Precipitation measurements with polarimetric radio occultations

Abstract

In 2009, the Spanish Ministry of Science and Innovation approved a proposal to modify the Global Positioning System (GPS) receiver and to allocate a Polarimetric (Pol) Radio Occultation (RO) antenna in the Spanish PAZ satellite. PAZ became an opportunity to test the new Pol-RO concept, which aims to capture ROs using a two orthogonal linear polarization antenna. The experiment has been named Radio Occultations and Heavy Precipitation with PAZ (ROHP-PAZ). The objective is to measure the phase difference between the horizontal and the vertical components of the incoming electromagnetic field that is induced when heavy precipitation flattened raindrops are present in the ray-path. This effect, widely studied in weather radar community, will be measured from space using GNSS signals for the first time with PAZ, which is planned to be launched in 2017. The main objective of this new concept is to enhance the RO capabilities by providing vertical precipitation information along with the current standard RO thermodynamic products (i.e. temperature, pressure and moisture). Until now, no other observing system has been able to provide simultaneous thermodynamic and precipitation information under extreme conditions. The high vertical resolution, global coverage and all-weather capability properties of the RO observations combined with vertical indication of precipitation intensity can be of great value for heavy rain characterization, and therefore for climate and weather forecast and research. The theoretical background for the technique, its feasibility and applications have been assessed in this dissertation. The theoretical basis has been developed combining electromagnetic propagation theory and cloud and precipitation microphysics. Forward scattering simulations at L-band have been obtained in order to relate the microphysics parameters with the expected Pol-RO observables. The feasibility has been addressed using coincident (in space and time) RO profiles and space-based precipitation observations. Such simultaneous observations allow for the characterization of actual RO measurements according to the coincident precipitation information. Finally, the applications have been investigated through realistic end-to-end simulations of the Pol-RO observations, which provide the anticipated Pol-RO products for different precipitation situations, regions, and seasons. Before the launch of the satellite, a field campaign has been conducted with the aim of starting the characterization of the polarimetric measurements. The engineering model of the PAZ antenna was placed at the top of a mountain peak in order to capture, for the first time, linear polarimetric GNSS signals at low grazing angle. This campaign has been useful to start identifying the hardware internal effects and unexpected precipitation features that will be affecting the Pol-RO observations. These effects have been incorporated to the simulations, hence providing valuable feedback to obtain more realistic Pol-RO products. Besides feedback, the data from the field campaign have shown the first observational evidence that precipitation and other hydrometeors induce a noticeable effect on the GNSS polarimetric signals. All these exercises yielded several relevant results. The noise level analysis from actual RO observations sensing precipitation scenarios has allowed to set a detectability threshold for the technique, indicating that a high percentage of moderate to heavy precipitation events will be detected with PAZ. Nevertheless, the integrated nature of the Pol-RO observable does not allow to distinguish between the contributions from the rain's intensity and extension, leaving an ambiguity in the provided product. In an attempt to solve such ambiguity, a tomographic approach has been proposed, which has yielded promising theoretical results. Moreover, it has been shown how the Pol-RO observables can be linked to physical precipitation parameters, such as the along-ray averaged rain rate, in a probabilistic way. The end-to-end simulation has also revealed that the ionosphere will induce a non-negligible depolarization, that will require calibration. Finally, the collocated data has shown the potential applications for Pol-ROs products.

Obtenir mesures simultànies de l’estat termodinàmic de l’atmosfera i de precipitació ha esdevingut un repte per la comunitat científica. Les missions espacials dedicades a obtenir perfils termodinàmics de l’atmosfera tenen problemes amb la presència de núvols gruixuts, ja que el medi esdevé opac a la radiació infraroja (que és la banda de l’espectre electromagnètic en la qual operen). Alternativament, es poden utilitzar radiosondes. Les radiosondes obtenen perfils termodinàmics de l’atmosfera amb molt alta resolució vertical, però tenen l’inconvenient que el seu llançament necessita certa infraestructura, i per tant les zones més remotes en queden al marge. Això inclou pràcticament la totalitat dels mars i oceans, i moltes zones sub-desenvolupades. Per tant, moltes de les zones amb precipitació extrema no poden ser caracteritzades amb aquesta tècnica. A més a més, la resolució temporal acostuma a ser molt baixa, ja que no se’n poden llençar moltes al dia degut a l’elevat cost econòmic que suposaria. Per altra banda, els radars que mesuren les estructures en tres dimensions de la precipitació no tenen la capacitat d’obtenir perfils de temperatura o pressió. Les estacions meteorològiques, que poden ser molt nombroses en segons quins territoris, estan limitades a mesures en superfície, i altra vegada, mars, oceans i regions sub-desenvolupades en queden al marge. Amb tot, les Radio Ocultacions Polarimètriques emergeixen com una tècnica a tenir en compte a l’hora de caracteritzar precipitació extrema. La seva cobertura global, alta resolució vertical i la capacitat de penetrar en núvols i precipitació la fa una tècnica molt atractiva en aquest sentit. Cada cop més estudis científics coincideixen en apuntar un augment en la freqüència d’aquests fenòmens extrems, i una caracterització acurada és necessària per millorar els models de predicció.