G33B-1231
All Quiet on the Seaward Ice Front?
Horizontal flow of Antarcticaês great ice streams is temporally variable on diurnal and semidiurnal time-scales associated with ocean tides. Sometimes the ice streams exhibit sudden movement pulsations lasting less than hours, but modulated strongly in the diurnal and sub-diurnal time frame ( e.g., Whillans Ice Stream (formerly B)). In other cases, ice-stream flow appears to vary continuously over time in concert with the tide ( e.g., Bindschadler Ice Stream (formerly D)) or with the spring-to-neap tidal cycle ( e.g., Rutford Ice Stream), and can wiggle from side to side in addition to surging forward in the direction of flow. In all cases, these ice-streams enter large floating ice shelves that, to a large degree, control their time-average discharge through the application of backpressure. The present study examines a series of continuous GPS measurements of ice- shelf flow at three stations near the seaward ice front of the Ross Ice Shelf associated with a measurement campaign lasting 23 days. The measurements reveal smooth, sinusoidal variability in the horizontal ice-shelf flow consistent with a linear response to ocean tidal fields. Given the fact that much of the ice shelfês flow speed is determined by ice-stream inflow velocity at the grounding line, and given that this velocity can be strongly pulsed (at least at the grounding line of the Whillans Ice Stream), why are there no velocity pulsations at the ice front? If temporal variability of ice-stream inflow is damped within the ice-shelf interior, temporal variations at the ice front will be –quiet" at least as far as the temporal signal introduced by ice-stream flow is considered. If temporal variability at the seaward ice front retains some of the pulsation associated with ice-stream inflow-- an unquiet seaward ice front--then ice-stream conditions have the ability to potentially alter ice-shelf flow at the sensitive iceberg-calving margin. This would constitute an important interrelationship between ice-stream and ice-shelf dynamism.
G33B-1232
Continuous Measurements of Ice Motion and Associated Seismicity at Bering Glacier, Alaska.
In April 2007, we established an array of GPS and seismic stations on the Bering Glacier, Alaska, to investigate the relationship between glacier motion and glacier-generated seismicity. Bering Glacier is North America's largest mountain glacier and has an area of more than 5000 km2. Dual-frequency GPS data were recorded continuously at 15 second intervals at five stations on the glacier from April to September. Four of the GPS glacier stations were established in a strain diamond located roughly halfway between the equilibrium line and the terminus, at a distance of 40 km from a GPS base station located near the terminus. These four GPS glacier stations were co-located with seismometers, which, together with a fifth seismometer located at center of the strain diamond, form a cross pattern seismic array with a 4-km aperture. The fifth GPS station is located 20 km up glacier from the strain diamond and seismic array, at a point where the upper icefield feeds into a narrow gate to the lower glacier. GPS antennas were fixed to tripods constructed of steel poles drilled 5-7 m deep into the surface of the glacier. This provides a stable reference relative to the glacier surface, which is subject to several meters of annual ablation at the elevation of the strain diamond. The GPS data have been processed using the GAMIT kinematic utility Track. The motion recorded at all sites is rapid (3+ m/day) but smooth and steady down to the temporal resolution of the data. Specifically, we find no evidence for sudden motion events in the timeseries, but rather find only small perturbations superimposed on slowly varying velocities. The seismic records from short period (L-22) and broadband (6TD) instruments reveal frequent icequakes including both emergent low frequency events and impulsive high frequency events. Many of the events recorded show strong time domain correlations across the array. We will construct a timeseries of seismicty using an automatic icequake detector, allowing comparison of the GPS and seismic timeseries. The effect of alternative processing methods for the GPS data, such as GYPSY precise point positioning analysis, will also be explored. http://www.gps.alaska.edu/chris
G33B-1233
Measuring Ice Mass Fluctuations in Southern Alaska and Evaluating the Potential Influence on Tectonic Earthquakes
