The intra-basin variability of snowmelt and meltwater runoff hydrology in an 8 km2 subarctic alpine tundra catchment was examined for the 2003 melt period. The catchment, Granger Creek, is within the Wolf Creek Research Basin, Yukon which is typical of mountain subarctic landscapes in north-western Canada. The study catchment was segmented into nine internally uniform zones termed Landscape Units (LUs) based on their similar hydrological, physiographic, vegetation and soil properties. Snow accumulation exhibited significant variability among the LUs, with greatest snow water equivalent in areas of tall shrub vegetation. Melt began first on southerly exposures and at lower elevations, yet average melt rates for the study period varied little among LUs with the exception of those with strong aspects. In LUs with capping organic soils, meltwater first infiltrated this surface horizon, satisfying its storage capacity and then percolated into the frozen mineral substrate. Infiltration and percolation into frozen mineral soils was restricted where melt occurred rapidly and organic soils were thin; in this case meltwater delivery rates exceeded the frozen mineral soil infiltration rate, resulting in high runoff rates. In contrast, where there were slower meltrates and thick organic soils, infiltration was unlimited and runoff was suppressed. The snow water equivalent had a large impact on runoff generation as soil storage capacity was quickly surpassed in areas of deep snow, diverting the bulk of meltwater laterally to the drainage network. A spatially distributed water balance indicated that snowmelt freshet was primarily controlled by areas with tall shrub vegetation that accumulate large quantities of snow and by alpine areas with no capping organic soils. The intra-basin water balance variability has important implications for modeling freshet in hydrological models.

The intra-basin variability of snowmelt and meltwater runoff hydrology in an 8 km2 subarctic alpine tundra catchment was examined for the 2003 melt period. The catchment, Granger Creek, is within the Wolf Creek Research Basin, Yukon which is typical of mountain subarctic landscapes in north-western Canada. The study catchment was segmented into nine internally uniform zones termed Landscape Units (LUs) based on their similar hydrological, physiographic, vegetation and soil properties. Snow accumulation exhibited significant variability among the LUs, with greatest snow water equivalent in areas of tall shrub vegetation. Melt began first on southerly exposures and at lower elevations, yet average melt rates for the study period varied little among LUs with the exception of those with strong aspects. In LUs with capping organic soils, meltwater first infiltrated this surface horizon, satisfying its storage capacity and then percolated into the frozen mineral substrate. Infiltration and percolation into frozen mineral soils was restricted where melt occurred rapidly and organic soils were thin; in this case meltwater delivery rates exceeded the frozen mineral soil infiltration rate, resulting in high runoff rates. In contrast, where there were slower meltrates and thick organic soils, infiltration was unlimited and runoff was suppressed. The snow water equivalent had a large impact on runoff generation as soil storage capacity was quickly surpassed in areas of deep snow, diverting the bulk of meltwater laterally to the drainage network. A spatially distributed water balance indicated that snowmelt freshet was primarily controlled by areas with tall shrub vegetation that accumulate large quantities of snow and by alpine areas with no capping organic soils. The intra-basin water balance variability has important implications for modeling freshet in hydrological models.

O estabelecimento de balanços hídricos no campo é difícil e dispendioso, sendo a variabilidade de seus componentes o maior problema para se obter resultados confiáveis. Esta variabilidade dos componentes é aqui apresentada para uma cultura de café desenvolvida no hemisfério sul, em um solo tropical com 10% de declividade. É concluído que a chuva deve ser medida com um número apropriado de repetições, que a irrigação pode introduzir grande variabilidade dos cálculos, que a evapotranspiração calculada a partir da equação do balanço hídrico tem coeficientes de variação muito altos, que o componente armazenamento de água no solo é o que mais contribui na propagação dos erros e que a enxurrada pôde ser satisfatoriamente controlada nesse declive por meio de práticas de manejo. / The establishment of field water balances is difficult and costly, the variability of its components being the major problem to obtain reliable results. This component variability is here presented for a coffee crop grown in the Southern Hemisphere, on a tropical soil with 10% slope. It is concluded that rainfall has to the measured with an appropriate number of replicates, that irrigation can introduce great variability into calculations, that evapotranspiration calculated from the water balance equation has high coefficients of variation, that the soil water storage component is the major contributor in error propagation calculations, and that the run-off could be satisfactorily controlled through crop management practices.

