Weather radars are increasingly used for the study of clouds, understanding the precipitation systems and also for forecasting very short range weather (one hour to a few hours). Now, Doppler Weather Radar (DWR) data are available in India and it is possible to study cloud properties at fine temporal and spatial scales. Radar is a complex system and calibration of a radar is not an easy job. But derived cloud properties strongly depend on the absolute magnitude of the reflectivity. Therefore, there is a need to check how data from two or more radars compare if they measure a common volume. Chennai and SHAR radars are within 66 km from each other, and the data collected during their calibration and intercomparison experiment in 2006 enables the comparison of their reflectivity(Z) values. Individual reflectivity are compared after plotting SHAR versus Chennai in a scatter plot. Fitting a least square linear best fit line shows that the intercept has a value around 6 dBZ and the slope of the line is 1.06. Thus, there is a trend as well, and the difference between the two radars increase with Z, and for Z around 40 dBZ (for SHAR DWR), the difference between the two is around 8.5 dBZ. Visual intercomparison also validated the results. Data from the two radars are compared with Precipitation Radar (PR) data on board TRMM satellite. TRMM radar slightly overestimates compared to Chennai radar above the range of 30 dBZ. After standardized, SHAR data is used for understanding the evolution and propagation of cloud systems. The diurnal variation in convection is strong in the study region, with increase around local evening and morning and weakening around midnight except in December. Average liquid water content in the clouds is about 0.5 gm/m3. There is some seasonal dependence but no clear dependence on cloud size. Smaller systems of May have more liquid water content compared to larger ones. For nowcasting vertically projected maximum reflectivity is taken. A threshold of 30 dBZ is set to identify the cloud systems. Both center of gravity tracking (CG) and cross-correlation (CC) methods are used to track them. Frequent merging and splitting is common in the clouds which makes storm tracking difficult. Tracking by CC is giving better result than that by the CG method in the case of large systems (i.e., clusters). For smaller systems (individual cloud systems), CC method gives better result than CG method but not as good as cluster.

A Bacia do Alto Tiete abriga cerca de 50% dos habitantes do Estado de São Paulo e é afetada freqüentemente por eventos de inundações. Uma das principais fontes de problemas é a alta impermeabilização devida à ocupação da superfície nas últimas décadas. Um dos seus tributários secundários, a bacia do Rio Pirajuçara se insere neste contexto e sofre com problemas da mesma natureza. A modelagem hidrológica permite uma análise do escoamento superficial nestes ambientes e é útil na previsão de vazões por meio de redes telemétricas e sensoriamento remoto. Entretanto, redes telemétricas apresentam problemas de representatividade espacial e exposição, radares meteorológicos, apesar da maior resolução espaço-temporal das estimativas de precipitação, possuem várias fontes de erros e incertezas. A principal delas se refere à relação ZR. Deste modo, a integração dessas medições e estimativas pode minimizar erros de ambas. O objetivo deste estudo é analisar aspectos hidrológicos da Bacia do Rio Pirajuçara por meio do modelo TOPMODEL com medições de vazão e precipitação disponíveis para 18 eventos monitorados entre outubro de 2008 a outubro de 2009. O modelo TOPMODEL foi calibrado com dez eventos e verificado com os demais. A calibração foi realizada com os dados da telemetria da Bacia do Alto Tietê, radar meteorológico de São Paulo e a combinação de ambos por meio da análise objetiva estatística. Os resultados da calibração indicam que o melhor desempenho foi obtido com radar meteorológico, com número de NASH de 0,51, menor erro quadrático médio e menor viés médio absoluto. A verificação também indicou o mesmo resultado com número de NASH de 0,69. As simulações indicam que apesar da utilização da precipitação média, o modelo TOPMODEL simulou adequadamente cerca de 75% das vazões de alerta. O trabalho evidencia as limitações da telemetria e seus impactos na integração com os dados do radar. / The Alto Tiete watershed is home for about 50% of the inhabitants of São Paulo State and is affected by recurrent flashfloods. One major source of difficulties is the high rate of soil impermeabilization caused by dense surface occupation in the last decades. One of its secondary tributaries, the Pirajussara watershed suffers with