Polar Marine Gravity fields from ERS-1

Abstract

We present new altimetric marine gravity fields of the polar oceans covering the entire Southern Oceans and the Arctic Ocean up to 81.5N derived from the ERS-1 geodetic mission and ERS-1/2 tandem mission. The fields include poorly charted areas which are covered by both seasonal and persistent sea ice. To retrieve gravity measurements over ice covered regions requires reprocessing of the full waveform telemetry data set to correct for errors in surface elevation measurements which occur over ice. In both of the polar oceans the new gravity fields have excellent spatial resolution (approaching 20 km in most areas and 30 km in perpetually ice-covered seas). Hence they provide an unprecedented view of tectonic fabric in polar seafloor. Signatures in the Arctic gravity field, such as that of an apparent fossil spreading ridge in the Canada Basin, provide valuable constraints on the tectonic opening of the basin. In the Antarctic, the gravity field of the Weddell Sea embayment reveals structures which reflect its complex, uncertain history. Gravity over the Amundsen Sea permits a significant, new southward tracing of fracture zones (e.g., the Pahemo, Endeavor) that abut Marie Byrd Land, conjugate to those adjacent to the Campbell Plateau. This result places significant new constraints on the early history of separation between the New Zealand micro-continent and Marie Byrd Land, West Antarctica; it indicates regional extension which is consistent with the Bellingshausen paleo-plate hypothesis.

Keywords: Altimetry, Gravity, Polar, Geophysics

Introduction

In fact, major, missing tectonic plate boundaries, presumably active during the Cretaceous and/or early Tertiary, have been hypothesised for both the Arctic(Vogt, et. al., 1979) and Antarctic(Stock and Molnar, 1987)(Cande, et. al., 1995). Proof that these plate boundaries actually existed and the actual delineation of their specific location has not been possible due to the acute sparsness of surface surveys. High-resolution marine gravity fields derived from ERS-1 altimeter data are now providing the answer to these long standing problems in the tectonic evolution of the polar regions.

Altimetry Over Sea Ice

Figure 1. Showing a typical sequence of radar altimeter echoes over (a) Open Ocean, (b) Sea Ice. The echoes are recorded at a rate of 20 Hz leading to a spacing, along the satellite ground track, of approximately 330m. The portion of the echo recorded corresponds to 32 metres in range. The dashed line, or tracking point, indicates the on-board estimate for the position of the leading edge. Whilst the tracking point for the open ocean data is correctly identified on the leading edge of the return, that for the returns from sea ice is in error by several metres.

Figure 1(a)shows a sequence of return echoes from the open ocean whilst figure 1(b) shows a sequence over an area covered by sea ice. The vertical dashed lines shown in figure 1 correspond to position of the leading edge of the return echo waveform determined by the tracker onboard ERS-1. For returns from the open ocean the onboard processor correctly identifies the location of the leading edge of the return echo, corresponding to a return from the sub-satellite point. Over sea ice, however, the position of the leading edge determined by the onboard processor is in error by several metres. The origin of these errors lies in the fact that the on-board processor assumes a model for the return echo based on reflection from open water.

This effect can be seen in Figure 2(a) where a dramatic increase in height noise accompanies the transition from open ocean to sea ice. The small variations in sea surface height, associated with tectonic features, are completely obscured. To reduce this noise it is necessary to reprocess the altimeter echoes recorded by the spacecraft. To achieve this requires the use of the full echo waveform (WAP) data rather than the ocean product (OPR) data set, normally used for open ocean gravity mapping, which contains only on-board estimates of surface height. By measuring the offset between the range to leading edge of the recorded echo and the range estimated by the on-board processor a correction can be computed(Laxon, 1994). By applying this correction to the on-board estimates the height noise is significantly reduced. Figure 2(b) shows the same height profile as Figure 2(a) but corrected for tracking errors. As can be seen the noise associated with tracker error has been significantly reduced. By selecting data in areas of seasonal sea ice the corrected profile can be compared with data obtained at a different time when no ice is present. The profile in figure 2(c) covers the same ground track as the data shown in figure 2(a) and figure 2(b) but was acquired when no ice was present. Such data can be used to assess how well the retracking algorithm is working when ice is present.

Figure 2: This figure shows height profiles obtained over the same ground track lying in an area of the Southern Ocean subject to seasonal ice cover. (a) On-board height estimates during a period when part of the ground track is ice covered. (b) Same height profile as (a) but with a tracking correction been applied. (c) height profile over the same ground track obtained when the entire length of the profile was ice free.

After the removal of outliers and the elimination of some spurious signals due to the ice cover individual height profiles are smoothed and then differentiated to extract along-track slopes. Grids of ascending and descending slopes are then generated and gravity anomalies calculated using the method of McAdoo and Marks(McAdoo and Marks, 1992).

ERS Marine Gravity Over Sea Ice

An error in ground command programming lead to the operation of ERS-1 in ice mode over most of the Arctic Ocean for several 35 day repeat cycles starting in June 1992 (Cycle 86). The processing of an incomplete data set from this cycle of ERS-1 lead to the first marine gravity field over the Arctic Ocean including permanently ice covered areas. Although the resolution of this field was only 70 km, it provided the first satellite altimeter derived marine gravity field over the Arctic Ocean and lead to significant geophysical discoveries (Laxon and McAdoo, 1994).