O cafeeiro é uma planta que tem seu crescimento e desenvolvimento fortemente afetado pelo regime hídrico, ora prejudicando, ora favorecendo a produção final, dependendo do estádio fenológico em que este se encontra durante uma possível seca. Para ter um conhecimento mais aprofundado do consumo de água do cafeeiro, assim como para um manejo da irrigação mais eficiente, uma ferramenta muito útil é o balanço hídrico, que pode ser medido no campo ou estimado através de modelos. O balanço hídrico de campo é demasiado trabalhoso, por isso os modelos são mais usados em virtude de sua rapidez e facilidade. No entanto, os modelos são frequentemente aplicados em condições agronômicas e ambientais diferentes das em que foram concebidos, necessitando de testes regionais. Um dos principais elementos de entrada para o cálculo do balanço hídrico é a precipitação pluviométrica, e o rigor em sua medida pode determinar se este será ou não condizente com a realidade, devendo sua variabilidade espacial ser levada em conta, o que não ocorre na maioria dos casos. O presente trabalho traz um estudo entre balanço hídrico medido no campo com café e os balanços hídricos climatológicos baseados na estimativa da evapotranspiração pelos métodos de Thornthwaite e Penman-Monteith, confeccionados em um programa computacional. Uma segunda parte trata do número ideal de pluviômetros a serem utilizados em uma área pequena, além das comparações destas medidas com duas estações meteorológicas. Ambos os estudos foram feitos para o município de Piracicaba - SP, com dados meteorológicos do período de 2003 a 2005. / The coffee plant has its growth and development strongly influenced by the water regime, either in favor or depressing the final yield, all depending on phonological stage. In order to have a better knowledge of the water requirement of the coffee crop, and also to have a more efficient irrigation management, a very useful tool is the water balance, which can be obtained through direct field measurements or through model estimation. The first are laborious and costly, while the second are simple and fast to be obtained. However, models are frequently applied to agronomic or environmental conditions that differ from those in which they were developed. One of the most important components for the calculation of water balances is the rainfall, and the precision of its measurement determines the confidence of the balances, indicating the need of taking into account rainfall variability, which is neglected in most cases. This study makes a comparison between climatologic water balances, based on the estimation of the evapotranspiration through the methods of Thornthwaite and Penman-Monteith, obtained though a computational program, and field water balances of a coffee crop. A second part of this study deals with the ideal number of pluviometers to be used in small areas, and the comparison of these measurements with two automatic meteorological stations. Both studies were carried out in Piracicaba, SP, Brazil, with data collected during 2003 - 2005.

O cafeeiro é uma planta que tem seu crescimento e desenvolvimento fortemente afetado pelo regime hídrico, ora prejudicando, ora favorecendo a produção final, dependendo do estádio fenológico em que este se encontra durante uma possível seca. Para ter um conhecimento mais aprofundado do consumo de água do cafeeiro, assim como para um manejo da irrigação mais eficiente, uma ferramenta muito útil é o balanço hídrico, que pode ser medido no campo ou estimado através de modelos. O balanço hídrico de campo é demasiado trabalhoso, por isso os modelos são mais usados em virtude de sua rapidez e facilidade. No entanto, os modelos são frequentemente aplicados em condições agronômicas e ambientais diferentes das em que foram concebidos, necessitando de testes regionais. Um dos principais elementos de entrada para o cálculo do balanço hídrico é a precipitação pluviométrica, e o rigor em sua medida pode determinar se este será ou não condizente com a realidade, devendo sua variabilidade espacial ser levada em conta, o que não ocorre na maioria dos casos. O presente trabalho traz um estudo entre balanço hídrico medido no campo com café e os balanços hídricos climatológicos baseados na estimativa da evapotranspiração pelos métodos de Thornthwaite e Penman-Monteith, confeccionados em um programa computacional. Uma segunda parte trata do número ideal de pluviômetros a serem utilizados em uma área pequena, além das comparações destas medidas com duas estações meteorológicas. Ambos os estudos foram feitos para o município de Piracicaba - SP, com dados meteorológicos do período de 2003 a 2005. / The coffee plant has its growth and development strongly influenced by the water regime, either in favor or depressing the final yield, all depending on phonological stage. In order to have a better knowledge of the water requirement of the coffee crop, and also to have a more efficient irrigation management, a very useful tool is the water balance, which can be obtained through direct field measurements or through model estimation. The first are laborious and costly, while the second are simple and fast to be obtained. However, models are frequently applied to agronomic or environmental conditions that differ from those in which they were developed. One of the most important components for the calculation of water balances is the rainfall, and the precision of its measurement determines the confidence of the balances, indicating the need of taking into account rainfall variability, which is neglected in most cases. This study makes a comparison between climatologic water balances, based on the estimation of the evapotranspiration through the methods of Thornthwaite and Penman-Monteith, obtained though a computational program, and field water balances of a coffee crop. A second part of this study deals with the ideal number of pluviometers to be used in small areas, and the comparison of these measurements with two automatic meteorological stations. Both studies were carried out in Piracicaba, SP, Brazil, with data collected during 2003 - 2005.

O estabelecimento de balanços hídricos no campo é difícil e dispendioso, sendo a variabilidade de seus componentes o maior problema para se obter resultados confiáveis. Esta variabilidade dos componentes é aqui apresentada para uma cultura de café desenvolvida no hemisfério sul, em um solo tropical com 10% de declividade. É concluído que a chuva deve ser medida com um número apropriado de repetições, que a irrigação pode introduzir grande variabilidade dos cálculos, que a evapotranspiração calculada a partir da equação do balanço hídrico tem coeficientes de variação muito altos, que o componente armazenamento de água no solo é o que mais contribui na propagação dos erros e que a enxurrada pôde ser satisfatoriamente controlada nesse declive por meio de práticas de manejo. / The establishment of field water balances is difficult and costly, the variability of its components being the major problem to obtain reliable results. This component variability is here presented for a coffee crop grown in the Southern Hemisphere, on a tropical soil with 10% slope. It is concluded that rainfall has to the measured with an appropriate number of replicates, that irrigation can introduce great variability into calculations, that evapotranspiration calculated from the water balance equation has high coefficients of variation, that the soil water storage component is the major contributor in error propagation calculations, and that the run-off could be satisfactorily controlled through crop management practices.