PART E—PROJECT DESCRIPTION E1 PROJECT TITLE: Prediction of Sea Level Change around Australia and its Calibration and Validation by Satellite-Geodetic Measurements E2 AIMS AND BACKGROUND Describe the background to the project. Include information about recent international progress in the field of the research, and the relationship of this proposal to work in the field generally. Refer only to refereed papers that are widely available to national and international research communities. E2.1 Problem Definition It is widely acknowledged that the Earth’s atmospheric temperature is increasing (eg. IPCC, 2001), thus causing previously ice-covered areas to melt (i.e., deglaciation). This is alleged to cause the phenomenon termed “sea level rise”. However, sea level will not necessarily rise in all parts of the globe. Indeed, it may even fall in some regions. This is because there is a redistribution of mass (due to movement from ice-covered regions to the oceans, changes in the ocean circulation, and isostatic rebound/depression due to surface un/loading), which causes the Earth’s centre of gravity (geocentre) and orientation of the spin axis to change (eg. Gasperani et al., 1986; James and Irvins, 1997; Mitrovica et al., 1991, 2001; Mitrovica and Milne, 1998; Sabadini and Vermeersen, 1997). In addition, the redistribution of mass alters the Earth’s gravitational field and hence the geoid (eg. Tamisiea et al., 2001). Since mean sea level mirrors the geoid (i.e., the flow of water is driven by gravity), the result is that sea level, though increasing in volume, will rise and fall by different amounts in different regions. Clearly, this will have differing consequences in different parts of the world (Figure 1). before after Figure 1. The concept of geocentre, spin axis, gravity and hence sea level change after deglaciation: the resulting change in the geoid will cause sea level to vary by different amounts in different areas, and may fall in some. Most previous investigations in this field have concentrated on historical changes in sea level back to the last Ice Ages (eg. Kaufmann, 2002; Donato et al., 2000; Feng and Hager, 1999), including studies around Australia (eg. Chappell, 1983; Lambeck and Nakada, 1990; Milne and Mitrovica, 1998b). The proposed project is novel in that it seeks to make future predictions of sea level change based on syntheses that have been calibrated and validated (albeit to a time-limited extent) by satellite geodetic data. This is a demanding project, principally because of the many issues to be considered and the Earth is a complex dynamical system (see Figure 3) with different geodynamic phenomena interacting in various ways (eg. Cazenave et al., 1999; Lambeck, 1998). According to ARC guidelines, this is classified as a combination of pure and strategic basic research; that is, it is a combination of curiosity-driven experimentation, but the outcomes should lead to improved decision making for Australia and its neighbouring countries. At present, many predictions of sea level change are based on the global estimate of an 1.8mm per annum rise (eg Douglas, 1995), thus neglecting the phenomena that cause it to fall or rise by different amounts in different areas (cf. Flemming et al., 1998; Mitrovica and Davies, 1995). This needs redressing. 10E2.2 Primary Aims and Outcomes This project will refine a synthetic model of the Earth’s mass distribution (based on the best data currently available) to simulate changes in the gravity field, spin axis and geocentre caused by deglaciation. The synthetic gravity field developed as part of ARC large grant A00001127 to Featherstone (see the progress report in Section D2) will be adapted for this purpose by adding ice sheets, which will then be ‘analytically’ melted and the resulting changes to the geoid (mean sea level) quantified as a function of position. These simulated changes in the geoid will be used to map the expected rises and falls in regional sea level as a function of time. The synthetic model will also be used to predict various scenarios based on different (realistic) rates and distributions of deglaciation and the associated isostatic rebound of the Earth’s crust after the removal of the ice masses (cf Figure 2), as well as the isostatic depression caused by the additional loading (cf. Johnston and Lambeck, 1999; Peltier, 1999). This forms the curiosity-driven component of the research program. The changes in sea level predicted by the synthetic Earth model will be calibrated and validated (cal-val) using contemporary estimates of (i) mass balance from satellite radar altimetry and (ii) temporal changes in the Earth’s gravity field from dedicated satellite gravity field missions (cf. Satoi et al., 2001). The upcoming ICESAT satellite altimeter mission will observe changes in the ice coverage very precisely and with a high spatial and temporal resolution. Since most of the resulting geodynamical processes leave their signatures in the gravity field of the Earth, the static and time-variable components of the Earth’s gravity field will be observed very