Stream water quality is controlled by the interaction of natural and anthropogenic factors over a range of temporal and spatial scales. Among these anthropogenic factors, land cover changes at catchment scale can affect stream water quality. This work aims to evaluate the influence of land use and seasonality on stream water quality in a representative tropical headwater catchment named as Córrego Água Limpa (Sao Paulo, Brasil), which is highly influenced by intensive agricultural activities and urban areas. Two systematic sampling approach campaigns were implemented with six sampling points along the stream of the headwater catchment to evaluate water quality during the rainy and dry seasons. Three replicates were collected at each sampling point in 2011. Electrical conductivity, nitrates, nitrites, sodium superoxide, Chemical Oxygen Demand (DQO), colour, turbidity, suspended solids, soluble solids and total solids were measured. Water quality parameters differed among sampling points, being lower at the headwater sampling point (0m above sea level), and then progressively higher until the last downstream sampling point (2500m above sea level). For the dry season, the mean discharge was 39.5ls (from April to September) whereas 113.0ls were averaged during the rainy season (from October to March). In addition, significant temporal and spatial differences were observed (P<0.05) for the fourteen parameters during the rainy and dry period. The study enhance significant relationships among land use and water quality and its temporal effect, showing seasonal differences between the land use and water quality connection, highlighting the importance of multiple spatial and temporal scales for understanding the impacts of human activities on catchment ecosystem services.

No abstract

Due to the importance of the species Araucaria angustifolia (Bertol.) Kuntze and on account of its presence in the "Núcleo Cunha", it was decided to study its influence on the process of rainfall redistribution. Daily throughfall (Tr) and stemflow of trees (Et) in a plot of 600 m 2 , equipped with 15 rain gauges and devices for collecting stemflow water on 19 trees were measured. A rain gauge installed in the open area located near the plot provided the rainfall data in the open (Pa). The interception (I) was estimated by the equation: I = Pa -(Et + Tr). The experimental period covered from October 2008 to March 2009. Sixty-eight rain events were recorded, resulting in 1,565.9 mm of precipitation in the open, and 163.4 mm were intercepted by the canopy. The gross precipitation is directly proportional to the rainfall in the open, since the intercept tends to decrease with the increase of rainfall classes. Negative interception events were recorded, showing another water entrance in the system. Palavras-chave:Araucaria angustifolia (Bertol.) Kuntze, gross precipitation, interception, throughfall, stemflow, precipitação efetiva, interceptação, transprecipitação, escoamento pelo tronco. 117 4 3 artigo anterior 9 Anais II Seminário de Recursos Hídricos da Bacia Hidrográfica do Paraíba do Sul: Recuperação de Áreas Degradadas Serviços

Given the intrinsic hydrological cycle made of large input of water vapour and intense precipitation producing large volumes of water and sediment, modelling runoff and water losses in humid tropical watersheds is important for forest and water resources management. For instance, reliable simulations of the water cycle in such environments are a prerequisite for predictions of water quality, soil erosion and the climate change effects on water resources. The distributed parameter, physically based, continuous simulation, daily time step AnnAGNPS model, was implemented in almost completely forested (98% of its area, 0.56 km) Cunha watershed (Brazil) to assess its capability to simulate hydrological processes under tropical conditions. The simulated surface runoff was compared to 4-year observations with statistical indices on several time scales. The model, running with default CN of forest, showed poor predictions of runoff. After increasing CN from 63 to 72 by calibration, the runoff prediction capability of AnnAGNPS was satisfactory on annual, seasonal and monthly scales, while daily runoff predictions were less accurate. Modelling water losses at event scale showed that the effect of forest vegetation on water retention during a single precipitation was more limited than for longer periods (months, seasons and years), since evapo-transpiration and interception account for small shares (>20%) of total precipitation. This study demonstrated that the AnnAGNPS model has reliable runoff prediction capacity in tropical forest watersheds at the annual and seasonal scales (E > 0.73), whereas daily runoff simulations are less accurate (E = 0.44). The use of this model may prove an important tool for water resource and territory management in tropical rainforests.

