Towards Improved Estimates of Upper Ocean Energetics

Preparing to load PDF file. please wait...

0 of 0
100%
Towards Improved Estimates of Upper Ocean Energetics

Transcript Of Towards Improved Estimates of Upper Ocean Energetics

TOWARDS IMPROVED ESTIMATES OF UPPER OCEAN ENERGETICS
A Thesis presented to the Faculty of California Polytechnic State University, San Luis Obispo
In Partial Fulfillment of the Requirements for the Degree Master of Science in Electrical Engineering
by Alexander Grant Wineteer
March 2016

© 2016 Alexander Grant Wineteer ALL RIGHTS RESERVED
ii

COMMITTEE MEMBERSHIP
TITLE: Towards Improved Estimates of Upper Ocean Energetics AUTHOR: Alexander Grant Wineteer DATE SUBMITTED: March 2016
COMMITTEE CHAIR: John Saghri, Ph.D. Professor of Electrical Engineering
COMMITTEE MEMBER: Jane Zhang, Ph.D. Professor of Electrical Engineering
COMMITTEE MEMBER: Ryan Walter, Ph.D. Assistant Professor of Physics
iii

ABSTRACT
Towards Improved Estimates of Upper Ocean Energetics Alexander Grant Wineteer
The energy exchanged between the atmosphere and the ocean is an important parameter in understanding the Earth’s climate. One way of quantifying this energy exchange is through the use of “wind work,” or the work done on the ocean by the wind. Since wind work is calculated according to the interaction between ocean surface currents and surface wind stress, a number of surface current decompositions can be used to decompose wind work calculations. In this research, geostrophic, ageostrophic, Ekman, and total current decompositions are all used to calculate wind work. Geostrophic currents are formed by the balance of surface pressure gradients and the Coriolis effect. Ageostrophic currents, on the other hand, are difficult to calculate because they are made up of many types of currents, and are generally defined as any current not in geostrophic balance. The main component of ageostrophic currents, Ekman currents, are used in this work to approximate ageostrophic currents. Ekman currents are formed by the balance of surface wind stress and the Coriolis effect. Finally, total currents are the sum of all currents in the ocean.
Using high resolution, global NASA ocean models, the wind work on the global oceans is estimated via a number of decompositions, with results finding about 3.2 TW, .32 TW, and 3.05 TW for total, geostrophic, and Ekman wind work respectively, when taking a 7 day window average of surface currents and a 1 day average of surface stress. Averaging period for currents is found to significantly affect the resulting calculated wind work, with greater than 50 percent difference between 1 and 15 days of averaging. Looking at the same total, geostrophic, and Ekman wind work results for 1 day averages of wind stress and surface currents finds 5.5 TW, .03 TW, and 6.3 TW respectively. This result indicates that high frequency currents are very important to wind work. Seasonally, wind work is found to be at a maximum during the Northern Hemisphere (NH) summer, and at a minimum during the NH winter months. To help motivate the funding of a Doppler Scatterometer, simulations are used to show the capabilities of such an instrument in measuring wind work. DopplerScat simulations find that a satellite capable of measuring coincident surface vector winds and surface vector currents, with 1.1 m/s wind speed error and .5 m/s current speed error, could estimate global wind work to within 2 percent accuracy on an 8 day average with daily global snapshots.
iv

ACKNOWLEDGMENTS To all the friends, family, mentors, and teachers that have supported me in my life:
I am truly fortunate to be surrounded by such amazing people. Thank you.
This work was carried out at Cal Poly, San Luis Obispo, under contract with Caltech and the National Aeronautics and Space Administration. Government sponsorship acknowledged. v

TABLE OF CONTENTS Page
LIST OF TABLES ........................................................................................................... viii LIST OF FIGURES ........................................................................................................... ix CHAPTER 1 INTRODUCTION ........................................................................................ 1
1.1 Ocean Model ............................................................................................................. 3 1.2 Data ........................................................................................................................... 6 1.3 DopplerScat ............................................................................................................. 12 CHAPTER 2 ANALYSIS ................................................................................................ 15 2.1 Currents ................................................................................................................... 15
2.1.1 Geostrophic Currents ........................................................................................ 17 2.1.2 Ageostrophic Currents ...................................................................................... 19 2.1.3 Ekman Currents ................................................................................................ 20 2.2 Kinetic Energy Flux and Power .............................................................................. 22 2.3 DopplerScat ............................................................................................................. 23 CHAPTER 3 LITERATURE REVIEW ........................................................................... 26 CHAPTER 4 DISCUSSION............................................................................................. 28 4.1 Currents ................................................................................................................... 28 4.1.1 Total Currents ................................................................................................... 28 4.1.2 Geostrophic Currents ........................................................................................ 29 4.1.3 Ekman Currents ................................................................................................ 31 4.2 Power....................................................................................................................... 32 4.2.1 Total (MITgcm) Power..................................................................................... 32 4.2.2 Geostrophic Power ........................................................................................... 34 4.2.3 Ekman Power.................................................................................................... 36 4.2.4 Ageostrophic Power ......................................................................................... 37 4.2.5 Zonal Averages................................................................................................. 38 4.3 Time Sensitivity ...................................................................................................... 39 4.3.1 Wind Work Time Series ................................................................................... 39 4.3.2 Averaging Period .............................................................................................. 42 4.4 Spatial Resolution Effects ....................................................................................... 46 4.5 DopplerScat Simulation .......................................................................................... 48
vi

