Geodesy and its future.
A contribution to the undergoing discussion within the IAG.
A. Dermanis (*) – F. Sansó
(*) On leave at the Politecnico di Milano
For some years now, geodesists share the feeling that their discipline, despite its great scientific achievements, is at a critical point of its historical development. Decisions or directions for research priorities will play a decisive role for the future outlook, or even for the very existence of Geodesy as a well-defined distinct science.
Geodesy is a science, which is served by a relatively small number of scientists and as such it has a problem of "recognition" among colleagues of related fields. It remains a little known field within the general scientific world and an almost unknown one, as far as the general public is concerned. This fact is creating an "identity problem" which becomes even more important in a changing world where "image making" is playing an important role in maintaining existence and success.
These problems have been of much smaller concern in the past, when geodesy found support from its great engineering partner "Surveying", which provided a secure academic shelter for the development of research and the training of new geodesists. Perhaps the most important source of the present uncertainty within Geodesy stems from the problems related to the changes that surveying, photogrammetry, etc. have undergone in the last years.
As a result of technological progress, traditional surveying practices have been trivialized thus posing a threat to the existence of surveying as a "university level" subject. Even more, in many countries, people without university education has always carried out a large part of the surveying activities. Surveyors had foreseen these developments and responded with a great shift of interest from traditional work to what may be today described as the new field of "geomatics" or "geoinformatics". Geodesists have responded to this shift with mixed feelings, some times enthusiastic, other times with concerns about the relative importance of geodesy in a field where emphasis is sliding from the "geometric" to the "qualitative" element. In such a situation they started to look in other directions and mostly in the family of the earth sciences, where geodesy also has an indisputable place, though not supported from a history of academic coexistence.
Despite these developments, geodetic work itself seems to be much needed, with secure role in future scientific activities. A problem could be that such work might be carried out without reference to geodesy as a distinct science. One hardly finds the word geodesy or geodetic in all kinds of references related to GPS. In fact researchers who do not identify themselves as geodesists, as a result of their former academic training carry out important geodetic work. Indeed the maintenance of an identity as a separate scientific field is not possible without a similar identity within either the academic world or in the eyes of the general public, whenever applications play an important role outside the scientific world itself. Unfortunately the latter does not sufficiently apply to geodesy.
The geodetic identity
At this point one could ask why is it important to maintain such an identity. It may well be in the best interest of society to carry out geodetic work within an interdisciplinary context, without reference to a particular distinct scientific field. Of course it is very unlikely that we as geodesists would subscribe to such a view (even if it were justified) and we would nevertheless continue to strive for. But this is certainly not the case here. The problems of geodesy are of a strategic nature and not of existence, as witnesses by the achievements of our work which have been made possible thanks to the technological advancements of the last years.
Maintaining the geodetic identity is a must, in order to meet the challenges of the future. For this reason it seems worthwhile to reflect on this identity and try to be more specific about what in fact is to be maintained.
Science is one and infinite, but the human mind is small. Historical development led first to the separation of science from philosophy and next to the division of science (natural philosophy!) into distinct scientific disciplines. This separation became inevitable in the face of the growth of scientific knowledge, which made necessary the creation of separate well-defined groups of scientists who carry out a particular part of the whole scientific research and work.
Such a separation did not occur in an arbitrary or even planned way, but it has been the fruit of historical development and it is not trivial to identify the elements that make a scientific discipline distinct and worth of a separate identity within the general framework of scientific activities. These elements we may identify either from an internal point of view, explaining how a discipline evolves and performs on its own, or from an external one, by defining its relation to other scientific disciplines or technological applications.
The internal point view
The internal development of Geodesy is characterized by its object and the methods applied for its development.
The object of Geodesy can be defined in the well-known Helmert’s words: "geodesy is the science of measuring and mapping the surface of the earth". Today we would add: "in relation to an extrinsic reference frame, as well as its variation with respect to time". Contrary to the times of Helmert, the shape of the "geometric" surface of the earth can nowadays be derived by determining very precisely the three-dimensional position of individual points in a world wide frame, connecting them by measurements which have very little to do with the gravity field.
This development raised the question whether geodesy is still the science that we traditionally identify with positioning and gravity field determination, as it was clear in the times of Helmert, because the third dimension could be only accessed by exploiting the gravity field. The fact is that in Helmert’s definition we should use "surfaces of the earth" (which he implicitly included in the term mapping) considering that what we are really interesting in, is not only the geometric surface, but also the physical one, as substantiated by the "geoid" (or telluroid) concept. In fact the geometric surface is the seat of a large number of natural phenomena, many of which depend upon the gravity field, and of human activities. Many of these change the shape of the geometric surface of the earth, without affecting significantly the gravity field. This is why knowing both the geometric surface and its physical counterpart provides an indispensable information to understand the real kinematics of the earth by separating its different causes.
