Submitted by Gabriela Lopes (gmachadolopesufpel@gmail.com) on 2018-08-13T18:52:36Z No. of bitstreams: 1 DISSERTAÇÃO VIVIANE DORNELES.pdf: 871631 bytes, checksum: b578ded3d286fade9470c2c6eb66553f (MD5) / Approved for entry into archive by Aline Batista (alinehb.ufpel@gmail.com) on 2018-08-16T19:21:09Z (GMT) No. of bitstreams: 1 DISSERTAÇÃO VIVIANE DORNELES.pdf: 871631 bytes, checksum: b578ded3d286fade9470c2c6eb66553f (MD5) / Made available in DSpace on 2018-08-16T19:21:09Z (GMT). No. of bitstreams: 1 DISSERTAÇÃO VIVIANE DORNELES.pdf: 871631 bytes, checksum: b578ded3d286fade9470c2c6eb66553f (MD5) Previous issue date: 2017-02-22 / O conhecimento do comportamento dos eventos extremos de precipitação e consequentes vazões observadas é de extrema importância para desenvolvimento de projetos de obras hidráulicas, tais como, drenagem agrícola, vertedouros de barragens, bueiros e drenagem urbana, canais de terraço, entre outros. Considerando a esparsa rede de monitoramento de vazão, disponibilizada no território brasileiro, é possível realizar a análise da vazão de projeto, quando não se tem dados históricos de vazão, por estudo de chuvas intensas. A dissertação, desenvolvida no curso de mestrado, no Programa de Pós-Graduação em Manejo e Conservação do Solo e da Água, foi elaborada a partir da pesquisa de chuvas intensas no município de Pelotas/RS, sendo composta por dois artigos científicos. Os artigos divididos em 1 e 2, foram realizados mediante a análise de registros pluviográficos e dados pluviométricos, respectivamente. Objetivou-se com este estudo determinar as equações intensidade-duração-frequência, do município de Pelotas/RS, utilizando a metodologia de interpretação e análise de pluviogramas e desagregação de chuvas diárias, bem como, comparar as estimativas das intensidades de precipitação por meio da aplicação da equação utilizada atualmente no município. Para isso, foram utilizados dados pertencentes à Estação Agroclimatológica mantida pela Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA) e pela Universidade Federal de Pelotas (UFPel), compondo 34 anos de séries históricas (1982 a 2015), utilizando-se as metodologias anteriormente mencionadas. As séries históricas foram submetidas à análise estatística a fim de identificar o modelo probabilístico teórico que representasse melhor o ajuste aos dados, e o método empírico de análise dos dados, por se ter disponíveis 34 anos de dados. As equações intensidade-duração-frequência obtidas pela análise de registros pluviográficos e dados pluviométricos, bem como, a pela metodologia híbrida (empírico) e convencional (teórico) permitem o cálculo das intensidades máximas de precipitação, para as durações de chuva de 5 até 1440 min, nos períodos de retorno de 2 a 100 anos. / Knowledge of the behavior of the extreme events of precipitation and consequent observed flows is of extreme importance for the development of projects of hydraulic works, such as agricultural drainage, spillways of dams, culverts and urban drainage, terrace channels, among others. Considering the sparse flow monitoring network available in the Brazilian territory, it is possible to carry out the analysis of the project flow, when there is no historical flow data, due to the study of heavy rains. The dissertation, developed in the master's degree program in the Postgraduate Program in Soil and Water Management and Conservation, was elaborated based on intense rainfall research in the city of Pelotas, Rio Grande do Sul, Brazil, and was composed of two scientific articles. The articles divided into 1 and 2, were carried out through the analysis of pluviographic records and rainfall data, respectively. The objective of this study was to determine the intensity-duration-frequency equations of the municipality of Pelotas / RS, using the methodology of interpretation and analysis of pluviograms and disaggregation of daily rains, as well as to compare the estimates of precipitation intensities by means of Application of the equation currently used in the municipality. For this purpose, data from the Agroclimatological Station maintained by the Brazilian Agricultural Research Company (EMBRAPA) and the Federal University of Pelotas (UFPel) were used, composing 34 years of historical series (1982 to 2015) using the aforementioned methodologies. The historical series were submitted to statistical analysis in order to identify the theoretical probabilistic model that best represented the adjustment to the data, and the empirical method of data analysis, because 34 years of data were available. The intensity-duration-frequency equations obtained by the analysis of pluviometric and rainfall data, as well as by the hybrid (empirical) and conventional (theoretical) methods allow the calculation of the maximum precipitation intensities for rainfall durations from 5 to 1440 Min, in the return periods of 2 to 100 years.

Several previous studies have found that various characteristics of the Indian monsoon rainfall have shown secular changes over the past century. In this study, using a gridded (1degree) daily rainfall dataset, we analyse the spatio-temporal characteristics of the intensity and duration of monsoon (June through September) rainfall for secular changes over the last 50 years. The characteristics of the duration of rain events are described by wet and dry spells. A wet/dry spell is defined as a period of consecutive days with rainfall above/below a particular threshold. We choose to use a threshold that is a function of the local climatological mean, given the spatial heterogeneity of mean monsoon rainfall. The wet and dry spells are then divided into three categories: short [1 to 7 days], moderate [8 to 10 days], long [11 and more days] and analysed for changes over the past 50 years [19512000]. We find that while the number of short duration wet spells show a significant increase over the last 50 years (~15% change), the number of long duration wet spells show a significant decrease (~25%). Furthermore, while the numbers of short duration dry periods have shown a significant increase, the moderate and long duration dry spells do not shown an appreciable change. This increase and decrease in the short and long duration wet spells offset each other and consequently the total number of rainy days during the season has not shown any significant change over the past 50 years. In addition to the duration of wet and dry spells, we also analysed for changes in the accumulated rainfall of the short, medium and long duration wet spells. Our analysis suggests that while the depth of accumulated rainfall in short duration wet spells has shown a significant increase (~20%), the depth of rain in the long duration spells has shown a significant decrease (~30%) in the past fifty years.

The scarcity of data poses a challenging obstacle for the study of natural disasters, especially in developing countries where the social vulnerability plays as important a role as the physical vulnerability.  The work presented in this thesis is oriented towards the demonstration of the usefulness of press archives as a data source for the temporal and spatial analysis of landslides in Tegucigalpa, Honduras for the period between 1980 and 2005.  In the last four decades, Tegucigalpa has been characterized by a disorganized urban growth that has significantly contributed to the destabilization of the city’s slopes.  In the first part of the thesis, a description of the database compilation procedure is provided.  The limitations of using data derived from press archives have also been addressed to indicate how these affect the subsequent landslide analyses.  In the second part, the temporal richness offered by press archives has allowed the establishment of rainfall thresholds for landslide occurrence.  Through the use of the critical rainfall intensity method, the analysis of rainfall thresholds for 7, 15, 30 and 60 antecedent days shows that the number of yielded false alarms increases with the threshold duration.  A new method based on the rainfall frequency contour lines was proposed to improve the distinction between days with and without landslides.  This method also offers the possibility to identify the landslides that may only occur with a major contribution of anthropogenic disturbances as well as those landslides induced by high-magnitude rainfall events.  In the third part, the matrix method has been employed to construct two landslide susceptibility maps: one based on the multi-temporal press-based landslide inventory and a second one based on the landslide inventory derived from an aerial photograph interpretation carried out in 2014.  Despite the low spatial accuracy provided by the press archives in locating the landslides, both maps exhibit 69% of consistency in the susceptibility classes and a good agreement in the areas with the highest propensity to landslides.  Finally, the integration of these studies with major actions required to improve the process of landslide data collection is proposed to prepare Tegucigalpa for future landslides.

I Sverige inträffar många skyfall och intensiva regn under sommarmånaderna. Det finns inga uppenbara geografiska mönster, vilket är en skillnad gentemot älv- eller sjööversvämningar där det vanligtvis är känt vilka områden som kan komma att översvämmas vid en viss vattennivå eller ett visst vattenflöde. För individer och samhällsaktörer innebär en skyfallshändelse i många fall en stor överraskning då skyfall utvecklas snabbt och dagens meteorologiska prognossystem i stort inte lyckas att prognosticera extrema regn korrekt med avseende på mängd, tid och plats. Vädervarningar kommer med kort varsel eller uteblir helt. Konsekvenserna av intensiv nederbörd och skyfall är främst översvämningar och erosionsskador på byggnader och infrastruktur, men även störningar och avbrott i olika samhällsfunktioner som kan påverka samhället och individer utanför det drabbade området. I denna avhandling har 15 år av försäkringsskadedata använts för att undersöka samband mellan nederbördsintensitet och skyfallsskador. Även påverkan av andra faktorer som topografi, bebyggelse och socioekonomiska aspekter har undersökts. Resultaten visar att regnintensitet under ett 60 minuters intervall i kombination med korta perioder av extrem intensitet, tillsammans med topografiska faktorer spelar en betydande roll vid uppkomsten av skador.

