Search

El empleo de sensores remotos junto a la información de los factores sociales propios de cada población permite el monitoreo del uso de los recursos naturales. El objetivo del presente estudio fue determinar el cambio y proyección a futuro de los usos y coberturas de suelo, y a la vez comprender, desde la perspectiva de los principales actores, los factores propulsores y subyacentes que impulsan estos cambios en la cuenca hidrográfica media-alta del río Mira. Para ello, se utilizaron imágenes multiespectrales, Landsat y Sentinel del año 1996, 2007 y 2018, a las cuales se las realizó un pretratamiento y tratamiento. Se efectuó una proyección de los cambios de coberturas y usos del suelo del 2018-2030 mediante el software TerrSet. Después, utilizando el método Delphi se identificaron los factores propulsores y subyacentes. Los resultados encontrados muestran que bosques y pastos presentaron una disminución sostenida, mientras que las áreas de cultivo y zonas urbanas aumentaron dentro de los periodos 1996-2018 y 2018-2030. Estos cambios se relacionan con el crecimiento urbano, agrícola, ganadero, minero y de la industria forestal; impulsados generalmente por el aumento poblacional, demanda de mercado, cambios de patrones de consumo, expansión de las carreteras e inexistencia de políticas ambientales., The use of remote sensing and information on social factors specific to each population allows the monitoring of natural resource use. The main objective of this study is to determine the change and future projection of land use and land cover and, at the same time, to understand, from the perspective of the main stakeholders, the driving and underlying factors of these changes in the middle-upper Mira river basin. For this purpose, multispectral, Landsat, and Sentinel images from 1996, 2007, and 2018 have been used and pre-processed and treated. Subsequently, a forecast of land cover and land use changes from 2018-2030 has been made using TerrSet software. The drivers and underlying factors have been identified using the Delphi method. The results show that forests and pastures presented a sustained decrease, while cropland and urban areas increased between 1996-2018 and 2018-2030. These changes are related to urban, agricultural, livestock, mining, and forest industry growth, which are generally driven by population growth, market demand, changing consumption patterns, road expansion, and a lack of environmental policies.

Extreme peak runoff forecasting is still a challenge in hydrology. In fact, the use of traditional physically-based models is limited by the lack of sufficient data and the complexity of the inner hydrological processes. Here, we employ a Machine Learning technique, the Random Forest (RF) together with a combination of Feature Engineering (FE) strategies for adding physical knowledge to RF models and improving their forecasting performances. The FE strategies include precipitation-event classification according to hydrometeorological criteria and separation of flows into baseflow and directflow. We used ∼ 3.5 years of hourly precipitation information retrieved from two near-real-time satellite precipitation databases (PERSIANN-CCS and IMERG-ER), and runoff data at the outlet of a 3391-km2 basin located in the tropical Andes of Ecuador. The developed models obtained Nash-Sutcliffe efficiencies varying from 0.86 to 0.59 for lead times between 1 and 6 h. The best performances were obtained for peak runoffs triggered by short-extension precipitation events (<50 km2) where infiltration- or saturation-excess runoff responses are well learned by the RF models. Conversely, the forecasting difficulty is associated with extensive precipitation events. For such conditions, a deeper characterization of the biophysical characteristics of the basin is encouraged for capturing the dynamic of directflow across multiple runoff responses. All in all, the potential to employ near-real-time satellite precipitation and the use of FE strategies for improving RF forecasting provides hydrologists with new tools for real-time runoff forecasting in remote or complex regions., Water Resources

La erosión del suelo es uno de los principales problemas ambientales a nivel global, la cual causa pérdida de la fertilidad del suelo y aumento en la producción de sedimentos. El objetivo del presente estudio es cuantificar y conocer la distribución espacial de la erosión hídrica en las diferentes coberturas de suelo en la cuenca media-alta del río Mira en Ecuador. Para ello se aplicó el modelo RUSLE en entorno SIG. Los resultados determinaron que la erosión varía entre 0 y 812,30 t/ha/año y que la erosión promedio es 32,79 t/ha/año. En el área de estudio únicamente las coberturas bosques y páramo presentaron una erosión media menor al umbral sostenible de erosión con valores inferiores a 10 t/ha/año. Por su parte, las áreas sin vegetación, y cultivos son las coberturas con más propensión tasas de erosión del suelo elevadas. Esto demuestra que la erosión dentro del área de estudio se encuentra entre las más altas del mundo. En consecuencia, se recomiendan alternativas para el manejo de uso de suelo en la cuenca hidrográfica para los tomadores de decisiones., Soil erosion is one of the leading environmental problems at a global level, which causes loss of soil fertility and an increase in the production of sediments. This study aims to quantify and understand the spatial distribution of water erosion over land cover in the upper-middle basin of the Mira River in Ecuador. The RUSLE model has been applied using remote sensing tools and GIS tools. The results show erosion varies between 0 and 812.30 t/ha/year, and the average erosion is 32.79 t/ha/year. The forest and paramo are the unique covers that present an average erosion lower than the sustainable threshold of erosion with values lower than 10 t/ha/year in the study area. On the other hand, areas without vegetation and crops have the highest propensity for high soil erosion rates. It shows that erosion within the study area is among the highest in the world. Consequently, decision-makers are recommended to seek alternatives for land use management in the watershed.

The Random Forest (RF) algorithm, a decision-tree-based technique, has become a promising approach for applications addressing runoff forecasting in remote areas. This machine learning approach can overcome the limitations of scarce spatio-temporal data and physical parameters needed for process-based hydrological models. However, the influence of RF hyperparameters is still uncertain and needs to be explored. Therefore, the aim of this study is to analyze the sensitivity of RF runoff forecasting models of varying lead time to the hyperparameters of the algorithm. For this, models were trained by using (a) default and (b) extensive hyperparameter combinations through a grid-search approach that allow reaching the optimal set. Model performances were assessed based on the R2, %Bias, and RMSE metrics. We found that: (i) The most influencing hyperparameter is the number of trees in the forest, however the combination of the depth of the tree and the number of features hyperparameters produced the highest variability-instability on the models. (ii) Hyperparameter optimization significantly improved model performance for higher lead times (12- and 24-h). For instance, the performance of the 12-h forecasting model under default RF hyperparameters improved to R2 = 0.41 after optimization (gain of 0.17). However, for short lead times (4-h) there was no significant model improvement (0.69 < R2 < 0.70). (iii) There is a range of values for each hyperparameter in which the performance of the model is not significantly affected but remains close to the optimal. Thus, a compromise between hyperparameter interactions (i.e., their values) can produce similar high model performances. Model improvements after optimization can be explained from a hydrological point of view, the generalization ability for lead times larger than the concentration time of the catchment tend to rely more on hyperparameterization than in what they can learn from the input data. This insight can help in th