Vazões líquidas e sólidas em bacias hidrográficas sob diferentes cenários ambientais no Sul da Amazônia

• Autor(a) principal: Borella, Daniela Roberta
• Data de Publicação: 2022
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFMT
• Texto Completo: http://ri.ufmt.br/handle/1/5539
• topic: Hidrologia; Hidrossedimentologia; Rio Teles Pires; Fronteira agrícola brasileira; Transição Cerrado-Amazônia; CNPQ::CIENCIAS EXATAS E DA TERRA::FISICA; Hydrology; Hyidrossedimentology; Teles Pires river; Brazilian agricultural frontier; Cerrado-Amazon transition

The increased demand for the natural resources of water and soil in the transition region of the Cerrado-Amazon biomes has stimulated the need to know water availability and the dynamics of hydro-sedimentological processes for proper management of water resources. Since, the use and occupation of the land without planning have generated great losses of soil by water erosion, increase of the load of sediments in the rivers, water pollution and reduction of the surface drainage. In this context, the present study evaluated: I - net flows (QL) obtained by traditional and acoustic methods by Doppler Effect; II - solid flows (Qs) collected by the method of equal increment of width with samples of solids in suspension (Qss) by vertical, composite, standard vertical and percentiles, and, III - determined the physiographic characteristics and established key curves for the flows liquid and solid. QL measurements and Qs collections were carried out monthly between 2018 and 2021 to represent the water seasonality and the hydrosedimentological dynamics of the hydrological sections of the Caiabi 1 and 2, Celeste, Nandico, Preto, Renato 1 and 2 rivers and Rosana stream, tributaries of the river Teles Pires. QL were obtained by MLN-7 hydrometric windlass and Doppler Acoustic Current Profiler (ADCP RiverRay), while Qs were obtained by suspended and drag sediment samplers, and total solid flows (Qst) were calculated. Regression models were adjusted to verify the agreement of the QL obtained between reel and ADCP, of the depths between graduated metal rod and ADCP, for estimates of Qss and Qst, and for the liquid and solid key curves. Statistical performances were evaluated by mean relative error (MBE), root mean relative error (RMSE), Willmott concordance index (dw), NashSutcliffe efficiency (NSE) and test of means. Physiographic characteristics were determined using QGIS software version 3.16.11. The QL obtained by windlass and ADCP showed similar dynamics in the water seasonality and in each hydrological section, with maximum and minimum peaks between the rainy and dry periods of the region, respectively. The QL ranged from 3.5 to 61.0 m3 s -1 for the windlass and 2.7 to 54.0 m3 s -1 for the ADCP, between hydrological sections, with dw of 0.9998 for QL and greater than 0.9995 for depth. The Qs showed similar dynamics between the sampling methods, following the maximum and minimum peaks of the QL in each hydrological section. The Qss and Qst ranged from 0.31 to 39.35 t d-1 and from 0.32 to 43.70 t d-1 , respectively, between the dry and rainy periods of the region and between the hydrological sections. As for the Qs estimates, the dw values were greater than 0.9815 for Qss and 0.9975 for Qst. The Renato River sub-basin had a regular drainage density = 0.505 km km-2 , an elongated shape (Kf = 0.19; Kc = 2.17 and Ic = 0.21) and the relief varied from flat to strongly undulating (0 to 30%). The regression models (linear, polynomial, potential and exponential) adjusted for QL showed satisfactory statistical performance, with dw between 0.8999 and 0.9599 and NSE between 0.60 and 0.83, while for Qs the models did not adjust in a satisfactory way. Measurements by the traditional method and ADCP were compatible to obtain QL in shallow rivers (up to 6.5 m deep) in the Teles Pires river basin. The dynamics of sediment production and transport was influenced by land use and occupation, drainage area and water seasonality in the region. The different ways of sampling suspended sediments are compatible to obtain suspended and total solid flow rates. The Renato river hydrographic sub-basin is little susceptible to the occurrence of floods, due to its elongated shape and regular drainage density. The regression models adjusted satisfactorily for QL and unsatisfactorily for Qss and Qst.