In southern Alaska between the Malaspina and Bering Glaciers large ice fluctuations occur directly above a shallow main thrust zone associated with subduction of the Pacific-Yakutat plate beneath continental Alaska. Recently the southern Alaskan glaciers have shown a tendency toward earlier glacier melt onset and longer ablation season resulting in increased glacier wastage. Although these glaciers are generally undergoing ice mass loss, the temporal and spatial pattern of surface elevation change is complex and many of the larger glaciers undergo quasi-periodic surges. We have used ICESat-derived elevations along with InSAR-derived digital elevation models (DEM), such as the SRTM-C,-X DEMs, to detect general patterns in ice elevation change for surfaces with variable slope and roughness with exact and near-repeat ICESat tracks. Rather than averaging over large regions or relying on crossovers, we exploited the potential of individual ICESat waveform returns to estimate glacier elevations and surface characteristics. Careful interpretation of the ICESat waveforms must take into account the potential effects of signal saturation, forward scattering due to clouds, and field of view shadowing on pulse shape and the resulting errors in elevation and relief measurements. We have used our ICESat minus ICESat and ICESat minus InSAR-derived DEM elevation change results, along with earlier ice change studies, to estimate ice load changes from 1988-2006 for the southern coastal Alaska glaciers between the Malaspina and Bering Glaciers. The ice load changes were input to finite element models to calculate displacement rates, incremental stresses, and change in the fault stability margin. In 2002-2006, for instance, the predicted displacement rates of the solid Earth due to average annual change in ice loads were up to 20 mm/yr for the vertical and 3 mm/yr for the horizontal. To empirically evaluate the influence of short-term ice fluctuations on fault stability, we compared the seismic rate from a reference background time period against other time periods with variable ice or tectonic change characteristics. For most months regional temperatures in 2002-2006 were warmer than the longer-term average monthly temperatures (1917-2006). We found that the frequency of small tectonic events (2.2 < M < 3.0) in the Icy Bay region increased significantly in the 2002- 2006 time interval relative to the reference time period of 1988-1992. After AEIC removed identified icequakes, the seismic rate change for other time periods (1993-1996, 1997-2001) relative to 1988-1992 was not statistically significant and the tectonic events did not show a seasonal dependence. However, we found that during 2002- 2006 more earthquakes occurred in the late summer and fall than during other seasons.
G33B-1234
Mechanisms for Tidally Induced Glacier Deformation and Flow Variations, East Greenland
Analysis of geodetic observations from Kangerdlugssuaq and Helheim glaciers indicates that the glacier snouts are in nearly free-floating and nearly fully-grounded conditions, respectively. GPS sites on the glacier surfaces exhibit a semidiurnal vertical motion that is in phase with the ocean tides. For Kangerdlugssuaq Glacier, vertical motion and tides are of nearly equal amplitude, while for Helheim Glacier vertical motion is ~25% of the tidal amplitude. The along-flow horizontal motion of these fast moving glaciers (v > 10~meters per day at the terminus) is also modulated by the ocean tides. For data from Kangerdlugssuaq Glacier in 2005, for example, a vertical semidiurnal tidal motion of ~2~m peak-to-peak leads to along-flow position deviations (relative to mean glacier flow of ~38~m~d-1) of ~0.5~m peak-to-peak, and flow speed variations of 10% peak-to-peak. The sense of the horizontal tidal response is such that the instantaneous flow velocity is at a minimum at ocean high tide. We consider four mechanisms for this tidal response: pressure drag, basal drag, gravitational forcing, and flexural deformation. Through consideration of these mechanisms, we use the geodetic data to place constraints on various physical parameters for the glacier. A preliminary analysis indicates that for Kangerdlugssuaq Glacier, only gravitational forcing and flexural deformation are significant contributions. We will present the geodetic data, describe the various mechanisms, and discuss the implications of our analysis.
G33B-1235
Strain Rate Distribution on an Active Ice Shelf Rift Derived by GPS
The majority of mass lost from the Antarctic ice sheet takes place at the fringing ice shelves via iceberg calving. Iceberg calving is controlled by the initiation and propagation of large scale rifts (fractures that penetrate through the entire ice shelf thickness), which precede large tabular iceberg detachment and can lead to ice shelf break- up. Our study area is the Amery Ice Shelf, East Antarctica, where we have observed over the past 5 Antarctic summer seasons an active rift system using a network of GPS and seismic stations. Here we report on the analysis of some of the GPS measurements. Strain rates are determined for a network of 11 sites observed over three weeks during the 2004/05 Antarctic summer period. In order to investigate possible changes in rift fracture mechanics, the results are combined with, and compared to, strain rates obtained in the 2002/03 season, when a sparser 6-station network was deployed for 46 days. Analysis of the network using a cumulative sum approach, obtained by differencing a pair of residual baseline time series situated approximately normal and parallel to the rift, is found to be an effective method to detect small baseline length changes.
