperature Record
0.01°C
0.001°C
raw temperature (red) with spikes and de‐spiked (blue)
2014 event oscillations ~30 days periodlate 2015 oscillations much shorter period
Instruments and Locations
This slide not for displayAll instruments less than 100 m vertical separation
ACO Microcat is 1.8 m above bottom (4728 m)
All WHOTS Seacats were 37 m above the seafloor.
Bottom depths:
WHOTS‐9 4631 m‐10 4720 m‐11 4671 m
SBE‐37 on outrigger
Simultaneous θ differences across 10‐12 km over three days (between three WHOTS moorings and the ACO during deployment/recovery operations)
WHOTS‐WHOTS differences 2‐11 mK (12 km separation across isobaths)
WHOTS‐ACO differences 3‐11 mK, WHOTS colder (9.6‐11.4 km separation ~along‐isobaths)
Note warming of 11 mK at WH‐11 over one day, with 4 mK warming at ACO lagging by one day All instruments near bottom, with vertical separation < 100 m
Cold events show increased variance an order of magnitude larger than measurement uncertainty