precisely by the CHAMP and GRACE, and ultimately GOCE satellite gravity field missions (cf. Bentley and Wahr, 1998). Importantly, this combination of data types will provide a better calibration and cross-validation of the techniques adopted, and hence a more reliable prediction of the change of environmental parameters. This forms the strategic-basic component of the research program. The resulting cal-val’ed synthetic model will then be used to extrapolate the expected regional changes in sea level from various scenarios (rates and locations) of deglaciation. As well as providing a useful dataset for generic Earth system science, the models can also be used for improved decision making and planning. Examples include improved coastal defences against storm surges, especially in the low-lying islands of the Asia-Pacific region, relocation of population centres for the longer term, and management/relocation of emergency services. E3 SIGNIFICANCE AND INNOVATION Describe how the research is significant and whether the research addresses an important problem. Describe how the anticipated outcomes advance the knowledge base of the discipline and why the project aims and concepts are novel and innovative. Detail what new methodologies or technologies will be developed. This research program indisputably addresses an important problem. Global climate change is probably the most well known environmental phenomenon facing society today, and more so in the future. Most people believe that sea level will rise with global warming and are totally unaware that it may also fall in some regions because of the complex interactions in Earth system dynamics. The importance of this project is that it will indicate whether sea level will rise or fall in Australasian, and thus educate people as to the regional impacts of global warming. The project will add significantly to the knowledge base in this discipline because of its novel approach of looking towards the future. As stated earlier, previous studies in the field have only considered the historical effects of deglaciation after the last Ice Ages (eg. Kaufmann, 2002). Admittedly, this was driven by the need to quantify these effects for geodynamic studies. The result of this research program may also provide some useful data to these geodynamic studies, but the primary focus is on the quantification of regional sea level changes for improved decision-making. 11E3.1 Significance The proposed program is significant in that it will provide explanation of the different estimates of sea level change observed at different places around the Earth (cf. Mitrovica and Davies, 1995). However, these observed differences can be explained (reasonably adequately) by other geophysical processes such as land subsidence or uplift, rather than real sea level change. Therefore, there is the need to separate these conflicting signals. One outcome of this program is that it will allow models to be constructed that will predict whether sea level should be falling or rising based on the changes in the geoid. These can be applied to field measurements to remove biases, thus generating a better estimate of the true sea level change. The project is also of practical significance to the Asia Pacific region, which characterised by several low-elevation island nations. The benefits of this research to Australia and the region are given in Section E5. E3.3 Innovation As stated, current predictions of sea level change assume a constant rise of ~1.8mm/annum. The innovative nature of this program lies in the fact that it is predictive of rise and fall (i.e., true change), whereas most previous studies have only looked back in time to the last Ice Ages (Section E2.1). For this predictive element to be valid, the synthetic model will be calibrated using data from new satellite geodetic techniques. The ICESAT mission data will be used to estimate rates of deglaciation, and the resulting gravity changes will be sensed by the highly precise dedicated satellite gravity field missions, GRACE, CHAMP and GOCE. No such program has been attempted, because of the lack of these precise datasets, and to the best of our knowledge, no such program has been proposed. Therefore, this program has the potential to place Australia as an important contributor to studies of global change. It will also allow Australia to provide better-informed aid to its Pacific neighbours. E4 APPROACH Outline the conceptual framework, design and methods and demonstrate that these are adequately developed, well integrated and appropriate to the aims of the project. E4.1 PHASE ONE: Refinement of the existing synthetic Earth gravity model (1 year) The synthetic gravity forward model developed as part of ARC large grant A00001127 (Section D2) uses the most up-to-date and realistic estimates of the Earth’s mass-density distribution to generate the Earth’s external gravitational field and the geoid (mean sea level) using Newton’s laws of gravitation. However, this model is essentially static in that it assumes that the rotation rate, spin axis and geocentre are