G33B-1236
GPS Data Acquisition and Processing for Glaciological Applications: A Community View of the Unique Applications, Challenges, and Solutions
The NSF funded glaciology community and UNAVCO have worked together for over 15 years on a variety of glacier and ice sheet applications primarily in Antarctica, Alaska, and Greenland. We provide highlights of robust solutions gained from this experience, including hardware and set-up considerations, advances in remote data retrieval with satellite based communications, power solutions related to collecting continuous data through the polar dark months, and emerging infrastructure from International Polar Year projects. We also discuss data processing related to fundamental differences between the data processing needs of glaciologists and the solid earth science community: station velocities are much larger and/or modulated by tidal effects often requiring sub-daily solutions, precision requirements are often less stringent, global reference frame issues are less of a concern while local reference frames are complicated by the ice dynamics, and baseline lengths tend to be much shorter (the case of a glacier where its easy to set up a local base on rock) or much longer (the case of an ice sheet with no practical base station locations for 100s of kms). Several data processing solutions are readily available, but there are also substantial data processing challenges facing the community, particularly to produce the time series desired from continuous data collection.
G33B-1237
Improving spatial and spectral estimates of terrestrial mass variations from GRACE data (while unlocking the secrets of the universe)
Geodesy and cosmology may be unlikely bedfellows, in this presentation we discuss the theoretical and practical advances that we have been making in adressing two problems common to both disciplines --- with examples. The first concerns the accurate (low bias, low variance) determination of mass variations from incomplete and noisy gravity data taken at satellite altitude, as is relevant to determine glacier mass fluxes. The second involves the estimation of the power spectrum of such fields from similar data. The first problem requires an estimate that is linear in the data, the second is quadratic. In both cases we take advantage of recently developed spherical localizing "Slepian" basis functions. For the first problem, our competition is from the algorithm developed by Swenson and Wahr, for the second we require comparisons with the maximum-likelihood method of spectral estimation currently in vogue in the cosmological community. We provide an intuitive and graphical discussion of the philosophy underlying the various methods, and illustrate our newest theoretical developements by showing them applied to data obtained as level-2 products from GRACE (with a special application to detecting the Sumatra-Andaman earthquake from space), as well as to the determination of the cosmic microwave-background spectrum (obtained using data from the WMAP satellite). http://www.frederik.net
G33B-1238
Decadal Changes in the 50-Year GECCO Ocean Synthesis
The German partner of the Estimating the Circulation and Climate of the Ocean (GECCO) consortium provided a dynamically consistent estimate of the time-varying ocean circulation over the 50-year period 1952-2001. The estimate results from a synthesis of most of the ocean data sets available during this 50-year period with the ECCO/MIT ocean circulation model. This GECCO estimate is analysed here with respect to decadal and longer term changes in in sea level and meridional overturning. The variability of the meridional overturning is decomposed into contributions from different processes. In agreement with other modeling results, changes in the models MOC strengths are strongly influenced by the southward communication of density anomalies along the western boundary originating from the subpolar North Atlantic which are mainly caused by changes in the Denmark Strait overflow and only marginally influenced by watermass formation in the Labrador Sea. The propagation of density anomalies along the southern edge of the subtropical gyre by baroclinically unstable Rossby waves is found to be at least equally important. Wind driven processes such as local Ekman transport explain a smaller fraction of the variability on those long time scales. Regional changes in sea level are predominantly associated with an intensification of the subtropical gyre circulation and a corresponding redistribution of heat. The horizontal advection of heat due to an increase in wind stress curl is found to explain a major fraction of the estimated regional sea level trends over the last 40 years. Estimated over the top 750 m depth, the increase in thermosteric sea level rise amounts to 1.3 mm/yr on average over the period 1992 through 2001. This corresponds to an increase in upper ocean heat content of 1.5x1022 J/yr and is in agreement with estimates of Willis et al. (2004). However, over the period 1962 through 2001 the global net thermosteric sea level rise is estimated as 0.92 mm/yr from top to bottom, which is three times the recent estimate from Antonov et al. (2005) (0.33 mm/yr). For the last decade, the corresponding global heat flux into the ocean of 1.5 W/m is twice as large as the recent estimate by Willis et al. (2004) due to the heat content change in deeper layers. http://www.ecco-group.org