Summarizing, we note that the interplay between gravity and geometry maintains its significance, to the extent that we could even arrive to the point of defining geodesy as the "science of determining heights on the earth".
Coming to the methods we have to distinguish between observation methods and methods of analysis, as well as, the reference frame concept bridging between the two. The importance of technological advances which have led the transition from traditional geodetic measurements (theodolites, levels gravimeters, etc.) to modern space techniques (GPS, Glonass, DORIS, PRARE, SLR, LLR, VLBI, altimetry, SAR interferometry, inertial systems, etc.) hardly needs to be emphasized. We can foresee some of the forthcoming developments (GPS without phase ambiguities, double-frequency altimetry and SAR, integrated geodetic surveying apparatus, kinematic gravimeters, SST and satellite gradiometry), but the past shows that there will also be pleasant surprises in the future.
The methods of data analysis are of course a driving force in the development of a science and for sure geodesy has succeeded in creating its own theoretical arsenal. A distinctive characteristic of geodetic analysis has always been to concentrate on the measurement process trying to model it as directly as possible. The combination of deterministic and stochastic variables (disturbances) has always being waving forth and back in the observation equations, raising many discussions in the geodetic community. The necessity of refining the physical model behind observations has always been driven by the accuracy and resolution of the measurements themselves, in an attitude different from a general physical approach where the theoretical model development is the main concern while observations are just a means of confirmation. In this sense geodesy is an applied science.
Driven by the tremendous improvement in the accuracy of measurements, we have been forced to introduce more and more unknown parameters in our models ending with an infinite dimensional theory of observational equations, where the traditional overdetermined nature of the problem is replaced with an underdetermined one. This has led us to the realm of signal analysis where we have to formulate our peculiar theory to take into account that most of the times we deal with spherical signals harmonic in space.
This wider point of view, together with forthcoming data acquisition systems is pushing geodetic theory to a solution of the problem of describing its object in a unified way from the global down to the local level. In this huge effort, which is also a big challenge from the numerical point of view, it is peculiar to geodetic methodology to try to attack the same problem by continuous or discrete models, switching from one to the other according to the specific task.
In all its historical development, the theory of positioning and determination of the earth surface and of gravity field never split apart and always benefited from each other producing a unique and lively approach. And it is a common belief that it will continue in this way.
The reference frame is more than simply one of the "objects" of geodesy, but it rather belongs to the realm of "methods" bridging between analysis and observations. In writing our observation equations we choose certain systems of coordinates in order to turn positions (as well as vectors) into numbers. But this cannot be done without choosing a physical reference system (e.g. an inertial one) to which the coordinate system is attached, together with all other coordinate systems obtained from it by a purely mathematical transformation. The reference system is physical in the sense that the physical laws, which govern the observation equations, hold in that particular form only in that system. It is fixed in a theoretical way by making a choice, which eliminates the rank deficiency in the whole system of observation equations describing all the data available in principle. It is materialized by choosing a particular set of actual observation equations, for instance a specific set of observations referring to a specific set of permanent geodetic stations (e.g. ITRF versus ITRS in IERS terminology).
It is important to underline that, in the present satellite era, we are living the unique experience that a real global reference frame can be realized. Thus all the traditional work of geodetic services in establishing control networks can be seen as a spread of the materialization of the global reference frame all over the world down to the local level.
From the internal point of view, the problem of determining earth rotation is identical to the problem of defining the reference frame in space-time. And this is so because the "earth rotation" determined in geodesy is in fact the rotation of the reference frame as materialized by a particular choice of, more or less, globally distributed permanent stations. A different choice of this materialized frame leads to a different earth rotation.
The external point of view
Geodesy, being a part of science as a whole, can also be identified through the relations it maintains with all other branches of science as well as technology. While examining these relations let us keep in mind that all "boundaries" are artificial and the question of assigning a specific object to one field or the other makes no sense; what we want to examine is rather the areas of common interest.
From a historical point of view we could say that geodesy has separated as an autonomous field from physics first and then from astronomy. Therefore it is quite obvious that it maintains important relations with these, for instance in the field of relativity or in that of celestial of mechanics. The relation between geodesy and mathematics and statistics hardly needs to be emphasized. Let us underline that geodesy not only has and is using a wide spread of mathematical and statistical methods, but it has and it is contributing to these sciences in specific items. Examples are tensor calculus, numerical analysis, statistical testing theory, generalized inverses, boundary value problems, stochastic approximation theory, integer estimation, etc.
Turning to fields, which are closer to geodesy in everyday life, let us look into earth sciences and engineering.