ESA continued, with assistance from MSSL, to try and rectify the problems in tracking over sea ice areas. After a number of software patches continuous tracking over sea ice was finally achieved in June 1993. Cycles 96 and 97 which followed provided the first complete coverage in Ocean mode and allowed the generation of an improved gravity field based on three cycles of data. In April 1994 ERS-1 started it's geodetic mission from which the current fields are generated. In addition ERS data from the tandem mission was added in selected areas, in particular in the Bellingshausen Sea, where we suspected the tectonically important signatures of fracture zones might be discovered.

Results

Figure 3. ERS Satellite marine gravity over the Arctic Ocean.

Figure 3 shows the marine gravity map for the Arctic Ocean for all areas up to 82N. The new gravity data reveal a wealth of detail on the structure of the Arctic ocean floor. The gravity field clearly shows the margin of the deep Arctic Basin. The Chukchi Borderland is revealed as a complex system of gravity highs and lows perhaps representing stretched continental crust. The southern tip of the Mendeleev ridge is also seen although it's gravitational signal is somewhat enigmatic. The Eastern margins of the Lomosonov and Nansen ridges lying in the deep Arctic basin also appear clearly in our gravity field. Over the broad Eurasian continental shelf numerous geological structures are observed the nature of which is still the subject of investigation. The a most significant discovery in the Arctic is linear gravity low trending roughly North-South in the Canada Basin. This lineation is now believed to represent the signature of the extinct spreading ridge, lying beneath several km of sediment, which was responsible for the opening of the Canada basin some 100 million years ago(Laxon and McAdoo, 1994).

Figure 4. ERS Satellite marine gravity over the Southern Ocean.

Figure 4 shows the new gravity field of the Southern Ocean. ERS-1 marine gravity has been used for all areas from 160E east to 0W. In other areas data from the Geosat geodetic mission are shown (McAdoo and Marks, 1992). Although the new areas revealed in our gravity data are smaller than in the Arctic, the results are highly significant from a tectonic point of view. Much of the Weddell sea is previously unmapped and a number of tectonically significant features are observed(McAdoo and Laxon, 1996). The poorly mapped continental margin of the West Antarctic is clearly observed in our gravity field. Perhaps the most significant result are the faint N-S trending gravity lineations on the Pacific margin of Antarctica (200 - 220E). These correspond to the fracture zones created by the separation of the Campbell Plateau from Antarctica some 80 million years ago. By matching the fracture zone traces adjacent to Antarctica with conjugate traces near the Campbell Plateau we have reconstructed the relative positions of the two continents some 67 million years ago. The reconstruction reveals an offset in the fracture zones further east which we attribute to the existence of the previously hypothesised Bellingshausen plate(McAdoo and Laxon, 1997).

Validation

Figure 5 shows a comparison between the ERS-1 marine gravity with that obtained from an aircraft survey carried out by the Naval Research Lab. Agreement between the longer wavelength components is extremely close. At shorter wavelengths the ERS-1 gravity reveal oscillations which are likely caused by height errors left by retracking over sea ice.

Figure 5. Comparison of ERS-1 and NRL airborne gravity over a trackline in the Canada Basin. The inset image shows the locations of the track (Airborne data are courtesy of Skip Kovacs and Jon Brozena, Naval Research Lab., Washington).

We estimated the resolution of the original Arctic Ocean gravity as around 70km. Figure 6 shows a coherence plot of the ERS-1 and NRL airborne survey data.This analysis shows that the two data types agree to wavelengths as short as 30-40km where coherency falls below 0.5. The limiting factor in the current fields appears to be the remaining noise on the retracked heights obtained over sea ice.

Figure 6. Coherence plot between ERS and NRL airborne gravity shown in figure 5.

Conclusions

Acknowledgement

References

Vogt, P.R., Taylor, P.T., Kovacs, L.C. & Johnson, G.L. 1979

Detailed aeromagnetic investigation of the Arctic Basin, JGR, 84, pp. 1071-1089.

Stock, J. & Molnar, P. 1987

Revised history of early Tertiary plate motion in the south-west Pacific. Nature, 325, pp. 495-499.

Cande, S.C., Raymond, C.A., Stock, J. & Haxby, W.F. 1995

Geophysics of the Pitman Fracture Zone and Pacific-Antarctic Plate Motions During the Cenozoic, Science, 270, pp. 947-953.

Laxon, S., 1994

Sea ice altimeter processing scheme at the EODC IJRS, 15, 4, pp. 915-924.

McAdoo, D.C. & Marks, K.M. 1992

Gravity Fields of the Southern Ocean From Geosat Data JGR, 97, B3, pp. 3247-3260.

Scott, R.F., et al. 1994

A comparison of the performance of the ice and ocean tracking modes of the ERS-1 radar altimeter over non-ocean surfaces.IJRS, 21, 7, pp. 553-556.

Laxon, S.W. & McAdoo, D. 1994

Arctic Ocean Gravity Field Derived From ERS-1 Satellite Altimetry.Science, 256, pp. 621-624.

McAdoo, D.C. & Laxon, S.W., 1996

Marine Gravity from Geosat and ERS-1 Altimetry in the Weddell Sea. inWeddell Sea Tectonics and Gondwana Breakup, (Eds. B.C. Storey, E.C. King and R.A. Livermore, The Geological Society, London, pp. 155-164.

McAdoo, D. and Laxon, S., 1997

Antarctic Tectonics : Constraints from a new ERS-1 Satellite Marine Gravity Field.Science, 276, 5312, pp. 556-561.

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry