7 research outputs found
Use of machine learning techniques for modeling of snow depth
Get PDFSnow exerts significant regulating effect on the land hydrological cycle since it controls intensity of heat and water exchange between the soil-vegetative cover and the atmosphere. Estimating of a spring flood runoff or a rain-flood on mountainous rivers requires understanding of the snow cover dynamics on a watershed. In our work, solving a problem of the snow cover depth modeling is based on both available databases of hydro-meteorological observations and easily accessible scientific software that allows complete reproduction of investigation results and further development of this theme by scientific community. In this research we used the daily observational data on the snow cover and surface meteorological parameters, obtained at three stations situated in different geographical regions: Col de Porte (France), Sodankyla (Finland), and Snoquamie Pass (USA).Statistical modeling of the snow cover depth is based on a complex of freely distributed the present-day machine learning models: Decision Trees, Adaptive Boosting, Gradient Boosting. It is demonstrated that use of combination of modern machine learning methods with available meteorological data provides the good accuracy of the snow cover modeling. The best results of snow cover depth modeling for every investigated site were obtained by the ensemble method of gradient boosting above decision trees – this model reproduces well both, the periods of snow cover accumulation and its melting. The purposeful character of learning process for models of the gradient boosting type, their ensemble character, and use of combined redundancy of a test sample in learning procedure makes this type of models a good and sustainable research tool. The results obtained can be used for estimating the snow cover characteristics for river basins where hydro-meteorological information is absent or insufficient
River runoff evaluation for ungauged watersheds by SWAP model. 2. Application of methods of physiographic similarity and spatial geostatistics
No full textRunoff evaluation for ungauged watersheds by SWAP model. 1. Application of artificial neural networks
No full textClimate change impact on streamflow in large-scale river basins: projections and their uncertainties sourced from GCMs and RCP scenarios
No full textClimate change impact on river runoff was investigated
within the framework of the second phase of the Inter-Sectoral Impact Model
Intercomparison Project (ISI-MIP2) using a physically-based land surface
model Soil Water – Atmosphere – Plants (SWAP) (developed in the Institute
of Water Problems of the Russian Academy of Sciences) and meteorological
projections (for 2006–2099) simulated by five General Circulation Models
(GCMs) (including GFDL-ESM2M, HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM-CHEM, and
NorESM1-M) for each of four Representative Concentration Pathway
(RCP) scenarios (RCP2.6, RCP4.5, RCP6.0, and RCP8.5). Eleven large-scale
river basins were used in this study. First of all, SWAP was calibrated and
validated against monthly values of measured river runoff with making use of
forcing data from the WATCH data set and all GCMs' projections were
bias-corrected to the WATCH. Then, for each basin, 20 projections of
possible changes in river runoff during the 21st century were simulated
by SWAP. Analysis of the obtained hydrological projections allowed us to
estimate their uncertainties resulted from application of different GCMs and
RCP scenarios. On the average, the contribution of different GCMs to the
uncertainty of the projected river runoff is nearly twice larger than the
contribution of RCP scenarios. At the same time the contribution of GCMs
slightly decreases with time
Impact of possible climate changes on river runoff under different natural conditions
No full textThe present study was carried out within the framework of the International
Inter-Sectoral Impact Model Intercomparison Project (ISI-MIP) for 11 large
river basins located in different continents of the globe under a wide
variety of natural conditions. The aim of the study was to investigate
possible changes in various characteristics of annual river runoff (mean
values, standard deviations, frequency of extreme annual runoff) up to 2100
on the basis of application of the land surface model SWAP and meteorological
projections simulated by five General Circulation Models (GCMs) according to
four RCP scenarios. Analysis of the obtained results has shown that changes
in climatic runoff are different (both in magnitude and sign) for the river
basins located in different regions of the planet due to differences in
natural (primarily climatic) conditions. The climatic elasticities of river
runoff to changes in air temperature and precipitation were estimated that
makes it possible, as the first approximation, to project changes in climatic
values of annual runoff, using the projected changes in mean annual air
temperature and annual precipitation for the river basins. It was found that
for most rivers under study, the frequency of occurrence of extreme runoff
values increases. This is true both for extremely high runoff (when the
projected climatic runoff increases) and for extremely low values (when the
projected climatic runoff decreases)
Применение методов машинного обучения для моделирования толщины снежного покрова