With geophysics of the solid earth we share the interest in reference systems and earth rotation, in plate tectonics, and more generally in the dynamics of earth deformation. Gravity determination and interpretation is also a fundamental tool to build integrated solid earth models.
With oceanography we share the analysis of altimetric data for the determination of the stationary surface of the ocean related to geostrophic circulation, as well as oceanic variability. Tides and height datums are also a shared scientific challenge. Determination of bathymetry is a common item too.
In climatology and environmental sciences we can contribute in important ways through the study of atmospheric soundings, ice cover and sea-level changes.
Apart from its relations with geosciences, geodesy has always formed an important basis for various engineering applications.
Navigation, as real-time kinematic positioning is naturally almost a part of geodesy, although its great importance for the general public makes it a subject in itself within an engineering environment. The shared used of GPS technology, as well as INS, gives the opportunity to strengthen and further exploit this traditional relation.
The application of geodetic methods of observation and analysis at a local level has always been considered as a child of geodesy going to the direction of cartography. Along this direction we go further into the problem of digitalization and manipulation of geographic information, ending again into the realm of surface representations, such as the creation of digital terrain models, as well as the representation of more general forms of quantitative and qualitative spatial data (like pollution, traffic load, land use, etc.).
We should not overlook the engineering applications of geodesy itself, such as the establishment of special purpose control networks at various scales down to even industrial applications, which we traditionally call surveying.
An interesting remark about a part of these disciplines, namely photogrammetry and remote sensing is that, although they appear to drift apart from geodesy as a result of technological developments, in fact they are incurring in the same type of problems from the methodological point of view. The interplay between discrete-continuous and deterministic-stochastic point of view in model development, is currently a common great challenge in both geodesy and these fields and a lot could be gained by sharing experiences and inspirations.
In any way geodesy provides the natural theoretical background for a new type of engineer devoted to the assessment, handling and representation of spatial information, which we now call geomatics or geoinformatics.
We have examined above the items which are of common interest between geodesy and other disciplines, to which geodesy can contribute significantly. This is for the advantage of science as a whole, and it comes from geodesy because geodesy is able to maintain an internal identity and methodological coherence. This is possible because objects and methods identified as the internal defining factors of geodesy are capable of providing an autonomous image of reality, free from hypotheses and assumptions coming from other fields. Of course having a deeper understanding of geophysical mechanisms influencing e.g. rotation or ocean dynamics, facilitates the process of setting up observational models. Nevertheless, the point is that geodesy could make it without them and this is providing a fully autonomous contribution to other earth sciences, as well as a means of independent testing of relevant hypotheses, such that the theory of plate tectonics.
For this reason geodesy is useful by itself and should not be drown into other geosciences.
In concluding we just concentrate on three important points:
The idea of giving a focus to all our activities is probably a good one. Certainly the idea that geodesy is providing the global reference frame for positioning and gravity field determination is a good banner for promoting our field. This will put geodesy in a unique and central position with respect to all other disciplines which need positions, gravity field and earth surface geometry.
While focusing however we should not forget that one of the reasons for having a unique global reference system is exactly to provide reference for the extensive activities carried out at a local level. Of course many of these activities are of an applied or even routine nature and we do not mean that any professional performing it is automatically a geodesist. However we do claim that an applied activity which increases our material knowledge of the geodetic aspects of the earth, carried out with a clear understanding of the geodetic theory behind it has to be accepted as a member of our family in the wide sense.
Many excellent colleagues from either geodesy or nearby disciplines have drifted away, because too many times we have put up barriers and boundaries and they have succeeded in fields, which we do not recognize anymore as of geodetic interest. We should go back and conquer what is lost, namely a good partnership with colleagues who after all are working for the determination of the earth surface (e.g. Photogrammetry, SAR techniques, etc.).
The point here is that this can be done to a mutual benefit, because although applications may seem diverse, we can meet at a higher level, exchanging theories and methodologies.
By reasoning on the general directions of geodesy one is immediately led to the problem of a new IAG structure in such a way that its tasks could be performed in the best way; and in fact this is the duty of the Review Committee. Without going into any specific recommendation we try here just to make some rather general suggestions:
The structure reflecting the new "focused" configuration should nevertheless be very flexible, backing any good spontaneous initiative, giving to young scientists an opportunity to contribute to IAG work in a fast way without bureaucratic obstacles.
Furthermore a place in IAG should be also given to people doing applied work in a critical and experience gaining way. As an applied science geodesy should also include the activities contributing to the knowledge of our object. In this sense we believe that services can do a very good job.
Researchers from affine sciences should find in IAG an open structure, where they can come and work with us, even on a temporary basis.
Hasta la geodesia siempre !