Entity

Time filter

Source Type


Melchiorre C.,Geological Survey of Norway | Castellanos Abella E.A.,Institute of Geology and Palaeontology IGP | van Westen C.J.,International Institute for Geoinformation Science and Earth Observation | Matteucci M.,Polytechnic of Milan
Computers and Geosciences | Year: 2011

This paper describes a procedure for landslide susceptibility assessment based on artificial neural networks, and focuses on the estimation of the prediction capability, robustness, and sensitivity of susceptibility models. The study is carried out in the Guantanamo Province of Cuba, where 186 landslides were mapped using photo-interpretation. Twelve conditioning factors were mapped including geomorphology, geology, soils, landuse, slope angle, slope direction, internal relief, drainage density, distance from roads and faults, rainfall intensity, and ground peak acceleration.A methodology was used that subdivided the database in 3 subsets. A training set was used for updating the weights. A validation set was used to stop the training procedure when the network started losing generalization capability, and a test set was used to calculate the performance of the network. A 10-fold cross-validation was performed in order to show that the results are repeatable. The prediction capability, the robustness analysis, and the sensitivity analysis were tested on 10 mutually exclusive datasets. The results show that by means of artificial neural networks it is possible to obtain models with high prediction capability and high robustness, and that an exploration of the effect of the individual variables is possible, even if they are considered as a black-box model. © 2010 Elsevier Ltd. Source


Franco R.D.V.,RDVGeoconsulting | Thouret J.-C.,CNRS Magmas and Volcanoes Laboratory | Delaite G.,CNRS Magmas and Volcanoes Laboratory | Van Westen C.,International Institute for Geoinformation Science and Earth Observation | And 5 more authors.
Special Paper of the Geological Society of America | Year: 2010

Studies of the type, extent, and volume of Holocene pyroclastic and lahar deposits have concluded that future eruptions of El Misti volcano, even if moderate in magnitude, will pose a serious threat to the city of Arequipa, Peru. After describing the most probable volcanic scenarios at El Misti, this paper concentrates on lahar and flood risk assessment. Scenarios were derived with the help of the simulation codes LAHARZ and TITAN2D. The lahar risk assessment varies significantly depending on the method selected. LAHARZ simulations indicate that a considerable part of the urban areas and infrastructure could be severely affected. Losses due to impacts inflicted by lahars in three selected parts of the urban area are estimated to be in the order of 40-100 million U.S. dollars. In the case of TITAN2D, the resulting laharaffected area only includes infrastructure assets mainly located along the Río Chili. Results indicate that although simulation codes could be useful tools in the analysis of lahar hazard scenarios, it is still premature to regard them as accurate sources of information for actual decision making related to risk mitigation at the local level. More research is required to further adjust simulation codes and refine risk scenarios. The first priority for the mitigation of the volcanic hazard faced by the city of Arequipa should be improvement of the risk map (a hazard map has already been drawn and is under scrutiny) and the preparation of contingency plans. © 2010 The Geological Society of America. All rights reserved. Source


Omo-Irabor O.O.,West African Management Coastal Initiative WACMI | Omo-Irabor O.O.,Delta State University, Abraka | Omo-Irabor O.O.,Contemporary University | Olobaniyi S.B.,West African Management Coastal Initiative WACMI | And 10 more authors.
Environmental Monitoring and Assessment | Year: 2011

Mangroves are known for their global environmental and socioeconomic value. Despite their importance, mangrove like other ecosystems is now being threatened by natural and human-induced processes that damage them at alarming rates, thereby diminishing the limited number of existing mangrove vegetation. The development of a spatial vulnerability assessment model that takes into consideration environmental and socioeconomic criteria, in spatial and non-spatial formats has been attempted in this study. According to the model, 11 different input parameters are required in modelling mangrove vulnerability. These parameters and their effects on mangrove vulnerability were selected and weighted by experts in the related fields. Criteria identification and selection were mainly based on effects of environmental and socioeconomic changes associated with mangrove survival. The results obtained revealed the dominance of socioeconomic criteria such as population pressure and deforestation, with high vulnerability index of 0.75. The environmental criteria was broadly dispersed in the study area and represents vulnerability indices ranging from 0.00-0.75. This category reflects the greater influence of pollutant input from oil wells and pipelines and minimal contribution from climatic factors. This project has integrated spatial management framework for mangrove vulnerability assessment that utilises information technology in conjunction with expert knowledge and multi-criteria analysis to aid planners and policy/ decision makers in the protection of this very fragile ecosystem. © 2010 Springer Science+Business Media B.V. Source


Bamutaze Y.,Makerere University | Tenywa M.M.,Makerere University | Majaliwa M.J.G.,Makerere University | Vanacker V.,University of Strasbourg | And 4 more authors.
Catena | Year: 2010

Water infiltration is an important hydrological process that influences runoff and soil loss patterns in mountain ecosystems. In this paper, we present results on spatial variation in infiltration in croplands on the volcanic soils of Mt. Elgon, in Eastern Uganda. Twelve experimental sites with slope gradients ranging from 12 to 32% were established. Infiltration tests were carried out with a double ring infiltrometer and three measurements were taken at the upper, middle and lower sections of each experimental site to assess the local variability of infiltration. In addition soil information was collected on each experimental site. The soil infiltration data were then evaluated to fit to four commonly used water infiltration models: (1) Philip (1957), (2) Green-Ampt (1911), (3) Horton (1940) and (4) Kostiakov (1932). The twelve experimental sites cover two cropping systems: annual (6 sites) and perennial (6 sites) crops. Based on the results, we examine the spatial variability of infiltration, the relationship of infiltration to landscape position, and the influence of soil composition on infiltration rates on the slopes. The factors affecting spatial variability of soil infiltration were analysed using correlation and regression techniques. Steady state infiltration rates generally increased with the slope gradient and were crop type independent (P < 0.05). The performance of the four applied water infiltration models was generally good with mean R2 values ranging from 0.79 to 0.87, although all the models tended to over-predict the steady state infiltration rates at most sites. Overall, the Philip's and Kostiakov gave better results than the Horton and Green-Ampt models in reproducing the infiltration process on Mt. Elgon. © 2009 Elsevier B.V. All rights reserved. Source


Knox N.M.,International Institute for Geoinformation Science and Earth Observation | Skidmore A.K.,International Institute for Geoinformation Science and Earth Observation | Schlerf M.,International Institute for Geoinformation Science and Earth Observation | de Boer W.F.,Wageningen University | And 4 more authors.
International Journal of Remote Sensing | Year: 2010

We analysed stability and predictive capabilities of known nitrogen absorption features between plant material prepared for NIRS (dried) and RS (fresh) studies. Grass spectra were taken of the plant canopy, and again after the grass sample was dried and ground. Models were derived using stepwise multiple linear regression (sMLR). Regression values (adj.r2) produced using the dried material were greater than those produced using canopy material. For dried material only wavebands from the SWIR region were selected. Wavebands selected by sMLR on canopy material were located in both the VNIR and SWIR regions. Using wavebands selected for dried material models produced low adj.r2 values when applied to canopy plant material; differences in adj.r2 values are smaller when wavebands selected in canopy material models are applied to dried material. Widening of nitrogen features produced higher adj.r2 values for both dried and canopy material. This work shows that obtaining models with high predictive capabilities for nitrogen concentration is possible, but waveband selection should not be limited to features identified by NIRS studies. To accommodate for variability in absorption features, and instrument errors, absorption features should be widened. © 2010 Taylor & Francis. Source

Discover hidden collaborations