4.6 Overall Results ........................................................................................................ 52 CHAPTER 5 CONCLUSION........................................................................................... 54 REFERENCES ................................................................................................................. 55
vii

LIST OF TABLES

Table

Page

Table 1: Variable statistics for Sea Surface Height. ........................................................... 5

Table 2: Model variables used for computations in this work............................................ 6

Table 3: Filter coefficients for 24 sample Hamming filter. ................................................ 8

Table 4: Wind work results for each current type and averaging period. Units in TW.... 53

viii

LIST OF FIGURES

Figure

Page

Figure 1: LLC grid layout. Faces labeled with numbers 1-5; the smaller cells represent grid cells/pixels. .................................................................................................................. 4 Figure 2: Latitude of grid cells in LLC-4320 1/48 degree model....................................... 6 Figure 3: Magnitude response for 24 sample Hamming filter. ........................................... 7 Figure 4: Phase response of 24 sample Hamming filter. ................................................... 8 Figure 5: Original Sea Surface Height map with ice appearing in red. ............................ 11 Figure 6: Sea Surface Height after first pass of ice removal. ........................................... 11 Figure 7: Final Sea Surface Height after filtering............................................................. 12 Figure 8: Process diagram of work done in this research. ................................................ 15 Figure 9: Rossby number for 1/48 degree model.............................................................. 19 Figure 10: Optimization for vertical viscosity results in Az of .014. ............................... 21 Figure 11: Comparison of Parzen and Gaussian filters of length 4. ................................. 24 Figure 12: Total currents as output by the MITgcm model. Seven day average during Nov 2012........................................................................................................................... 28 Figure 13: Geostrophic currents as calculated based on a seven day average of Sea Surface Height during Nov 2012. ..................................................................................... 29 Figure 14: Geostrophic currents plotted over sea surface height in the Southern Ocean off the tip of Southern Africa. Seven day averages shown............................................... 30 Figure 15: Geostrophic vectors plotted over 3D sea surface height data. Again, off the tip of Southern Africa with 7 day averages. ..................................................................... 30

ix

Figure 16: Ekman currents as calculated based on a seven day average of surface stress during Nov 2012. .............................................................................................................. 31 Figure 17: Left to right: Total model surface currents (7d avg), surface stress (1d avg), and KE flux computed from the previous two. Single average from Nov 2012. ............. 32 Figure 18: Zoom of Figure 17 for the Gulf Stream. Total surface currents, surface stress, and KE flux from left to right. ............................................................................... 33 Figure 19: Left to right: Geostrophic surface currents computed from a 7d avg of sea surface height, 1d avg of surface stress, and KE flux computed from the previous two. Single averages from Nov 2012........................................................................................ 34 Figure 20: Zoom plot of Figure 19 for the Gulf Stream. Geostrophic currents, surface stress, and KE flux computed from the previous two from left to right ........................... 35 Figure 21: Surface Ekman currents, surface stress, and kinetic energy flux computed from the previous two. ...................................................................................................... 36 Figure 22: Zoom plot of Figure 21 for the Gulf Stream. From left to right: Ekman currents, surface stress, and KE flux computed from the previous two. .......................... 37 Figure 23: Comparison of KE flux for Ekman currents (left) and Ageostrophic currents (right). ............................................................................................................................... 38 Figure 24: Zonal averages of kinetic energy flux for total model, Ekman, geostrophic, and the total approximation geostrophic+Ekman. ............................................................ 39 Figure 25: Time series for total power over 15 months on 1/24th degree data. Seven day window averaging used for currents, one day for surface stress. ..................................... 41
x
Surface StressCurrentsOceanEkman CurrentsAgeostrophic Currents