Accelerated soil erosion is one of the major threats to agricultural production in Ethiopia and the Harerge region is not exceptional. It is estimated that about 1.5 billion tones of soil is being eroded every year in Ethiopia. In the extreme cases, especially for the highlands, the rate of soil loss is estimated to reach up to 300 t ha-1yr-1 with an average of about 70 t ha -1yr-1 which is beyond any tolerable level. The government have made different attempts to avert the situation since 1975 through initiation of a massive program of soil conservation and rehabilitation of severely degraded lands. Despite considerable efforts, the achievements were far bellow expectations. This study was aimed at assessing the effect of some soil properties, rainfall intensity and slope gradients on surface sealing, soil erodibility, runoff and soil loss from selected sites in the Harerge region, eastern Ethiopia, using simulated rainfall. Soil loss was also estimated for the sites using Soil Loss Estimation Model for Southern Africa (SLEMSA) and the Universal soil Loss Equation (USLE). Moreover, the effectiveness of various rates and patterns of wheat residue mulching in controlling soil loss was also evaluated for one of the study sites, (i.e. Regosol of Alemaya University), under both rainfall simulation and field natural rainfall conditions. For most of the erosion parameters, the interaction among soil texture, slope gradient and rainfall intensity was significant. In general however, high rainfall intensity induced high runoff, sediment yield and splash. The effect of slope gradients on most of the erosion parameters was not significant as the slope length was too small to bring about a concentrated flow. The effect of soils dominated by any one of the three soil separates on the erosion parameters was largely dependent on rainfall intensity and slope gradient. The soils form the 15 different sites in Harerge showed different degrees of vulnerability to surface sealing, runoff and sediment yield. These differences were associated with various soil properties. Correlation of soil properties to the erosion parameters revealed that aggregate stability was the main factor that determined the susceptibility of soils to sealing, runoff and soil loss. This was in turn affected by organic carbon content, percent clay and exchangeable sodium percentage (ESP). Soils with relatively high ESP such as those at Babile (13.85) and Gelemso (7.18) were among the lowest in their aggregate stability (percent water stable aggregates of 0.25 –2.0mm diameter); and have highest runoff and sediment yield as compared to other soils in the study. Similarly, most of those soils with relatively low ESP, high organic carbon content (OC%) and high water stable aggregates such as Hamaressa, AU (Alemaya University) vertisol and AU regosol were among the least susceptible to sealing and interrill erosion. Nevertheless, some exceptions include soils like those of Hirna where high runoff was recorded whilst having relatively high OC%, low ESP and high water stable aggregates. Both the SLEMSA and USLE models were able to identify the erosion hazards for the study sites. Despite the differences in the procedures of the two models, significant correlation (r = 0.87) was observed between the values estimated by the two methods. Both models estimated higher soil loss for Gelemso, Babile, Karamara and Hamaressa. Soil loss was lower for Diredawa, AU-vertisol and AU-Alluvial all of which occur on a relatively low slope gradients. The high soil loss for Babile and Gelemso conforms with the relative soil erodibility values obtained under rainfall simulation suggesting that soil erodibility, among others, is the main factor contributing to high soil loss for these soils. The difference in the estimated soil losses for the different sites was a function of the interaction of the various factors involved. Though the laboratory soil erodibility values were low to medium for Hamaressa and Karamara, the estimated soil loss was higher owing to the field topographic situations such as high slope gradient. SLEMSA and USLE showed different degrees of sensitivities to their input variables for the conditions of the study sites. SLEMSA was highly sensitive to changes in rainfall kinetic energy (E) and soil erodibility (F) and less sensitive to the cover and slope length factors. The sensitivity of SLEMSA to changes in the cover factor was higher for areas having initially smaller percentage rainfall interception values. On the other hand, USLE was highly sensitive to slope gradient and less so to slope length as compared to the other input factors. The study on the various rates and application