Votre recherche
Résultats 29 ressources
-
Abstract Floods are the most common and threatening natural risk for many countries in the world. Flood risk mapping is therefore of great importance for managing socio-economic and environmental impacts. Several researchers have proposed low-complexity and cost-effective flood mapping solutions that are useful for data scarce environments or at large-scale. Among these approaches, a line of recent research focuses on hydrogeomorphic methods that, due to digital elevation models (DEMs), exploit the causality between past flood events and the hydraulic geometry of floodplains. This study aims to compare the use of freely-available DEMs to support an advanced hydrogeomorphic method, Geomorphic Flood Index (GFI), to map flood-prone areas of the Basento River basin (Italy). The five selected DEMs are obtained from different sources, are characterized by different resolutions, spatial coverage, acquisition process, processing and validation, etc., and include: (i) HydroSHEDS v.1.1 (resolution 3 arc-seconds), hydrologically conditioned, derived primarily from STRM (NASA) and characterized by global coverage; (ii) ASTER GDEM v.3 with a res. of around 30 m (source: METI and NASA) and global coverage; (iii) EU-DEM v. 1.1 (res. 1 arc-second), Pan-European and combining SRTM and ASTER GDEM, customized to obtain a consistency with the EU-Hydro and screened to remove artefacts (source: Copernicus Land Monitoring Service); (iv) TinItaly DEM v. 1.1, (res. 10 m-cell size grid) and produced and distributed by INGV with coverage of the entire Italian territory; (v) Laser Scanner DEM with high resolution (5 m cell size grid) produced on the basis of Ground e Model Keypoint and available as part of the RSDI geoportal of the Basilicata Region with coverage at the regional administrative level. The effects of DEMs on the performance of the GFI calibration on the main reach of the Basento River, and its validation on one of its mountain tributaries (Gallitello Creek), were evaluated with widely accepted statistical metrics, i.e., the Area Under the Receiver Operating Characteristics (ROC) curve (AUC), Accuracy, Sensitivity and Specificity. Results confirmed the merits of the GFI in flood mapping using simple watershed characteristics and showed high Accuracy (AUC reached a value over 0.9 in all simulations) and low dependency on changes in the adopted DEMs and standard flood maps (1D and 2D hydraulic models or three return periods). The EU-DEM was identified as the most suitable data source for supporting GFI mapping with an AUC > 0.97 in the calibration phase for the main river reach. This may be due in part to its appropriate resolution for hydrological application but was also due to its customized pre-processing that supported an optimal description of the river network morphology. Indeed, EU-DEM obtained the highest performances (e.g., Accuracy around 98%) even in the validation phase where better results were expected from the high-resolution DEM (due to the very small size of Gallitello Creek cross-sections). For other DEMs, GFI generally showed an increase in metrics performance when, in the calibration phase, it neglected the floodplains of the river delta, where the standard flood map is produced using a 2D hydraulic model. However, if the DEMs were hydrologically conditioned with a relatively simple algorithm that forced the stream flow in the main river network, the GFI could be applied to the whole Basento watershed, including the delta, with a similar performance.
-
Abstract Reference evapotranspiration (ETo) is one of the most important factors in the hydrologic cycle and water balance studies. In this study, the performance of three simple and three wavelet hybrid models were compared to estimate ETo in three different climates in Iran, based on different combinations of input variables. It was found that the wavelet-artificial neural network was the best model, and multiple linear regression (MLR) was the worst model in most cases, although the performance of the models was related to the climate and the input variables used for modeling. Overall, it was found that all models had good accuracy in terms of estimating daily ETo. Also, it was found in this study that large numbers of decomposition levels via the wavelet transform had noticeable negative effects on the performance of the wavelet-based models, especially for the wavelet-adaptive network-based fuzzy inference system and wavelet-MLR, but in contrast, the type of db wavelet function did not have a detectable effect on the performance of the wavelet-based models.
-
Abstract A quantitative and qualitative understanding of the anticipated climate-change-driven multi-scale spatio-temporal shifts in precipitation and attendant river flows is crucial to the development of water resources management approaches capable of sustaining and even improving the ecological and socioeconomic viability of rain-fed agricultural regions. A set of homogeneity tests for change point detection, non-parametric trend tests, and the Sen’s slope estimator were applied to long-term gridded rainfall records of 27 newly formed districts in Chhattisgarh State, India. Illustrating the impacts of climate change, an analysis of spatial variability, multi-temporal (monthly, seasonal, annual) trends and inter-annual variations in rainfall over the last 115 years (1901–2015 mean 1360 mm·y −1 ) showed an overall decline in rainfall, with 1961 being a change point year (i.e., shift from rising to declining trend) for most districts in Chhattisgarh. Spatio-temporal variations in rainfall within the state of Chhattisgarh showed a coefficient of variation of 19.77%. Strong inter-annual and seasonal variability in regional rainfall were noted. These rainfall trend analyses may help predict future climate scenarios and thereby allow planning of effective and sustainable water resources management for the region.
-
Abstract Analyzing intra-annual stream flow can reveal the main causes for runoff changes and the contributions of climate variability and human activities. For this purpose, the Mann–Kendall and cumulative rank difference (CRD) tests, and the double mass curve method, were applied to a time series of hydro-meteorological variables from 1971 to 2010 in the Tajan River basin in Iran. Results indicated that runoff changes in the wet and dry seasons after 1999 had significant respective decreasing and increasing trends, at the 0.01 confidence level, due to dam construction. In the pre-dam period (1991–1998), the results of the double mass curve method showed that climate variability and human activities contributed 57.76% and 42.24%, respectively, to the runoff decrease during the wet season. For the post-dam period (1999–2010), climate variability and anthropogenic activities contributed 24.68% and 75.32%, respectively, to the wet season runoff decrease of 116.55 mm. On the other hand, in the same period during the dry season, climate variability contributed −30.68% and human activities contributed 130.68% to the runoff increase of 41.45 mm. It is evident that runoff changes in both wet and dry seasons were mainly due to human activities associated with dam construction to meet water supply demands for agriculture.