A physically based semi-distributed model, TOPMODEL, has been applied to one catchment in Cunha, São Paulo, Brazil. The model uses a topographic index which highlights significant hydrological areas within the catchment. The model simulate streamflow and analyse the behaviour of the catchment. The first part of this paper concerns to the first values of calibration parameters, experiments the validity of the premisses of the model, and, mainly, tries to understand how does the hydrology of D catchment works. The results are more efficient on simulation of isolate events than on events of long period. The simulation of hydrology behaviour subject to clear-cutting has showed a diminution of the time of rise and a significant increase of streamflow simulated: from 17 to 44 %, in comparaison to D forest catchment. They also show that from 7 to 93 % of the streamflow is generated from saturated contributing areas (sources areas). It's advisable to avoid clearcutting. If necessary, the riparian zone has to be preserved; and where they have been destroyed, they have to be recovered urgently. Agricultural lands or grasslands have to use the best management practices, in order to keep the natural conditions of infiltration.

Tropical montane cloud forests (TMCF) have paramount importance with regards to hydrological and ecological functions. Despite such importance in terms of water provision, they are one of the most endangered ecosystems on the planet. One important topic that is still unclear is related to TMCF regeneration: can it recover the hydrological function of such critical ecosystems? In the present paper, the hydrological effects of the natural regeneration of TMCF were examined. We selected three catchments with different degrees of TMCF, and we examined the relationship between the amount of TMCF over time and hydrological variables such as water yields and baseflow using a time series (1984–2011). Our results suggest that the increase in TMCF cover can contribute to increases in water yield. This pattern was found in a catchment with an increase of approximately 20% of forest cover. In catchments with TMCF increase lower than this threshold, water yield and flow path were influenced by hydroclimatic variables such as air temperature and rainfall. The effect of the reduction in water yield, in accordance with the increase in mean temperature, sheds light on possible consequences of climate change on ecosystem services produced in foggy environments. In light of these findings, aiming to increase the water yield, we recommend that regions frequently affected by fog are prioritized for forest restoration or conservation. To ensure significant effects, regeneration and/or successional development of TMCF has to reach at least 20% of a catchment area.

EFFECT OF SOIL USE ON THE QUALITY OF WATER RESOURCE IN WATERSHED USING MULTIVARIATE STATISTICAL ANALYSIS  MARCELO ZANATA1; TERESA CRISTINA TARLÉ PISSARRA2; ANTONIO SERGIO FERRAUDO3; MAURICIO RANZINI4 E SERGIO CAMPOS5 1 Pesquisador Científico, Instituto Florestal de São Paulo, Batatais-SP, marcel_zanata@hotmail.com2 Profa. Doutora, Faculdade de Ciências Agrárias e Veterinárias –UNESP, Jaboticabal, SP, teresap@fcav.unesp.br3 Prof. Doutor, Faculdade de Ciências Agrárias e Veterinárias –UNESP, Jaboticabal, SP, fsajago@gmail.com4 Pesquisador Científico, Instituto Florestal de São Paulo, Batatais-SP, ranzini@ig.com.br5 Prof. Titular, Faculdade de Ciências Agronômicas–UNESP, Botucatu, SP, seca@fca.unesp.br  1 ABSTRACT Changes in soil use  in  protected areas modify the landscape, and therefore, monitoring water quality in headwaters is essential to improve the population health and control water pollution. The objective of this study was to determine environmental changes as a result of soil use, evaluating physical and chemical characteristics of the water of the drainage network in watersheds.  The study was conducted to determine the effects of soil use in watersheds with first order streams, related with management practices of water quality, according to the strategic  importance  for the supplying source. Watersheds are predominantly covered by sugar cane, reforested with Pinus and native forest.   Temperature (TºC), pH, electrical conductivity (EC) and total dissolved solids (TDS) were the parameters determined in the   headwaters. Dissolved oxygen (DO) and turbidity (T) were obtained in the lab. The majority of variables and factors were explained by changes in soil use associated with management practices. Two factors resulting from the multivariate analysis showed direct correlation with SDT, CE and pH variables, which shows surface runoff with possible water nitrification of a watershed planted with sugarcane. The second factor showed direct correlation between TCº and T variables, while both were negatively correlated with DO, which shows the seasonality effect on one of the watersheds reforested with Pinus.  The results highlight   management practices in  conservation and maintenance planning of forested areas surrounding headwaters as indicators for higher quality of natural water. Keywords: Pinus, sugar cane, nitrification, seasonal variation ZANATA, M.; PISSARRA, T. C. T.; FERRAUDO, A. S.; RANZINI, M.; CAMPOS, S.EFEITO DO USO DO SOLO NA QUALIDADE DO RECURSO HÍDRICO UTILIZANDO ANÁLISE ESTATÍSTICA MULTIVARIADA  2 RESUMO Mudanças no uso do solo ao longo das áreas protegidas modificam a paisagem, assim o monitoramento da qualidade das águas das nascentes torna-se essencial para melhorar a saúde da população e controlar a poluição dessas águas. O objetivo deste trabalho foi determinar alterações ambientais devido ao uso do solo, avaliando características físicas e químicas da água da rede de drenagem de microbacias. O trabalho foi realizado para estudar os efeitos do uso do solo em microbacias de primeira ordem, relacionado com práticas de gestão da qualidade da água, de acordo com a importância estratégica para a fonte de abastecimento; as microbacias são predominantemente ocupadas com cana de açúcar, reflorestadas com Pinus e com mata nativa. Os parâmetros, de temperatura (T), pH, condutividade elétrica (CE) e sólidos dissolvidos totais (SDT) foram determinados nas águas das nascentes. O oxigênio dissolvido (OD) e a turbidez (t) foram obtidos em laboratório. As variáveis e os fatores foram explicados por mudanças decorrentes do uso do solo associados às práticas de gestão. Dois fatores resultantes da análise multivariada, mostraram correlação direta com as variáveis SDT, CE e pH, evidenciando o escoamento superficial com possível nitrificação da água de uma microbacia plantada com cana de açúcar. O segundo fator mostrou correlação direta entre as variáveis T e t enquanto ambos estavam negativamente correlacionada com OD, mostrando o efeito da sazonalidade sobre uma das microbacias reflorestadas com Pinus. Os resultados destacam as práticas de gestão no planejamento da conservação e a manutenção de áreas florestais ao redor das nascentes, como indicadores para melhor qualidade da água natural. Palavras-chave: Pinus, cana de açúcar, nitrificação, sazonalidade.