patterns of wheat residue on runoff and soil loss both in the laboratory rainfall simulation and under field natural rainfall conditions revealed that surface application of crop residue is more effective in reducing soil loss and runoff than incorporating the same amount of the residue into the soil. Likewise, for a particular residue application method, runoff and soil loss decreased with increasing application rate of the mulch. However, the difference was not significant between 4 Mg ha-1 and 8 Mg ha-1 wheat straw rates suggesting that the former can effectively control soil loss and can be used in areas where there is limitation of crop residues provided that other conditions are similar to that of the study site (AU Regosols). The effectiveness of lower rates of straw (i.e. less than 4 Mg ha-1 ) should also be studied. It should however be noted that the effectiveness of mulching in controlling soils loss and runoff could be different under various slope gradients, rainfall characteristics and cover types that were not covered in this study. Integrated soil and water conservation research is required to develop a comprehensive database for modelling various soil erosion parameters. Further research is therefore required on the effect of soil properties (with special emphasis to aggregate stability, clay mineralogy, exchangeable cations, soil texture and organic matter), types and rates of crop residues, cropping and tillage systems, mechanical and biological soil conservation measures on soil erosion and its conservation for a better estimation of the actual soil loss in the study sites. Copyright 2004, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. Please cite as follows: Bobe, BW 2004, Evaluation of soil erosion in the Harerge region of Ethiopia using soil loss models, rainfall simulation and field trials, PhD thesis, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-08022004-141533 / > / Thesis (PhD (Soil Science))--University of Pretoria, 2004. / Plant Production and Soil Science / unrestricted

From the Proceedings of the 1973 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - May 4-5, 1973, Tucson, Arizona / Hydrologic data collected from three small urban watersheds and one rural watershed were analyzed for the purpose of investigating the effect of urbanization on runoff. A procedure developed by the Soil Conservation Service was used to explain the relationship between the amount of rainfall and runoff. It was noted that the runoff curve number, a parameter of the method, increased as the percentage of impervious area increased. Also, there was evidence that a linear relationship existed between the runoff volume and its corresponding peak rate.

From the Proceedings of the 1973 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - May 4-5, 1973, Tucson, Arizona / An infiltrometer was designed to more closely simulate natural storm characteristics and still maintain sufficient portability to be used in various test sites in the field. In addition to portability, a relatively large test plot can be used over a relatively long duration. The instrument is designed to produce rainfall intensities of 2 to 6 inches per hour which are comparable to natural storm intensities found in northern Arizona. Capillary tubes produce water drops of equivalent kinetic energy at impact to natural raindrops. Errors due to lateral flow are minimized through peripheral wetting. Mounting the infiltrometer on a four-wheel drive vehicle allows nearly the portability of a hand carried unit with a greater water carrying capacity and allows the equipment to be large enough to test a representative plot.

From the Proceedings of the 1978 Meetings of the Arizona Section - American Water Resources Assn. and the Hydrology Section - Arizona Academy of Science - April 14-15, 1978, Flagstaff, Arizona / The runoff curve number rainfall- runoff relationships may be defined in two ways: (1) by formula, which uses total storm rainfall and a curve number, but not intensity or duration descriptors; and (2) rainfall loss accounting using a 4, rate and a specific intensity duration distribution of the function i(t) = 1.5P(5(1 +24t /T)-(1/2)-1) /T, where i(t) is the intensity at time t for a storm of duration T. Thus, the curve number method is found to be a special case of φ index loss accounting. The two methods are reconciled through the identity 1.2S = φT, leading to the relationship CN - 1200/(12 +φT). The effects of rainfall intensity on curve number are felt through deviations from the necessary causative intensity - duration curve. Some sample alternate distributions are explored and the effects on curve number shown. Limitations are discussed.

