For over thirty years, single-beam mechanically steered radars have dominated the field of atmospheric observations, and since then, newer improved technologies have emerged that could provide a replacement for aging radars. Phased array radar technology offers meteorologists and scientists a unique opportunity to enhance weather forecasting through rapid electronic adaptive scans. Multiple array geometries exist for phased array radars (i.e., spherical, cylindrical, and planar); however, this work concentrates on enhancing the performance of planar antenna architectures. Planar phased array radar antennas have been under scrutiny due to the challenges posed when trying to satisfy all polarimetric weather requirements met by conventional parabolic dish reflectors (e.g., co-polarized beam mismatch under 0.1 dB, input isolation higher than 40 dB, cross-polarized radiation under -40 dB). This dissertation takes a fresh look into the electromagnetic characteristics of traditional antennas used in planar phased array geometries and provides mathematical insight to prove their performance, limitations, and advantages. The metrics used to evaluate essential performance characteristics were bandwidth, scanning range, polarization, co-polarized beam match, cross-polarization, isolation, and intrinsic cross-polarization (IXR).

The antennas presented in this work (i.e., Horus, Polarimetric Atmospheric Imaging Radar (PAIR), and Horus-ONR) were validated by comparing the results of predictive simulating tools against physical antenna measurements. The Horus antenna was made using aperture coupling feeding technique with stacked microstrip patches. It achieved a fractional bandwidth of 15.4%, a co-polarized beam mismatch of 0.08 dB, and scanned cross-polarization levels of -29 dB, based on Ludwig’s third definition of polarization for θ = ± 45°. The PAIR antenna was made using balanced probe-fed stacked microstrip patches and it totaled fractional bandwidths of 7.7%, co-polarized beam mismatch of 0.21 dB, and -40 dB cross-polarization within the required imaging field of view. Lastly, the Horus-ONR antenna. Its design follows Horus guidelines for manufacturing but improves bandwidths up to 24.8% by trading the scanned co-polarized beam mismatch and cross-polarization required for weather missions. Other antenna architectures proposed for future phased array radar developments are the ultra-low cross-polarization microstrip patch (ULCP-MPA) and a dielectric covered slot antenna (ULCP-DCSA). The ULCP-MPA and the ULCP-DCSA can achieve cross-polarization levels of -40 dB for θ = ± 45°. The antenna designs presented in this dissertation show the lowest scanned cross-polarizations with highly calibratable polarization and might be the best planar radiating elements present in literature so far, despite not achieving all polarimetric weather requirements for multi-function phased array radars. Microstrip patch antennas offer a scalable, low profile solution with excellent polarization diversity and reasonable scanned bandwidths for multi-function, planar phased array radar platforms of the future.