similar problems. Hydrological modeling allows the analysis of runoff and other variables in these basins. It also useful for streamflow forecast based on telemetric networks and remote sensing measurements. However, surface networks lack spatial representativity and exposure is a also a issue, weather radars, in spite of their much higher spatial and temporal resolution rainfall estimation, are affect by several sources of errors and uncertainties; the most significant one being the ZR relationship. Thus, the integration of these measurements and estimates can minimize errors of both. The goal of the present work is to analyze the surface hydrology of the Pirajussara watershed based on the TOPMODEL, streamflow and rainfall measurements available for eighteen events between October 2008 and October 2009. The TOPMODEL was calibrated with ten events and verified with the remaining events. The calibration was performed with the Alto Tiete telemetric measurements of streamflow and rainfall only, the São Paulo weather radar (SPWR) only and a combination of both through the statistical analysis scheme. Calibration results show a better performance for the SPWR with a NASH number of 0.51, least SME and mean bias. On the other hand, the verification also indicated better results for the SPWR with a NASH number of 0.69. The simulations indicate that in spite of the use of the mean rainfall over the watershed, the TOPMODEL performed adequately for 75% of the streamflow alerts. It is also evident the limitation of the available network and its impacts on the integration to the SPWR rainfall data.

La question de la prévision des évènements fortement précipitants se produisant sur le bassin Méditerranéen est au coeur du programme international HyMeX (Hydrological cycle in Mediterranean EXperiment, http://www.hymex.org/) dont l'un des objectifs est d'améliorer la prévision et la prévention des risques hydrométéorologiques du bassin méditerranéen dans le contexte du changement climatique. Durant l'automne 2012, une campagne de mesures de deux mois dite "Période d'Observation Spéciale" (SOP1) a été menée afin de documenter les conditions propices à la formation et au développement des évènements convectifs de type cévenol souvent responsables de crues dévastatrices. Pendant cette SOP1 un dispositif instrumental sans précédent a été déployé avec notamment pour la première fois sur le sol français un imageur à haute résolution spatio-temporelle permettant d'observer les décharges électriques en trois dimensions : le Lightning Mapping Array (LMA). Cet instrument a été combiné aux radars du réseau ARAMIS de Météo-France, et plus précisément aux radars Doppler à diversité de polarisation dans le Sud-Est de la France qui offrent la possibilité d'obtenir des informations sur le type et la distribution des hydrométéores au sein des systèmes précipitants. La production d'éclairs étant le résultat d'une électrisation issue des interactions microphysiques (collisions entre graupels et cristaux de glace en suspension), une description détaillée des différents types d'hydrométéores présents dans les nuages convectifs est essentielle. Dans cette optique, les algorithmes d'identification des hydrométéores développés par Météo-France ont été évalués puis améliorés. Les résultats de cette étude ont montré que les restitutions entre les différents radars étaient plutôt cohérentes, à condition que l'information sur l'altitude de l'isotherme 0°C soit correcte. Ce travail a ensuite été complété par la création, via une méthode originale, de composites 3D d'hydrométéores permettant de décrire la microphysique majoritairement présente dans les systèmes convectifs observés pendant la campagne HyMeX. La deuxième partie de ce travail s'est basée sur l'exploitation de la synergie radar-LMA sur une ligne de grain observée durant la SOP1 de HyMeX. Les principales informations déduites de ce couplage ont mis en exergue l'importance des processus microphysiques intervenant dans l'électrisation du nuage d'orage, ainsi que l'impact du relief sur l'activité électrique globale du système convectif. Sur les quatre heures de données analysées du 24 Septembre 2012, le déclenchement et la propagation des éclairs ont majoritairement été observés dans les espèces microphysiques que sont le graupel, les cristaux de glace et dans une moindre mesure la grêle. Cette étude souligne également le rôle important de la topographie sur l'activité électrique et montre que le passage d'un faible relief peut dramatiquement influencer la distribution et l'intensité des éclairs dans les régions convectives. / The Hydrological cycle in Mediterranean Experiment (HyMeX, http://www.hymex.org/) is a 