invariant. While this was acceptable for the gravity field studies for which it was designed, the proposed project will refine this synthetic Earth model to the fourth dimension of time and will also generalise it to consider the viscoelastic (isostatic) responses due to the removal of loads from the Earth’s surface (due to deglaciation) and the addition (due to mass redistribution), while preserving the mass of the Earth. For a pseudo-static case, this is a trivial process. The ice sheets can be added to the synthetic Earth model based on the most up-to-date estimates of ice-sheet thickness (eg. Kaufmann, 2002) and the corresponding geoid deduced. All investigators have a large amount of experience in this aspect. Some additional, but simple, work will be needed to generate the changes to the location of the geocentre and the spin axis (Figure 1). These computations will be based upon the laws of conservation of angular momentum, as well as the fundamental equations defining the Earth’s moments and tensors of inertia (eg Marchenko and Schwintzer, submitted). The problem becomes more complicated when the viscoelastic response of the Earth is taken into consideration. Specifically, as the ice masses are removed from the Earth, the crust rebounds at a rate governed by the local mechanical properties of the crust. This phenomenon is well known as postglacial rebound and is currently observed at a maximum rate of ~10mm/annum in Fennoscandia (eg. Milne et al., 2000). Similarly, as the additional masses are redistributed around the Earth, these cause an additional load that depresses the crust (eg. Johnston, 1993). Importantly, these rebound and 12depression effects change the moments and tensors of inertia of the Earth, its external gravity field and long-term tides. Accordingly, there is an iterative element to this process. The above are important considerations if the results of this research are to be compared with contemporary and future tide gauge estimates of sea level because the vertical motion of the crust, due to the additional loading of the sea water, must be separated from the sea level change measurements. Therefore, the synthetic Earth model will be adapted quite considerably to consider viscoelastic effects of surface loads. Even a cursory glance of the literature in this area (see the reference list) shows that this is a complex problem, with widely differing philosophies and models used. Accordingly, these different models will be programd into the synthetic Earth model and numerous simulations run. Importantly, the comparison with contemporary satellite-geodetic data (phase two of the proposed program) will be used to select the appropriate model. Coupled with the above are the variable timescales over which these geodynamic processes occur (Figure 2), which needs to be considered during the four-dimensional construction of the synthetic Earth model. Of these processes, the ocean circulation and Earth rotation components are relatively straightforward to model using existing techniques (eg. Milne and Mitrovica, 1998a). As stated, however, the postglacial rebound and loading effects are problematic and controversial. These effects will be given more attention, where the synthetic Earth model is programmed to consider the various approaches, and simulations run. Figure 2. Time and length scales of Earth rotation, ocean circulation and postglacial uplift/depression that will be implemented in the synthetic Earth model E4.2 PHASE TWO: Calibration and Validation using Satellite Geodetic Data (2.5 years) Two distinct types of satellite geodetic missions are planned in the near future; these are the GRACE, CHAMP and GOCE gravity field missions and the polar ice-shield ICESAT mission. These will provide extremely detailed information about contemporary changes in the ice-ocean mass balance and the resulting temporal and spatial signature in the Earth’s gravity field. As such, they will be used to calibrate and cross-validate the synthetic Earth model so that its longer-term predictions (phase three) are scientifically justifiable. ICESAT: This mission will measure the distance of the ice sheet to the spacecraft with a precision better than 10 cm. As such, the variability of the glacial ice mass thickness can be inferred from these data. The change of the ice-shield thickness can be converted into variations of sea level. Using standard topographic gravity reduction techniques, this ice thickness change and the sea level variation will be converted into changes of gravity anomalies at sea level. Stokes’s theory can convert this change of gravity anomalies in changes of the disturbing potential. Keller will use his European contacts to access the ICESAT data, which has been granted tentatively at this stage, and will be confirmed after the Announcement for Opportunity has been made for this mission. CHAMP, GRACE and GOCE: These are three dedicated satellite gravity field missions, operated by Germany, the USA and the EU, which will measure the Earth’s external gravitational field to 13unprecedented levels of precision. This