G33B-1239
Climate Interpretations of GRACE Gravity Field Data: Implications of the Accuracy of the ICE- 5G(VM2)GIA Model for the Inference of Mass Loss from the Antarctic Ice Sheet
Application of the ICE-5G(VM2)GIA model to the interpretation of RL04 GRACE data from the CSR demonstrates that this model delivers a highly accurate prediction of the time dependent gravity field over the North American continent. In this region the signal is dominated by the contribution of the ongoing process of glacial isostatic adjustment to the removal of the ancient Laurentide ice-sheet that once covered the region. When the GRACE field is filtered by the removal of the ICE-5G(VM2) prediction, significant residuals exist over both Greenland and Alaska that must be attributed to the melt-back of land ice from these regions due to high latitude climate warming. The corrections in these regions due to Late PLeistocene deglaciation are small. For Antarctica, however, the inferred rate of mass loss depends entirely upon the GIA correction, with a state of near modern day mass balance being inferred in the absence of the application of a GIA correction and a significant rate of mass loss being implied when such a correction is applied to the raw data. Using the latest masscon solutions for the inference of the rate of surface mass loss from Antarctica (Luthcke et al., personal communication), corrected for the influence of GIA using the ICE-5G(VM2) model, one infers a rate of mass loss from Antarctica of approximately 105 Gt/annum. The issue of the accuracy of the ICE-5G(VM2) model for Antarctic applications is therefor extremely important.The presentation will address this issue in detail.
G33B-1240
Using GRACE and ICESat mission data to estimate GIA on Antarctica
The contribution of Glacial Isostatic Adjustment (GIA) to present-day mass change in Antarctica has a magnitude comparable to the total signal variations of current gravity and altimetry satellite missions. As a consequence, final ice mass balance estimates strongly depend on the assumed Pleistocene ice models and on the local Earth structure that together parameterize the GIA component. Using the latest results from the GRACE and ICESat missions, different GIA models over Antarctica are validated. The two missions provide entirely different measurement types, with GRACE measuring changes in gravity (mass) and ICESat measuring changes in surface height. Using only one of the data sets, it is difficult to separate changes in surface mass from those of GIA. With two data sets measuring the same phenomenon, however, the GIA and surface mass signals should be separable. We will present results from different forward GIA models, where both the Pleistocene ice model and the local Earth structure are varied within a range of plausible scenarios. The signature of GIA will be compared to both gravity and altimetry measurements in an attempt to reconcile the large differences still present between various established GIA models and to allow for a more accurate estimate of ice-mass balance in Antarctica.
G33B-1241
Mass balance in the Mediterranean Sea basin from GRACE
We compare mass variations retrieved from the GRACE (Gravity Recovery And Climate Experiment) mission over the Mediterranean Sea basin to bottom pressure from the ECCO (Estimating the Circulation and Climate of the Ocean) and MERCATOR operational ocean models, and to soil water content from GLDAS (Global Land Data Assimilation System) and ECMWF (European Centre for Medium-range Weather Forecasts) global hydrology models. We show the agreement between ECCO and MERCATOR ocean bottom pressure and GRACE recovered mass variations from mascon and spherical harmonic solutions.
G33B-1242
Arctic Warming, Greenland Melt and Moulins
Air temperatures on the Greenland ice sheet have increased by 4 deg. C since 1991. The ice sheet melt area increased by 30% for the western part between 1979-2006, with record melt years in 1987, 1991, 1998, 2002, 2005, and possibly the most extreme melt year in 2007. The increasing trend in the total area of melting bare ice is unmistakable at 13% per year, significant at a probability of 0.99. Hence, the bare ice region, the wet snow region, and the equilibrium line altitude have moved further inland and resulting in increased melt water flux towards the coast. Warm and extended air temperatures are to blame for 1.5 m water equivalent surface reduction at the long-term equilibrium line altitude, 1100 m elevation at 70 deg. N during summer 2007. Increase in ice velocity in the ablation region and the concurrent increase in melt water suggests that water penetrates to great depth through moulins and cracks, lubricating the bottom of the ice sheet. New insight was gained of subsurface hydrologic channels and cavities using new instrumentation and a video system during the melt peak in August 2007. Volume and geometry of a 100 m deep moulin were mapped with a rotating laser, and photographs with digital cameras. Sub-glacial hydrologic channels were investigated and filmed using a tethered, autonomous system, several hundred meters into the ice. These new results will be discussed in view of the rapid increase in melt area and mass loss of the Greenland ice sheet due to increasing air temperatures.