RESUMOApesar da reconhecida importância da Mata Atlântica, há uma escassez de estudos utilizando o sensoriamento remoto como ferramenta para identificação e classificação dos diferentes estágios sucessionais de seus remanescentes florestais. Neste estudo comparamos o desempenho de diferentes métodos para classificação de estágio sucessional e investigamos a existência de sazonalidade na resposta espectral de uma floresta tropical densa na Mata Atlântica. Usamos amostras de treinamento de três estágios sucessionais obtidas a partir de uma ortofoto de 2010 e selecionamos imagens Landsat 5 TM para os anos de 2009, 2010 e 2011, considerando os meses de maiores e menores médias históricas de temperatura e precipitação. Para avaliação da sazonalidade da resposta espectral usamos o teste de Mann-Whitney, comparando cada banda do espectro eletromagnético e estágios sucessionais entre as épocas de aquisição das imagens. Para classificação da cobertura vegetal usamos três Índices de vegetação (NDVI, EVI e NDMI) e Análise Discriminante Quadrática (QDA). Comparamos a acurácia dos classificadores a partir de matrizes de validação cruzada. Nossos resultados mostram diferenças significativas entre os estágios sucessionais para todas as bandas espectrais, com melhor distinção na época de menores temperaturas e precipitação. QDA foi o classificador com maior acerto global (92%), seguido por NDMI (68%), NDVI (67%) e EVI (59%). Concluímos que QDA é, dentre os classificadores avaliados, o mais eficiente para classificação sucessional da floresta e que imagens obtidas em época de menor precipitação e temperatura geram uma melhor distinção entre estágios sucessionais para essa fisionomia florestal. PALAVRAS-CHAVE: Análise discriminante quadrática, Floresta tropical, Índices de vegetação. ABSTRACTDespite the recognized importance of the Atlantic Forest, there is a shortage of studies using remote sensing as a tool to identify and classify the different successional stages of its forest remnants. In this study we compared the performance of different classifiers on the determination of successional stages and investigated the existence of seasonality in the spectral response of a dense tropical Atlantic Forest area. We used training samples of three successional stages gathered by an orthophoto from the year of 2010 and selected Landsat 5 TM images for the years of 2009, 2010 and 2011, considering the months of higher and lower historical averages of temperature and precipitation. To evaluate the seasonality in forest spectral response we used the Mann-Whitney test, comparing each band and successional stage between the two studied periods. To classify the vegetation cover we used three indices (NDVI, EVI and NDMI) and a Quadratic Discriminant Analysis (QDA). We compared classifiers accuracy using a cross validation matrix. We found significative differences between successional stages on all electromagnetic spectral zones, with finer definitions between stages at the samples of colder and drier periods. QDA was the classifier with high...