10-year research program focusing on the quantification and understanding of the water cycle in the Mediterranean at various time and spatial scales with particular emphasis on high-impact weather events. This study takes place within the framework of the first HyMeX field phase (HyMeX-SOP1), which was conducted in the autumn 2012. The unique and extensive dataset collected during this field campaign offers the possibility to further investigate the complex relationships between cloud microphysics and lightning at play within mesoscale convective systems observed in southern France. With this regard, the present study make the use of a Lightning Mapping Array (LMA) along with operational dual-polarization weather radar. The first instrument allows documenting the three-dimensional lightning activity, whereas the second has the ability to determine the type of hydrometeors within cloud systems. Since the production of lightning is the result of an electrification created by microphysical collisions between graupels and ice crystals in suspension, a highly detailed description of hydrometeor types within convective clouds is needed. With this respect, an improved version of Météo-France hydrometeor classification algorithm was developed and evaluated so as to be able to discriminate between a large number of microphysics species. Overall hydrometeor species retrieved from a pair of neighbouring radars within a common sampling area are consistent from one to another. This study has however pointed out the need to check the consistency related to the identification of 0°C isotherm derived from numerical weather prediction model outputs before to perform hydrometeor identification. As a follow up to this work, a novel interpolation method allowing the remapping of single-radar hydrometeor fields onto a common Cartesian grid was developed in order to get access the three-dimensional hydrometeor distribution within HyMeX convective systems. Another part of this work aims at combining LMA and polarimetric radar observations to infer relationships between the total lightning activity, microphysics, and kinematics within the intense bow-echo system observed above the complex terrain of southern France during HyMeX. Using the synergy between LMA and polarimetric radar data, it is underlined that microphysical processes involved in cloud electrification, along with the impact of the topography play at part onto the global lightning activity. Based on a 4h analysis on the 24 Setptember 2012, it is found that lighting initiation and propagation take preferentially place within graupel, ice and to a lesser extent hail regions. This study also highlights the important role of topography on lightning activity and shows that even a small hill can dramatically influence the distribution and intensity of lightning within convective areas.

Over the last two decades, dual-polarimetric weather radar has proven to be a valuable instrument providing critical precipitation information through remote sensing of the atmosphere. Modern weather radar systems operate with high sampling rates and long dwell times on targets. Often only limited target information is desired, leading to a pertinent question: could lesser samples have been acquired in the first place? Recently, a revolutionary sampling paradigm – compressed sensing (CS) – has emerged, which asserts that it is possible to recover signals from fewer samples or measurements than traditional methods require without degrading the accuracy of target information. CS methods have recently been applied to point target radars and imaging radars, resulting in hardware simplification advantages, enhanced resolution, and reduction in data processing overheads. But CS applications for volumetric radar targets such as precipitation remain relatively unexamined. This research investigates the potential applications of CS to radar remote sensing of precipitation. In general, weather echoes may not be sparse in space-time or frequency domain. Therefore, CS techniques developed for point targets, such as in aircraft surveillance radar, are not directly applicable to weather radars. However, precipitation samples are highly correlated both spatially and temporally. We, therefore, adopt latest advances in matrix completion algorithms to demonstrate the sparse sensing of weather echoes. Several extensions of this approach are then considered to develop a more general CS-based weather radar processing algorithms in presence of noise, ground clutter and dual-polarimetric data. Finally, a super-resolution approach is presented for the spectral recovery of an undersampled signal when certain frequency information is known.