Get PDFSnow exerts significant regulating effect on the land hydrological cycle since it controls intensity of heat and water exchange between the soil-vegetative cover and the atmosphere. Estimating of a spring flood runoff or a rain-flood on mountainous rivers requires understanding of the snow cover dynamics on a watershed. In our work, solving a problem of the snow cover depth modeling is based on both available databases of hydro-meteorological observations and easily accessible scientific software that allows complete reproduction of investigation results and further development of this theme by scientific community. In this research we used the daily observational data on the snow cover and surface meteorological parameters, obtained at three stations situated in different geographical regions: Col de Porte (France), Sodankyla (Finland), and Snoquamie Pass (USA).Statistical modeling of the snow cover depth is based on a complex of freely distributed the present-day machine learning models: Decision Trees, Adaptive Boosting, Gradient Boosting. It is demonstrated that use of combination of modern machine learning methods with available meteorological data provides the good accuracy of the snow cover modeling. The best results of snow cover depth modeling for every investigated site were obtained by the ensemble method of gradient boosting above decision trees – this model reproduces well both, the periods of snow cover accumulation and its melting. The purposeful character of learning process for models of the gradient boosting type, their ensemble character, and use of combined redundancy of a test sample in learning procedure makes this type of models a good and sustainable research tool. The results obtained can be used for estimating the snow cover characteristics for river basins where hydro-meteorological information is absent or insufficient.На основе открытых данных гидрометеорологических наблюдений на трёх водно-балансовых стационарах, расположенных в различных физико-географических условиях, исследована возможность применения современных методов машинного обучения для моделирования динамики снежного покрова. Эффективность использования ансамблевой модели градиентного бустинга над решающими деревьями выше, чем моделей одиночного решающего дерева или адаптивного бустинга для всех исследуемых объектов
Implementation of an ensemble of hydrological models to study the impact of climate variability and change on water resources in páramo ecosystems in Colombia. Case study, Chuza watershed, Chingaza Páramo
No full textilustraciones, diagramas, fotografías, mapas, tablasA través del tiempo, la modelación hidrológica ha sido utilizada con diferentes propósitos,
entre ellos, la evaluación del impacto de escenarios de cambio climático en los recursos
hídricos de una cuenca. Sin embargo, lo usual ha sido emplear un único modelo
hidrológico, lo cual puede derivar en errores y sesgos debido a su incertidumbre
estructural. Por ello, en el presente estudio se realizó la evaluación de los posibles
impactos del cambio climático en la cuenca del río Chuza en el páramo de Chingaza, parte
de uno de los sistemas de abastecimiento de agua potable de la ciudad de Bogotá. Esta
evaluación se realizó a través de la construcción de un ensamble hidrológico por la
metodología de Granger Ramanathan en su variante C. Los escenarios de cambio
climático empleados se obtuvieron de proyecciones de dos GCMs: CMCC-ESM2 y ECEarth3-CC, y se construyeron con la metodología delta change para las trayectorias SSP2-
4.5 y SSP5-8.5. Como resultados, en primer lugar, se obtuvo que el ensamble construido
mejora las métricas individuales encontradas, tanto para el periodo de calibración como de
validación. Respecto a los posibles impactos del cambio climático, se observa que la
trayectoria SSP2-4.5 proyecta caudales mayores que la trayectoria SSP5-8.5, y que para
el 25% de los escenarios futuros el caudal disminuye, especialmente para la trayectoria
menos optimista. Los resultados obtenidos representan solamente un posible futuro acerca
del posible riesgo de disminución de caudales que podría afectar el abastecimiento de la
ciudad de Bogotá en el largo plazo (Texto tomado de la fuente).Historically, hydrologic modeling has served different purposes, among them, the
assessment of the impact of climate change scenarios. However, a single hydrological
model is usually employed, which can lead to errors and biases due to the structural
uncertainty of the model. Therefore, the present study has evaluated the possible impacts
of climate change in the Chuza river basin in the Chingaza páramo, part of one of the water
supply systems of the city of Bogotá. This through the construction of a hydrological
ensemble using the Granger Ramanathan methodology in its variant C. The climate change
scenarios were obtained from projections of two GCMs: CMCC-ESM2 and EC-Earth3-CC
and constructed with the delta change methodology for the SSP2-4.5 and SSP5-8.5
trajectories. First, it was obtained that the constructed ensemble improves the metrics
obtained for individual models for both the calibration and validation periods. Regarding the
possible impacts of climate change, it was observed that the SSP2-4.5 trajectory projects
higher flows than the SSP5-8.5 trajectory, and that for 25% of the future scenarios the flow
decreases, especially for the less optimistic trajectory. The results obtained represent only
a possible future about the potential risk of flow decrease that could affect the water supply
of the city of Bogota in the long term.MaestríaMagíster en Ingeniería - Recursos HidráulicosLa metodología se planteó en cuatro fases, representadas en la Figura 3-1, alineadas con
los objetivos presentados en la introducción y guiadas por lo planteado en el Protocolo de
modelación hidrológica e hidráulica propuesto por IDEAM (2018) y en lo propuesto por
Beven (2012). De igual manera, en la Figura 3-2 se presenta la metodología secuencial
que rigió lo desarrollado en el presente documento.Hidrología y meteorologí
