The feasibility study

A precipitationdetectionandmeasurement experimentis plannedfor theSIR-C/X-SAR mission. This studywasconductedto determineunderwhatconditionsanoff-nadir experiment is feasible.We investigatedthesignal-to-clutteratio, thesignal-to-noiseratio,andtheminimum detectablerain rate. Availablemodels,usedin previousstudies,wereusedfor thesurfaceclutter and the rain echo. The study also considersthe attenuationof the returnsat X band. We concludethatanoff-nadir rain-measurement experimentis feasibleonly for rainrates> 10mrn/hr for look angles> 60°. For therangeof look angles5°< 0_< 50°, therain raterequiredis very high for adequatesignal-to-clutterratio, andhencethe feasibility of theexperiment. 1.0 INTRODUCTION Thedetectionand measurement of precipitationfrom spaceplatformswill be important in a wide spectrumof meteorologicalproblems. A precipitationdetectionand measurement experimentwasacceptedfor the SIR-C/X-SARmission. Although climatologicalapplications requireobservingthe rainfall overa long periodof time, theresultsobtainedfrom this proof-of-concept experiment will helpdemonstratesynthetic-aperture radar(SAR)capabilitiesfor short-termrain measurement andallow partial verification of therain-retrievalalgorithm. The resultsmayalsobeusefulfor meteorological research, but this is doubtful in view of the limited time for the experiment. We expectthatthe SIR-C/X-SAR sensorscanmeasurerain quantitativelyfor moderate to intenserain rates( > 10mm/hr), at high off-vertical angles. It is proposed,if possible,to comparethe observedSIR-C/X-SAR radarbackscatter power with in situ data from calibrated ground-basedradars. Rain-rate estimation from backscattersignal strength is the simplest lnethod for precipitation detection. However,the backscattermethodcanunderestimatethetrue rain rate (Goldhirsh1988)whenusedwithout correctingfor attenuation.SincetheX-SAR operatesat 9.7 GHz, whereattenuationcancauseproblems,correctionfactorsarenecessary. The report outlinesa studyto determinethe feasibility of rain retrievalusingthe X-SAR at off-vertical angles of incidence. The lower magnitude of backscatter at C band makes this frequency impractical. 2.0 GEOMETRY For vertical incidence and a narrow transmitted pulse, the backscatte,ed power from the hvdrometeors arrives at the receiver prior to the large surface contribution. At non-vertical ,.) viewing angles, however, the surface return arrives simultaneously with the desired echo ( see Fig. la). Thus, the surface clutter can mask the returns due to the rain at the surface. The geometry shown in Fig. lb is for non-vertical viewing angles making the plane-earth assumption. From elementary geometry we can compute the dimensions of the observed volume as a function of the look angle. The beam below the cloud top is divided into range cells corresponding to a slant-range resolution of 15 m. The swath R_ on the ground for vertical beamwidth _v and satellite height h is R s = h [ tan(0 I) tan(0)] (I) where