will allow temporal changes in the Earth’s gravity field due to geodynamic and hydrodynamic processes to be observed for the first time. Current simulations of these data (eg Rummel et al., 2001) show that they will be able to resolve the gravity field to a spatial resolution of a few hundreds of kilometres with a temporal resolution of a few months. Keller, by virtue of his association with the GRACE mission, will supply these data. A comparison of the time-variable part of the gravity field from the GRACE/CHAMP/GOCE with icemass balance from ICESAT will separate the ice mass change signature from other sources of the time variability of the gravity field. The relations between the low order spherical harmonic coefficients with the principal moments and tensors of inertia of the Earth will be used to derive variations in the location of the geocentre and the directions of the principle axes of inertia, hence the geocentre and spin axis (cf. Marchenko and Schwintzer, submitted). The cal-val of the synthetic Earth model (Section E4.1) will be achieved using forward and inverse approaches (given below). Importantly, this research project aims to use a combination of both data types for a more reliable and cross-validated prognoses of sea level change. As well as using two (largely independent) data sources, this will be done as a two-step procedure, thus allowing further cross validation. E4.2.1. Forward computation The modelling of the interaction of the changes of polar ice-mass balance with other geodynamical processes (Figure 3) will predict the resulting signatures in the Earth’s gravity field and hence the regional sea level. This is essentially the prediction phase. This is necessary because geodynamical processes interact in very complicated ways. Partially, these processes can be observed directly; partially, these interactions can be mathematically modelled. These will be applied as necessary. Figure 3. System Earth; in black: Some of the geodynamical processes interacting with the change of polar ice mass balance; in dark grey: Observation techniques, which can partially monitor some of these processes; in light grey: Models, which can predict the timely development of geodynamical processes from given observations. In the forward step, observed satellite data will be fed into standard models (i.e., those used routinely by the relevant scientific communities) to predict changes in steady state ocean circulation, sea level change, and changes in the Earth's rotation. This will give justifiable cal-val of the predictions made using the synthetic Earth model. As stated, the non-steric part of the sea level change has two major sources: 141. Melting of the polar ice-shields and the redistribution of the water masses to the oceans. 2. Land height change due to postglacial rebound and depression due to additional water loading. Figure 4 (left) The relationships between sea level change and ocean circulation The isostatic uplift and depression can be predicted from multi-layered viscoelastic Earth models using prehistoric glaciation records and estimates of rheological parameters as input (eg Donato et al., 2000). A secular trend of land uplift and uplift rates at different locations will be deduced from this model. The ice-mass changes detected from ICESAT can be converted into ocean water inflow using standard mass-conservation models. The resulting effect from land uplift/fall and sea level change from water inflow into the ocean basins gives an estimation of the non-steric part of the sea level change. Importantly, the water from the melted polar ice shields will not distribute uniformly over the oceans but will interact with the existing ocean circulation models and alter them (Figure 4). Ocean circulation models can provide this interaction. Therefore, standard models endorsed by the physical oceanographic community will provide this component. Figure 5 (right). Earth rotation changes The mass redistribution will also alter the gravity field and the changed shape of the geoid will influence the steady state ocean circulation. Therefore, synthetic Earth model must detect the mass changes determined by ICESAT and to predict changes in the geoid. These geoid changes will be assimilated in the ocean circulation models. As a result, prediction of the changes in ocean circulations with a spatial resolution of about 1000 km and a time resolution of about a decade can be expected. The mass redistribution due to the melting of the polar ice caps and the resulting postglacial land uplift also modify the principal axis and principal moments of inertia of the Earth. These will strongly affect the behaviour of the Earth's rotation. Using relations between Stokes constants in the spherical harmonic expansion of the gravitational potential and the principal moments of inertia, well known from potential theory, the change in the gravitational potential can be related to the 15changes in the length of the day (LOD). The ice-mass-induced influence on the Earth rotation can be deduced from the synthetic model of the ice mass change. More