Cada vez são mais comuns problemas relacionados a movimentos de massa nas encostas de clima tropical no Estado do Rio de Janeiro, especialmente na Serra do Mar, provocados por acumulados pluviométricos intensos. A ocupação humana desordenada de áreas sensíveis a tais processos geomorfológicos, bem como as condicionantes geológicas, geomorfológicas, pedológicas e de uso e cobertura do solo são apontadas como fatores cruciais na explicação desses processos. O maior conhecimento da dinâmica pluviométrica bem como suas interações com tais aspectos físicos ligados ao relevo parece ser a chave dessa maior compreensão desses fenômenos. Assim foram realizadas pesquisas relacionadas aos volumes e intensidades das chuvas na região do Alto Curso do Rio São João, bem como uma análise dos movimentos de massa identificados através de imagens de satélite e in loco, como forma de fornecer subsídios à melhor gestão do espaço dessas regiões montanhosa estão vulneráveis a movimentos de massa. A correlação entre os acumulados e a intensidade pluviométrica com fenômenos climáticos de escala global, como El Niño e La Niña também foi contemplada nessa pesquisa, mostrando uma relação mais alta com relação à intensidade da chuva mensal para anos de El Niño e para anos de La Niña uma reduzida ocorrência dessas intensidades pluviométricas. Os estudos revelaram que os tipos de solos e sua cobertura e uso têm uma grande influência na deflagração de movimentos de massa. Foram observados um número reduzido de movimentos de massa em áreas naturais e uma maior proporção desses movimentos em áreas utilizadas para a atividade da pecuária na região. Grande parte dos movimentos de massa ocorreram em áreas de Cambissolos (áreas mais elevadas) e Latossolos (áreas de encostas em menores altitudes). Ambos os solos são mais espessos do que os encontrados em áreas mais declivosas, apresentando maior acúmulo de materiais a serem mobilizados durante grandes acumulados pluviométricos, gerando movimentos de massa. A análise mostrou também que áreas mais chuvosas e com maior ocorrência de acumulados pluviométricos extremos, acima de 100 mm/dia e acima de 30mm/mês concentraram um número maior de movimentos de massa, como a região mais próxima da estação de Quartéis (porção leste). Por outro lado áreas bastante elevadas, com altas declividades, porém com predomínio de Mata Atlântica e áreas com solos menos espessos, como os Neossolos Litólicos, se mostraram com um número reduzido desses processos. Enfim esse estudo mostrou a necessidade de se gerir melhor os espaços dessas áreas sensíveis sob o ponto de vista geomorfológico, até por que são áreas na periferia de regiões densamente habitadas e cujas demandas tendem a se tornar cada vez mais marcantes, o que pode gerar problemas locais, atingindo sua população e economia, com sérias conseqüências para o ambiente. / Are increasingly common problems related to mass movements on the slopes of tropical climate in the State of Rio de Janeiro, especially in the Serra do Mar, caused by intense rainfall accumulated. The disorganized human occupation of sensitive areas such geomorphological processes and the geological conditions, geomorphology, soil and use and land cover are cited as crucial factors in explaining these processes. The understanding of rainfall dynamics as well as their interactions with such physical aspects related to the relief seems to be the key to this greater understanding of these phenomena. So were performed research related to volume and intensity of rainfall in the Upper St. John River Course, as well as an analysis of mass movements identified through satellite imagery and on-site as a way to provide subsidies to better management of these space mountainous regions as vulnerable to mass movements. The correlation between rainfall intensity and accumulated with global-scale climatic phenomena like El Niño and La Niña was also considered in this study, showing a higher ratio with respect to the intensity of monthly rainfall for El Niño years and La Niña years a reduced occurrence of rainfall intensities. Studies have shown that the types of soils and their cover and land use have a great influence on outbreaks of mass movements. We observed a small number of mass movements in natural areas and a higher proportion of these movements in areas used for livestock activity in the region. Most mass movements occurred in areas of Cambisols (higher areas) and Oxisols (slopes in areas of lower altitudes). Both soils are thicker than those found in hilly areas, with higher accumulation of materials to be deployed during large accumulated rainfall, generating mass movements. The analysis also showed that areas with increased wet and accumulated occurrence of precipitation extreme excess of 100 mm/day and above 30mm/month concentrated a larger number of mass movements, as the region closest to the station Quartéis (East portion).On the other hand very high areas with steep slopes, but with a predominance of Atlantic Forest and areas with thinner soils, such as Entisols, were with a small number of these processes. Finally this study showed the need to better manage these areas, sensitive areas under the geomorphological point of view, even for areas that are on the outskirts of densely populated and whose demands tend to become increasingly salient, which can cause problems locations, reaching its population and economy, with serious consequences for the environment

