**5. Conclusions and outlook**

Scope of this chapter was to give an overview on some very powerful and quite recent Remote Sensing possibilities emerged exploiting GNSS observations, which complement the atmospheric and Earth's surface remote sensing traditionally performed by dedicated payloads and instrumentation.

GPS ground receivers can provide valuable and accurate information on integrated precipitable water vapor, considering that single receivers or fairly dense networks are available in many part of the world, providing a quite cheap and reliable source of information. As described previously, many investigations have been carried out in this respect to develop processing techniques, to validate the results through comparisons with independent sources and to exploit the final product. For instance, ZTD or IPWV data from a GPS ground based network can be assimilated into Numerical Weather Prediction models, or integrated with additional sources of IPWV to produce two-dimensional water vapour fields, leading to improved products.

As far as the tomographic approach for the retrieval of Neutral Atmospheric Refractivity maps is concerned, we demonstrated that it is possible (and with a good level of accuracy) as long as some tricks are taken into account. In particular it has to be outlined that, in order to make neutral atmospheric tomography more effective, the choice of the GNSS network topology is a key aspect. A good horizontal receiver's distribution guarantees a good retrieval of horizontal gradients. A good vertical receiver's distribution guarantees also a good retrieval of vertical gradients. Even if our network topology was not optimal for tomographic purposes, the inclusion of measurements (even if not very accurate) performed by two receivers placed at higher heights and of the low elevation observations, demonstrate this aspect. Since a suitable vertical receiver distribution is difficult to implement, the availability of quasi-horizontal observations is necessary. Then, limb sounding Radio Occultation observations are necessary in order to guarantee good observations coming also from low elevation angles (this aspect has already been demonstrated by Foelsche and Kirchengast, 2001 and Notarpietro et al., 2008).

GNSS signals reflected off the Earth surface which represent an error source for navigation purposes, are instead useful for characterizing land and sea surfaces both from a monitoring and early-warning point of view. In particular the possibility of extracting information about the sea height and roughness, the soil moisture content, the snow and ice cover state have been successfully proven. Presently, no operative missions exist but many experimental activities have been carried out and the interest of national space agencies is constantly growing. From our point of view, we put some efforts in developing an instrument capable of collecting reflected GNSS signals, since we believe in the potentialities of this technique.

We definitely believe that the "expansion" of GNSS sources expected when also the European GALILEO, the Indian IRNSS and the Chinese BEIDOU navigation satellite systems will be deployed, together with the consequent availability of Radio Occultation observations, and the consequent availability of "vertical" and "horizontal" observations, will improve definitively all the techniques here presented.