G33B-1243
Relative Importance of Ocean Mass and Volume Changes to Global Sea Level Rise
Sea level is an integrated indicator of climate variability, reflecting changes in the dynamic and thermodynamic in atmosphere, ocean and cryosphere. The rate of sea level rise and its causes is a topic of active debate. We examine the relationship between 50 year long records of global sea level (GSL) calculated from 1023 tide gauge stations and global ocean heat content (GOHC), glacier and ice sheet melting. The lack of consistent correlation between changes in GOHC and GSL during the period 1955-2003 argues against GOHC being the dominant factor in GSL as is often thought. We provide clear evidence of the substantial and increasing role in GSL from the eustatic component (47 per cent) compared with the contribution from increasing heat content (25 per cent), suggesting that the primary role is being played by the melting glaciers and ice sheets. There remains about 23 per cent of GSL rise unaccounted for by the best estimates of both eustatic and thermosteric effects. This fraction also exhibits large variability that is not readily associated with known causes of sea level variability. The most likely explanation of this unknown fraction is underestimated melting, climate- driven changes in terrestrial storage components and decadal time scale variability in global water cycle. This argues for a concerted effort to quantify changes in these reservoirs.
G33B-1244
Building a Climate Record of Sea Level Change
Sea level change is a sensitive indicator of climate change because it responds both to changes in ocean temperature (where most of the excess heat from climate change is being absorbed) as well as to exchanges of water mass between the continents and the oceans (which is dominated over long time periods by the melting of ice in glaciers and ice sheets and changes in the water cycle). Therefore, the record of sea level change collected by the TOPEX/Poseidon and Jason-1 (and soon Jason-2) satellite altimeter missions is critically important for climate change studies. However, the use of satellite altimeter data for this purposes pushes the accuracy limits of the measurements, and thus great care must be taken to monitor the performance of the instruments and their biases, investigate anomalies, improve the measurement corrections, and place the measurements in a well- defined Earth-fixed reference frame. We have carefully built a publicly-accessible climate record of sea level change using TOPEX/Poseidon and Jason-1 altimeter data, which provides many different sea level products for use by various investigations. In addition, results from calibrating this record using tide gauges are shown. This climate record will help scientists understand how the Earth is responding to climate change, improve predictions of future sea level change, as well as ascertain the socio-economic impacts. http://sealevel.colorado.edu
G33B-1245
Improved GRACE Measurements of Climate Change: Alternative and Advanced Processing Techniques
As the GRACE project matures, innovative analysis algorithms and approaches have begun to receive significant attention for potentially providing improved accuracy and spatio-temporal resolution. Example of these include non-spherical harmonic basis functions (such as mascons), use of intersatellite range or range-acceleration data in addition to the standard range-rate data, and others. The analysis group at JPL is unique in having a very large number of these capabilities in a single software system which allows clear and controlled comparisons of each approach, which we believe is critical to providing accurate estimates of true uncertainties in geophysical estimates. We have previously discussed our various mascon results, which utilize global 2 and 4 degree spherical caps, but recently we have also focused on uses of the intersatellite range acceleration data. We have computed various solutions (spherical harmonic, mascon, and others) based on this data type, which by its nature is more spatially localized than the integral data types. In assessing these solutions for Greenland, Antarctica, Alaska, and smaller glaciated areas, our analyses appear to demonstrate improved spatial resolution and very significantly reduced aliasing noise. In this talk we will provide a complete description of our results, including careful comparisons with range-rate based spherical harmonics and mascons.
G33B-1246
Thermal Wind Forcing, Atmospheric Angular Momentum and Earth Rotation: Origin of the Earth's Delayed Response to ENSO
Interannual length-of-day variations and ENSO indices such as the Southern Oscillation Index (SOI) and Nino 3.4 SST are well correlated as a consequence of angular momentum conservation. During an El Nino event, the westerly winds increase, which raises the atmospheric angular momentum (AAM); as a result, the solid Earth must slow down, which increases the duration of the day (length-of-day: LOD). However, a lag has been observed with the SOI and Nino 3.4 SST leading the LOD and AAM series by one – two months; to date no dynamical explanation has been offered. The dominant excitation mechanism of interannual LOD is the wind term driven largely by thermal winds arising from the poleward gradient of tropical temperature (TT). We show that the TT gradient (TTG), which peaks 1-2 months after the Nino 3.4 SST anomaly, is the source of the thermal winds that drive the LOD anomaly and account for this well-known ENSO-Earth rotation lag. This work serves to underscore the global signature of climate processes such as ENSO, and highlights the utility of geodetic methods for studying both natural and anthropogenic climate variations and their interactions with broader aspects of the Earth system.
