Análise de incertezas do balanço hídrico climatológico espacializado

• Autor(a) principal: Carvalho Neto, Romário Moraes
• Data de Publicação: 2016
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/15101
• topic: Incertezas; Análise de incertezas; Disponibilidade hídrica; Déficit hídrico; Excesso hídrico; Balanço hídrico climatológico espacializado; BHC; Python; Thornthwaite; Uncertainties; Uncertainties analysis; Water availability; Water deficit; Water surplus; Spatilized climatic water balance; CWB; CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL

The spacialized Climatic Water Balance (CWB) is a model that simulates the availability of water in the soil, for plants, in a space distributed manner. It is important to understand the possible uncertainties of this spatialization, so the opportunities of its use in the public policies can be discussed, as well as the advantages of its use, its limitations, having a satisfying result in its use and that its optimization may be allowed. Searching to know the uncertainties in the CWB's spatial distribution, this thesis aims to evaluate such uncertainties due to: 1) the spatialization methods and 2) the density of information used for the spatialization. To address the matter of the uncertainties regarding the spatialization methods, two methods were analyzed. The first one calculates the CWB punctually at stations and then spatializes these values by interpolation (Calculation-Interpolation principle, CI), while the second method is to interpolate first the CWB's variables (precipitation and evapotranspiration) and then calculates it for each pixel (Interpolation-Calculation principle, IC). In addition, the influence of the interpolators were also analyzed. To analyze the uncertainties relating to the density of information, the strategy of comparing the differences arising from the results of the precipitation and the evapotranspiration interpolations and the calculation of the spacialized CWB was used, deleting stations and analyzing the error generated by this decrease of information density. This analysis was first done with the suppression of one station and then, by removing two, three and so on, until the remaining of 3 stations, referring to the minimum number of points required to perform interpolation. To make possible the spatialization of the CWB in a distributed way, a tool in the PythonTM programming language, using the package ArcPy® was created to perform the calculations of the CWB. The study area of this work was the plain area of Veneto, Region of Italy. The results showed that although the analyzes have indicated a trend of smaller uncertainties in the IC method in relation to the CI, these differences were not statistically significant at the 5% level. It was also observed that the CI method brings more uncertainties to the spatialization, particularly when there is water deficit in the CWB and/or ground recharge, by smoothing such balance values between stations, not properly representing the CWB in these areas. The uncertainty analysis performed in this study was also able to show which months can carry greater uncertainty into their spatializations, both P, ETo and the CWB and that the high variability of precipitation carries uncertainties in their spatial distribution. The spatial representation of the CWB showed, for this study, to have greater uncertainty at the beginning of the dry season, when starts the reservoir drawdown, or the beginning of the rains that cause the filling of the reservoir in the soil. The estimated uncertainties to the stations reduction from 15 to 3, ranged from 3 to 27% for precipitation, from 1 to 36% for ETo and 1 to 88% for CWB, considering 16 stations as the truth reference. As the IC method allows the CWB spatialization with different scenarios of Available Water Capacity (AWC), which is not feasible with the CI method, since CI considers only the AWC at the station's place, the use of the IC method was more suitable to represent the CWB at smaller scales (larger areas). This possibility provides more options for the application of the spatialized CWB in public policies, allowing the generation of crops scenarios in a more detailed and dynamic way than the CI method, besides the possibility of its adequacy to the reality of each soil type.