This thesis examines the role of rainfall variability and uncertainties on the spatial scaling structure of peak flows using the Whitewater River basin in Kansas, and Iowa River basin in Iowa as illustrations. We illustrate why considering individual hydrographs at the outlet of a basin can lead to misleading interpretations of the effects of rainfall variability. The variability of rainfall is characterized in terms of storm intensity, duration, advection velocity, zero-rain intermittency, variance and spatial correlation structure. We begin with the simple scenario of a basin receiving spatially uniform rainfall of varying intensities and durations, and advection velocities. We then use a realistic space-time rainfall field obtained from a popular rainfall model that can reproduce desired storm variability and spatial structure. We employ a recent formulation of flow velocity for a network of channels and calculate peak flow scaling exponents, which are then compared to the scaling exponent of the channel network width function maxima. The study then investigates the role of hillslope characteristics on the peak flow scaling structure. The basin response at the smaller scales is driven by the rainfall intensities (and spatial variability), while the larger scale response is dominated by the rainfall volume as the river network aggregates the variability at the smaller scales. The results obtained from simulation scenarios can be used to make rigorous interpretations of the peak flow scaling structure obtained from actual space-time model, and actual radar-rainfall events measured by the NEXRAD weather radar network. An ensemble of probable rainfall fields conditioned on the given radar-rainfall field is then generated using a radar-rainfall error model and probable rainfall generator. The statistical structure of ensemble fields is then compared with that of given radar-rainfall field to quantify the impact of radar-rainfall errors on 1) spatial characterization of the rainfall events and 2) scaling structure of the peak flows. The effect of radar-rainfall errors is to introduce spurious correlations in the radar-rainfall fields, particularly at the smaller scales. However, preliminary results indicated that the radar-rainfall errors do not significantly affect the peak flow scaling exponents.

The Shared Mobile Atmospheric Research and Teaching Radar (SMART-R) is a truck-mounted C-band, Doppler radar that was deployed during the Dynamics of the Madden-Julian Oscillation (DYNAMO) / Cooperative Indian Ocean Experiment on interseasonal variability in the year 2011 (CINDY2011) campaign on Addu Atoll, Maldives. One of SMART-R’s objectives was to provide continuous volume scans of precipitating clouds during all phases of the Madden-Julian Oscillation (MJO) for the full duration of the campaign. Data from SMART-R is available for 2 October 2011 through 9 February 2012. Every 10 minutes a full volume scan was produced, which was subsequently run through quality control algorithms that, among other filters, performed a calibration correction, noise filtering, and an attenuation correction. It was observed that data from SMART-R appeared to be slanted towards the WNW, and after analysis, a 0.75◦ tilt correction was applied towards azimuth 285◦. The data was then converted into Cartesian coordinates and an additional noise filter was applied. NETCDF files with radial velocities and corrected reflectivity were produced. From the reflectivity observations, a suite of products including rain maps, echo- top heights and convective/stratiform separations were produced. A modified version of the convective/stratiform separation was developed in an attempt to classify shallow and weak convection more correctly. The modified algorithm utilizes an isolation parameter set to 10 km to the north, south, east, and west, a 10-dBz echo-top height threshold set to 9 km, and a 16-dBz reflectivity threshold at 3 km to ensure only isolated, shallow, and weak rain originally classified as stratiform, is reclassified as convection. Analyses of these products clearly suggest two MJO events occurring in October and November as indicated by the Wheeler and Hendon Multivariate MJO index. While stratiform rain almost always encompassed a larger area of the radar domain, convective rain was the larger producer of rain with the exception of active MJO periods. In addition, echo-top height counts are observed to increase in both vertical structure and frequency as the MJO initiates and becomes active over the radar domain. Possible connections are also made between echo-top height data and humidity retrievals from soundings launched on Addu Atoll. It appears that during MJO initiation, convective echo tops lead the moistening of the mid troposphere, while during suppressed phases, the convective echo tops lag behind the moistening of the mid troposphere. Wind shear also appears to be weaker during an active MJO event, and increase as the active MJO exits the region. From these observations, as well as other rain statistics including the diurnal cycle, indicators for a localized MJO index are proposed that are based on local radar and sounding data, rather than satellite and reanalysis observations of wind and outgoing long-wave radiation.