directly, the viscoelastic Earth model gives indications about the change of the principal moments of inertia. Using standard gyro theory, the change of the moments of inertia can be translated into changes in LOD (Figure 5). Therefore, a fairly precise estimate of the secular rate of change of LOD can be expected as a by-product of these studies. E4.2.2 Inverse computation The observation of the signatures from GRACE/CHAMP/GOCE, ICESAT, tide-gauge, GPS and VLBI observations will be used to estimate uncertainty parameters in the above-mentioned models. This is essentially the cal-val stage. For geodynamical phenomena, such as ocean circulation, long-term tides and Earth rotation, well established observational and mathematical models are available to the geophysical and geodetic community, which will be adopted for this study. The information they provide is located on different time-scale domains: generally, the observations cover the short-term and the models the long-term. Therefore, the observations can be used to calibrate the models and thus give a long-term prediction about the evolution of regional sea level change. Figure 6. Observation of steady-state ocean circulation and sea level change from satellite geodesy The typical observation method for estimating ocean circulation is to measure sea surface heights from satellite radar altimetry (eg Kuhn, 2001; Kuhn et al., 2001; cited in his publication list) and compare these to geoid heights coming from dedicated gravity field satellite missions (Figure 6). The resulting sea surface topography information and its changes can be converted into circulation velocities and velocity changes. This can be compared to the model predicted circulation data in order to improve the parameters of the circulation model used, if necessary. Worldwide Earth rotation observations made by the GPS and VLBI satellite- and space-geodetic techniques, respectively, are collected and processed by the International Earth Rotation Service (IERS). The resulting Earth rotation time-series will be compared with the predicted change rates of LOD from the synthetic Earth model. This comparison may give hints how to improve the rheological parameters in the viscoelastic components of the synthetic Earth model. That is, these parameters will be varied as a function of location so as to give a rotation rate that is consistent with contemporary observations. E4.3 PHASE THREE: Extrapolation of sea level change over time in Australia (6 months) Once the synthetic Earth model has been developed and cal-val’ed using satellite geodetic data, it will be used to run numerous simulations to predict the changes to current sea level around Australasia. These data will then be used for scenario planning in terms of what areas are most vulnerable to flooding (and what areas will be less vulnerable to flooding in the case of sea level fall). These scenarios can then be used by Government and State agencies to anticipate flood defences and auxiliary activities, such as emergency services planning. While we will run the scenarios, the model 16will also be supplied to interested parties so that they can vary the parameters for themselves. This will help open the model to a wide variety of scientific and practical scrutiny. E4.4 Work completed to date As stated, the static component of the synthetic Earth model has been developed and successfully applied to the cal-val of existing gravity field determination techniques. From our literature reviews, we are now confident that we understand the key components that must be adapted in this model. Given that we have constructed the static synthetic Earth model, we anticipate that these adaptations will be relatively straightforward. As stated, access to the GRACE, CHAMP, GOCE and ICESAT datasets will be facilitated via Keller. As such, Keller is a key member of the research team from both the scientific and pragmatic viewpoints. Other key contributors (eg. Kurt Lambeck, Jerry Mitrovica and Glen Milne), while not justifying full investigator status, may be consulted via email and conference attendance during the program if necessary because of their experience in historical modelling of sea level change (see the reference list). E5 NATIONAL BENEFIT Describe the expected outcomes and the likely impact of the proposed research. Describe how the research project might result in economic and/or social benefits for Australia. The national and economical benefits to Australia from this project will be realised in the longer term (say, >10 years). However, sound planning, based on scientific fact, is indisputably valuable. The ultimate benefit is that Australia will be well positioned to plan for changes in the shape of the coastline and changes in the areas more and less likely to flood. This is important because the large majority of the Australian population inhabits coastal and low-lying areas. The social and economic benefits are clear. People will be less likely to be displaced by floods, emergency services will be better located and prepared, and insurance claims (and