G33B-1247
Detection of Characteristic Precipitation Anomaly Patterns of El Nino / La Nina in Time- variable Gravity Fields by GRACE
GRACE (Gravity Recovery and Climate Experiment) satellites, launched in March 2002, have been mapping monthly gravity fields of the Earth, allowing us to infer changes in surface mass, e.g. water and ice. Past findings include the ice mass loss in southern Greenland (Luthcke et al., 2006) and its acceleration in 2004 (Velicogna and Wahr, 2006), crustal dilatation by the 2004 Sumatra Earthquake (Han et al., 2006) and the postseismic movement of water in mantle (Ogawa and Heki, 2007). ENSO (El Nino and Southern Oscillation) brings about global climate impacts, together with its opposite phenomenon, La Nina. Ropelewski and Halpert (1987) showed typical precipitation patterns in ENSO years; characteristic regional-scale precipitation anomalies occur in India, tropical and southern Africa and South America. Nearly opposite precipitation anomalies are shown to occur in La Nina years (Ropelewski and Halpert, 1988). Here we report the detection of such precipitation anomaly patterns in the GRACE monthly gravity data 2002 - 2007, which includes both La Nina (2005 fall - 2006 spring) and El Nino (2006 fall - 2007 spring) periods. We modeled the worldwide gravity time series with constant trends and seasonal changes, and extracted deviations of gravity values at two time epochs, i.e. February 2006 and 2007, and converted them into the changes in equivalent surface water mass. East Africa showed negative gravity deviation (-20.5 cm in water) in 2006 February (La Nina), which reversed to positive (18.7 cm) in 2007 February (El Nino). Northern and southern parts of South America also showed similar see-saw patterns. Such patterns closely resemble to those found meteorologically (Ropelewski and Halpert, 1987; 1988), suggesting the potential of GRACE as a sensor of inter-annual precipitation anomalies through changes in continental water storage. We performed numerical simulations of soil moisture changes at grid points in land area incorporating the CMAP precipitation data, NCEP/NCAR temperature data, and potential evapotranspiration calculated after Thornswaite (1942). We took out the soil moisture anomalies in 2006 February and 2007 February by modeling its time series in the same way as gravity, and confirmed that they are quantitatively consistent with GRACE gravity deviations. Out study demonstrates that satellite gravity data can detect not only of global warming signals in high latitude regions but also inter-annual climate changes in low and middle latitude continental regions.
G33B-1248
Modes of variability of atmospheric excitation functions for polar motion
We undertake a study to examine the modes of atmospheric regional variability that drive polar motion. From atmospheric surface pressure fields in the NCAR-NCEP reanalysis we have calculated atmospheric-mass polar motion excitations on a fine-resolution network of sectors over the extended 60-year period 1948-2007. Earlier work demonstrated the intramonthly variability in these sectors, and highlighted regions that were particularly important to polar motion excitation, like the atmosphere over central Eurasia, but was based typically on a shorter period and a coarser network. Here we expand the study as well to examine monthly and interannual time scales. We utilize the complex empirical orthogonal function method to select the vector modes that explain independently the most variance of polar motion excitation. Modes comprise a measure of variance explained (the eigenvalue), a geographical structure (the eigenvector), and a time series (mode loadings), so that the regional and temporal importance of the atmosphere with respect to the polar motion function are revealed. Regionally, only the atmosphere over land is important, as the Inverted Barometer model, which greatly reduces all variability over the ocean, is appropriate at the time scales considered here. The variability appears strong over portions of Eurasia and North America in particular for most modes, but has some southern hemisphere variability as well for the first mode too. The time series of these modes, especially when seasonally stratified, may reveals some influence of climate patterns like the El Nino/Southern Oscillation, North Atlantic, Arctic, Antarctic, and Pacific-North American Oscillations, based on their indices.