In this cumulative dissertation, different features and methods are presented to assess and process multi-sensor derived radar data for climatological analysis. The overall objectives were to appraise the limitations of an hourly radar-based quantitative precipitation estimate (QPE) product and to develop and apply reasonable approaches to process these data. Hence the spatial and temporal limitations of radar-derived precipitation rates are discussed in the context of climatological applications, and two types of climatologies are obtained, first a climatology of daily precipitation fields and second a long term precipitation climatology. These relate to questions concerning the methodologies rather than climatological significance or assessment of precipitation and its role in the water balance. Current radar data availability limits such a hydro-climatic analysis. The thesis consists of three peer-reviewed publications. All investigations in this thesis are based on the RADOLAN rw-product of the German Weather Service (DWD) for an extended study region including the Free State of Saxony, Germany, for the period from April 2004 to November 2011. The first publication is dedicated to the classification of daily precipitation fields by unsupervised neural networks. In the presented work, the quality of the radar-derived precipitation rates is analysed by a temporal comparison between recording and non-recording gauges and the corresponding pixels of the RADOLAN rw-product on hourly and daily bases. The analysis shows that a temporal aggregation of the original product should be limited to a temporal scale up to 24 h because of the processing algorithms and the reappearance of previously suppressed errors. Nevertheless, an unsupervised neural network was successfully used for the classification of daily patterns. The derived daily precipitation classes and corresponding precipitation patterns could be assigned to properties of the associated weather patterns and seasonal dependencies. Hence, it could be shown that the classified patterns not only occurred by chance but by statistically proven properties of the atmosphere and of the season. The second publication is primarily concerned with two tasks: first, the pixel-wise fitting of mixture distributions on the bases of the obtained patterns from the first publication, and second, the analysis of spatial consistency of the radar-derived precipitation data set. The fitted parametric distribution functions were analysed in terms of Akaike\'s information criterion and the Kolmogorov-Smirnov test. These benchmarks showed, that the performances are best for mixture distributions derived by an initial classification by an unsupervised neural network and cluster analysis, and by gamma distributions. These results underline the significance of the derived precipitation classes obtained in the first publication. Furthermore, the Kolmogorov-Smirnov test indicates that independent of the distribution function, the radar-derived daily precipitation rates under the assumption of the deployed parametric distribution function has the best or most natural order of precipitation rates at spatial scales from 2 to 4 km for daily precipitation fields. Thus, it is recommended to use the original radar product at these scales rather than at 1 km resolution for daily precipitation sums. In the last publication, the focus shifts from daily to long-term precipitation climatology. The work introduces a rapid and simple approach for processing radar-derived precipitation rates for long-term climatologies. The method could successfully be applied to the radar-derived precipitation rates by excluding or correcting the errors that reappear due to temporal aggregation. Despite the fact that the approach is empirical, the introduced parameters could almost be objectively derived by means of simulation and optimisation. This could be achieved by utilising the reasonable relationship between elevation and precipitation rates for longer periods. Finally, the obtained results are compared to two independently derived precipitation data sets. The comparison shows good agreement of the precipitation fields and illustrates a reasonable application of the introduced procedure. The presented results support the application of the approach for precipitation aggregates of, at least, annual or longer periods. However the derivation of climatologies led to satisfactory results at the respective temporal scales, though the influence of radar-specific errors can only be minimized to a certain degree. Further studies have to prove if an application independent processing of radar-derived precipitation rates leads to higher qualities and validities of the derived data in time and space.