hopefully premiums) should be reduced or maintained at current levels. Australia will also be in a position to advise our near neighbours with very low-lying topography, notably the many island nations in the Asia Pacific region, on effective prediction and management strategies. This will also reduce Australia’s support needed in times of environmental crisis for our neighbours. E6 COMMUNICATION OF RESULTS Outline plans for communicating the research results. The normal academic avenues of publication in conference proceedings and peer-reviewed journals will be used to disseminate the results of this research program. Importantly, the peer-review process provides some form of quality control and solicits useful feedback on the project; likewise with conference attendance. The Geodesy Group's web (http:/www.cage.curtin.edu.au/~geogrp/) will be used for the rapid dissemination of results. Featherstone has also produced 'popularised' explanations of geodetic concepts for the layperson. One notable example is Featherstone (1996 - The Australian Surveyor), which gives a description of the Geocentric Datum of Australia. This strategy will be replicated to educate the wider population about this program. Ultimately, we would like to publish a paper in one of the high impact, prestigious journals such as Nature or Science. We feel that this is warranted given that studies on the historical aspects of this project have been published in these forums (eg. Chappell, 1983; Milne et al., 2001; Mitrovica et al., 2001). Importantly, the project is sufficiently novel in that it seeks to undertake predictions rather than historical analyses, which should alone justify its reporting in these media. Of course, other highimpact journals such as the Journal of Geophysical Research, Geophysical Journal International, Earth ad Planetary Science Letters, Journal of Geodesy and Journal of Climate Change will also be targeted for reporting the specific details of the techniques used. These journals will be targeted 17because they have reported the historical aspects of sea level change since the last Ice Ages and we wish to communicate our predictive strategies to this same readership. E7 DESCRIPTION OF PERSONNEL Summarise the role, responsibilities and contributions of each Chief Investigator and Partner Investigator. Summarise the roles and levels of involvement of other participants, such as technical staff. Different members of the proposed research team have complementary experience in the different fields required to successfully undertake this project. Accordingly, we believe that we have assembled the most appropriate team of experts to undertake the project. In addition, each member has a good working knowledge of all aspects of the project. Therefore, rather than relying on one person to undertake each aspect, true collaboration will be achieved. Dr Michael Kuhn (PDF) will take most of the intellectual responsibility for the project, duly assisted by Featherstone and Keller. He is an early career researcher (ECR), who is already making a positive impact on and is well regarded in geodesy. He has demonstrated experience in satellite altimetry (eg Kuhn 2001, Kuhn et al. 2001) and sea level change observation (Bosch et al., 2000). Importantly, he was employed as the Research Associate on Featherstone’s ARC large project on synthetic gravity field modelling (Section D2), of which this project is an extension. He is intimately involved with the software and techniques used to construct the existing static synthetic Earth model. As such, it will be very efficient for him to continue to develop and refine this model for the proposed project. Professor Will Featherstone (CI), as well as working on all the functions in Section E4, Will, as CI, will be responsible for the overall management of the program. He has experience of successfully managing ARC projects (having been awarded over $1M in ARC grants since 1993). He has also earned a reputation for delivering results in a timely and professional fashion (eg. AUSGeoid98 from ARC large grant A49331318). His scientific role in this project will be in the development of the new synthetic Earth model (having a base degree in geophysics and planetary physics). As he was instrumental in the intellectual development of this project and the existing synthetic Earth model, he will be able to drive its future refinement. Featherstone will also be involved in the dissemination of the research results to the Australian community via his contacts in academia and government. Professor Wolfgang Keller (PI), as well as providing key data to the project, will bring his extensive mathematical expertise to the project. He has an excellent track record of transferring new techniques from mathematics to solve geodetic problems (as demonstrated by his publication record). Therefore, he will be instrumental in the first two phases of the project, from the mathematical modelling of the synthetic Earth model, to the provision of the calibration data. His contacts with the suppliers of the data in Europe will also be