The rainfall specifics characteristics vary from one region to another. The knowledge of the rainfall potential to cause erosion is necessary to plan agricultural and civil engineering activities. For Quarai and Rio Grande (RS, Brazil), were determined the rainfall erosivity and its relationship with precipitation and rainfall coefficient, rainfall patterns and rainfall return period. Were used rainfall data charts of 38 years of Quarai (1966-2003) and 23 years of Rio Grande (1957, 1959-1978 e 1980-1981). For each erosive rainfall were separated the segments of the rainfall chart with the same intensity and the data registered in worksheet. With the software Chuveros were estimated the mean monthly and annual rainfall erosivity, the EI30 index in the International System of Units and the rainfall patterns. The mean monthly values of precipitation and erosivity index were expressed as percentages of the mean annual values of precipitation and erosivity index, respectively, to obtain the curve of accumulated distribution of precipitation and erosivity index in function of time. The rainfall coefficient (Rc) was calculated. Were performed Pearson correlations and linear regressions between the erosivity index EI30 and the mean annual values of precipitation and rainfall coefficient.. The rainfall return period were calculated for 2, 5, 10, 20, 50 e 100 years. The mean annual values of EI30 for Quarai and Rio Grande were 9292.1 e 5135.0 MJ mm ha-1 h-1 ano-1, respectively. Were obtained the equations EI30 = -754.37 + 13.50 p (r2 = 0.85) e EI30 = - 47.35 + 82.72 Rc (r2 = 0.84) for Quarai. For Rio Grande the equations were not significant. In relation to the total of the rainfalls studied for each place, 44.3% of the number and 90.4% of the volume were erosive in Quarai, and 32.6% of the number and 99.3% of the volume were erosive in Rio Grande. The method of extreme distribution type I was adequate for obtaining the curves of intensity-duration-frequency. The rainfall return periods may be calculated by the equations using the values of the parameters found, or by the figures of intensity-durationfrequency. / As características específicas das chuvas variam de uma região a outra. O conhecimento da potencialidade das chuvas em causar erosão é necessário para planejar atividades agrícolas e de engenharia civil. Para as localidades de Quarai e Rio Grande (RS), foram determinados a erosividade da chuva e a relação com a precipitação e o coeficiente de chuva, os padrões da chuva e o período de retorno da chuva. Utilizaram-se dados pluviográficos de 38 anos de Quarai (1966-2003) e 23 anos de Rio Grande (1957, 1959-1978 e 1980-1981). Para cada chuva erosiva foram separados os segmentos do pluviograma com a mesma intensidade e registrados os dados em planilha. Com o programa Chuveros foram calculadas a erosividade mensal, anual e média das chuvas pelo índice EI30 no Sistema Internacional de Unidades e os padrões de chuva. Os valores médios mensais da precipitação e do índice de erosividade foram expressos como percentagens do valor médio anual da precipitação e do índice de erosividade respectivamente, para obter a curva de distribuição acumulada da precipitação e do índice de erosividade em função do tempo. O coeficiente de chuva (Rc) foi calculado. Foram realizadas correlações de Pearson e regressões lineares simples entre o índice de erosividade EI30 e os valores médios anuais de precipitação e de coeficiente de chuva. O período de retorno foi calculado para 2, 5, 10, 20, 50 e 100 anos. Os valores médios anuais de EI30 para Quarai e Rio Grande foram 9292,1 e 5135,0 MJ mm ha-1 h-1 ano-1, respectivamente. Para Quarai, obtiveram-se as equações EI30 = -754,37 + 13,50 p (r2 = 0,85) e EI30 = -47,35 + 82,72 Rc (r2 = 0,84). Para Rio Grande as equações não foram significativas. Em relação ao total das chuvas estudadas em cada localidade, 44,3% do número e 90,4% do volume foram erosivas em Quarai, e 32,6% do número e 99,3% do volume foram erosivas em Rio Grande. O método da distribuição extrema tipo I foi adequado para obter as curvas de intensidade duração-freqüência. Os períodos de retorno da chuva podem ser calculados através das equações utilizando os valores dos parâmetros achados, ou pelos gráficos das curvas de intensidade-duração-freqüência.