No description available.

Orientador: Abel Maia Genovez / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Civil, Arquitetura e Urbanismo / Made available in DSpace on 2018-08-14T03:35:17Z (GMT). No. of bitstreams: 1 Campos_ElaineFrancode.pdf: 1908892 bytes, checksum: a9423fde9edbc3d1ba0202ab8f3be9b6 (MD5) Previous issue date: 2009 / Resumo: A precipitação é um componente do ciclo hidrológico que apresenta grande variabilidade espacial. Postos pluviométricos fornecem registros válidos apenas para um pequeno entorno do instrumento. A baixa densidade e as curtas séries de dados dos postos pluviométricos em bacias brasileiras tem sido fatores determinante nas incertezas dos resultados de diversos modelos hidrológicos que têm sido aplicados no país. Nesse contexto, o estudo de estimativas espaciais de precipitação podem ser extremamente úteis na busca de hidrogramas de projeto. Essas estimativas, embora a precariedade dos dados observados, devem fornecer uma boa noção da distribuição espacial das chuvas. Assim, com este trabalho, utilizam-se as estimativas da distribuição espacial da precipitação, obtidas a partir de dados de radar meteorológico, que são comparados aos dados da rede pluviométrica. Os dados da distribuição espacial da precipitação são usados como dados de entrada do modelo hidrológico distribuído Storm Water Management Model (SWMM), o qual foi aplicado a uma pequena bacia urbana da cidade de Campinas - SP, com área de drenagem de 7,59 km2. Os resultados são analisados comparando os hidrogramas simulados com os observados. Foram realizados estudos de caso na Bacia Hidrográfica Ralph Stettinger pertencente à Bacia do Ribeirão das Anhumas, na cidade de Campinas, estado de São Paulo, Brasil, que conta com boa rede pluviométrica para aferir as estimativas da distribuição espacial da chuva e com dados fluviométricos de várias enchentes ocorridas na bacia. Os dados de radar utilizados foram os do radar meteoro lógico localizado na cidade de São Paulo. A precipitação acumulada estimada pelo radar meteorológico para os nove eventos representou satisfatoriamente a chuva, quando comparados com os valores medidos nas estações pluviográficas, na qual, as diferenças obtiveram valor médio de 28%. Os hidrogramas resultantes da aplicação dos dados de radar reproduziram resultados satisfatórios se comparados aos resultados gerados pelo uso de dados de pluviógrafos, pois o coeficiente de eficiência obtido usando dados de radar foi de 0,69, já o coeficiente de eficiência encontrado usando dados de pluviógrafos foi de 0,85. / Abstract: The precipitation is one variable of the hydrological cycle that shows a large spatial variability. Conventional rain gauges only provide valid records from the positions next to the instrument. The low density of rainfall data recording posts in Brazilian watersheds has been the determinant factor of the uncertainties on the results of various hydrological models that have been applied in this country. In this context, the study of spatial estimates of precipitation can be extremely useful to find more accurate hydrographs of projects. These estimates, despite these poor data observed, may provide a good knowledge of the spatial rain distribution at the environment. With this study, the estimates of the spatial distribution of precipitation, using a weather radar, may be compared with the rainfall data from the recording stations. The data of the spatial distribution of precipitation is used as an input at the distributed hydrological model "Storm Water Management Model (SWMM)", which was applied to a urban watershed in the city Campinas - SP, with a drainage area in about 7,59 km2. The results are analyzed by comparing , the simulated hydrographs to observed flow rates. Studies of case were made in the watershed Ralph Stettinger belonging to Ribeirão das Anhumas watershed, which has a good pluviometric record to measure the estimates of the spatial distribution of rainfall and with tluviometric data from several flooding, occurred in this watershed. The radar data used was from the weather radar located in the city of São Paulo. The accumulated precipitation that the weather radar predicted for the nine events, represented the rain as well when compared with the values already measured in the rain gauges, which the average value differences was 28%. The resultant hydrographs of the application of radar data reproduced satisfactofy results if compared to the generated results by the rain gauges, because the obtained efficiency coefficient using radar data was 0,69, otherwise, the efficiency coefficient that was found using rain gauges data was 0,85. / Universidade Estadual de Campi / Recursos Hidricos, Energeticos e Ambientais / Mestre em Engenharia Civil