essential in terms of acquisition and determination of any nuances in the data. Research Associates (unnamed as yet): The program will require two full-time ARC Research Associates (incrementing from ARC level A1) to assist the team on most of the aspects in Section E4, to help write and test software, and to run the numerically intensive computations. If Kuhn is not awarded a PDF, he will be appointed to the physical geodesy position. However, we will still need the physical oceanographer (the second RA) to program and run the ocean circulation models. One PhD student may work on projects that are not critical to the ultimate outcome of the program. An Australian Postgraduate Award will be sought for this aspect, should a suitable candidate present him/herself. 1819 E8 REFERENCES (Include a list of all references within the page limit of ten pages.) Bentley, C.R. and J.M. Wahr (1998) Satellite gravity and the mass balance of the Antarctic ice sheet, J Glaciology, 44: 207-213. Cazenave, A. et al (1999) Global-scale interactions between solid Earth and its fluid envelopes at the seasonal time scale, Earth Planet Sci Lett, 171: 549-559. Chappell, J (1983) Evidence for smoothly falling sea level relative to north Queensland, Australia, during the past 6,000 yr, Nature, 302: 406-408. Donato, G.D., L.L.A. Vermeersen & R. Sabadini (2000) Sea-level changes, geoid and gravity anomalies due to Pleistocene deglaciation by means of multi-layered analytical Earth models, Tectonophysics, 320: 409-418 Douglas, B.C. (1995) Global sea level change: determination and interpretation, Rev Geophys Suppl: 1425-1432. Fang, M & Hager B.H. (1999) Postglacial sea level: energy method, Global and Planetary Change, 20: 125-156. Flemming, K., et al. (1998) Refining the eustatic sea-level curve since the Last Glacial Maximum using far- and intermediate-field sites. Earth Planet Sci Lett, 163: 327-342. Gasperini, P., R. Sabadini & D.A. Yuen (1986) Excitation of the Earth’s rotational axis by recent glacial discharges, Geophys Res Lett, 13: 533-536. IPCC (2001) IPCC Third Assessment Report – Climate Change 2001: The Scientific Basis, Intergovernmental Panel on Climate Change. (http://www.ipcc.ch/) James, T.S. & Ervins, E.R. (1997) Global geodetic signatures of the Antarctic Ice Sheet, J Geophys Res, 102: 605-633. Johnston, P. & Lambeck, K. (1999) Postglacial rebound and sea level contributions to changes in the geoid and the Earth’s rotation axis, Geophys J Int, 136: 537-558. Johnston, P., (1993) The effect of spatially non-uniform water loads on predictions of sea level change. Geophys J Int, 11: 615-634. Kaufmann, G. (2002) Predictions of secular geoid changes from Late Pleistocene and Holocene Antarctic ice-ocean amss balance, Geophysical Journal International, 148: 340-347. Lambeck, K (1988) Geophysical Geodesy, Oxford University Press, Oxford. Lambeck, K. & Nakada, M. (1990) Late Pleistocene and Holocene sea-level change along the Australian coast. Palaeogeog. Palaeoclimat. Palaeoecol., 89: 143-176. Marchenko, A.N. and Schwintzer, P (submitted) The Earth’s Tensor of Inertia: Estimation from Recent Gravity Field Solutions, Journal of Geodesy Milne G.A. et al., (2001) Space-geodetic constraints on glacial isostatic adjustment in Fennoscandia, Science, 291: 2381- 2385. Milne, G.A. & Mitrovica, J.X. (1998) Postglacial sea-level change on a rotating Earth, Geophys J Int, 133: 1-19. Milne, G.A. & Mitrovica, J.X. (1998) The influence of a time-dependent ocean-continent geometry on predictions of postglacial sea level change in Australia and New Zealand, Geophys Res Lett, 25: 793-796. Mitrovica, J.X., G A Milne & J L Davies (1991) Glacial isostatic adjustment on a rotating Earth, Geophys J Int, 147: 562- 578. Mitrovica, J.X. & Davis, J.L. (1995) Present-day post-glacial sea level change far from the Late Pleistocene ice sheets: implications for recent analyses of tide gauge records. Geophys Res Lett, 22: 2529-2532. Mitrovica, J.X. & G A Milne (1998) Glaciation-induced perturbations in the Earth’s rotation: a new appraisal, J Geophys Res, 103: 985-1005. Mitrovica, J.X., J L Davies & I I Shapiro (2001) Recent mass balance of the polar ice sheets inferred from patterns of global sea level change, Nature, 409: 1026-1029. Nakada, M. & Lambeck, K. (1987) Glacial rebound and relative sea-level variations: a new appraisal, Geophys J R Astr Soc, 90: 171-224. Peltier, W.R. (1999) Global sea level rise and isostatic adjustment, Global and Planetary Change, 20: 93-123. Rummel R, Balmino G, Johnhannessen J, Visser P, Woodworth P (in press) Dedicated gravity field missions - principles and aims. Journal of Geodynamics. Sabadini, R., Yuen, D.A. & Boschi, E. (1982) Polar wandering and the forced responses of a rotating, multilayered, viscoelastic planet, J. Geophys Res, 87: 2885-2903. Satoi et al. (2001) On the observed annual gravity variation and the effect of sea surface height variations, Phys Earth and Planet Interiors, 123: 45-63. Tamisiea, M.E., et al. (2001) Global geoid and sea level changes due to present-day ice mass fluctuations, J Geophys Res, 106: 30,849-30,865