A Bacia do Alto Tiete abriga cerca de 50% dos habitantes do Estado de São Paulo e é afetada freqüentemente por eventos de inundações. Uma das principais fontes de problemas é a alta impermeabilização devida à ocupação da superfície nas últimas décadas. Um dos seus tributários secundários, a bacia do Rio Pirajuçara se insere neste contexto e sofre com problemas da mesma natureza. A modelagem hidrológica permite uma análise do escoamento superficial nestes ambientes e é útil na previsão de vazões por meio de redes telemétricas e sensoriamento remoto. Entretanto, redes telemétricas apresentam problemas de representatividade espacial e exposição, radares meteorológicos, apesar da maior resolução espaço-temporal das estimativas de precipitação, possuem várias fontes de erros e incertezas. A principal delas se refere à relação ZR. Deste modo, a integração dessas medições e estimativas pode minimizar erros de ambas. O objetivo deste estudo é analisar aspectos hidrológicos da Bacia do Rio Pirajuçara por meio do modelo TOPMODEL com medições de vazão e precipitação disponíveis para 18 eventos monitorados entre outubro de 2008 a outubro de 2009. O modelo TOPMODEL foi calibrado com dez eventos e verificado com os demais. A calibração foi realizada com os dados da telemetria da Bacia do Alto Tietê, radar meteorológico de São Paulo e a combinação de ambos por meio da análise objetiva estatística. Os resultados da calibração indicam que o melhor desempenho foi obtido com radar meteorológico, com número de NASH de 0,51, menor erro quadrático médio e menor viés médio absoluto. A verificação também indicou o mesmo resultado com número de NASH de 0,69. As simulações indicam que apesar da utilização da precipitação média, o modelo TOPMODEL simulou adequadamente cerca de 75% das vazões de alerta. O trabalho evidencia as limitações da telemetria e seus impactos na integração com os dados do radar. / The Alto Tiete watershed is home for about 50% of the inhabitants of São Paulo State and is affected by recurrent flashfloods. One major source of difficulties is the high rate of soil impermeabilization caused by dense surface occupation in the last decades. One of its secondary tributaries, the Pirajussara watershed suffers with similar problems. Hydrological modeling allows the analysis of runoff and other variables in these basins. It also useful for streamflow forecast based on telemetric networks and remote sensing measurements. However, surface networks lack spatial representativity and exposure is a also a issue, weather radars, in spite of their much higher spatial and temporal resolution rainfall estimation, are affect by several sources of errors and uncertainties; the most significant one being the ZR relationship. Thus, the integration of these measurements and estimates can minimize errors of both. The goal of the present work is to analyze the surface hydrology of the Pirajussara watershed based on the TOPMODEL, streamflow and rainfall measurements available for eighteen events between October 2008 and October 2009. The TOPMODEL was calibrated with ten events and verified with the remaining events. The calibration was performed with the Alto Tiete telemetric measurements of streamflow and rainfall only, the São Paulo weather radar (SPWR) only and a combination of both through the statistical analysis scheme. Calibration results show a better performance for the SPWR with a NASH number of 0.51, least SME and mean bias. On the other hand, the verification also indicated better results for the SPWR with a NASH number of 0.69. The simulations indicate that in spite of the use of the mean rainfall over the watershed, the TOPMODEL performed adequately for 75% of the streamflow alerts. It is also evident the limitation of the available network and its impacts on the integration to the SPWR rainfall data.