%0 Journal Article %1 2011AGUFM.H11B1052M %A Molina, J. M. %A Ramirez, J. A. %A Raff, D. A. %B AGU Fall Meeting Abstracts %D 2011 %I American Geophysical Union %J AGU Fall Meeting Abstracts %K Climate ENSO Flows HYDROLOGY Modeling NOAA Ocean/atmosphere PDO Stochastic USA generation myown streamflow %P H11B-1052 %T A Multivariate Downscaling Model for Nonparametric Simulation of Daily Flows %V 2011 %X A multivariate, stochastic nonparametric framework for stepwise disaggregation of seasonal runoff volumes to daily streamflow is presented. The downscaling process is conditional on volumes of spring runoff and large-scale ocean-atmosphere teleconnections and includes a two-level cascade scheme: seasonal-to-monthly disaggregation first followed by monthly-to-daily disaggregation. The non-parametric and assumption-free character of the framework allows consideration of the random nature and nonlinearities of daily flows, which parametric models are unable to account for adequately. This paper examines statistical links between decadal/interannual climatic variations in the Pacific Ocean and hydrologic variability in US northwest region, and includes a periodicity analysis of climate patterns to detect coherences of their cyclic behavior in the frequency domain. We explore the use of such relationships and selected signals (e.g., north Pacific gyre oscillation, southern oscillation, and Pacific decadal oscillation indices, NPGO, SOI and PDO, respectively) in the proposed data-driven framework by means of a combinatorial approach with the aim of simulating improved streamflow sequences when compared with disaggregated series generated from flows alone. A nearest neighbor time series bootstrapping approach is integrated with principal component analysis to resample from the empirical multivariate distribution. A volume-dependent scaling transformation is implemented to guarantee the summability condition. In addition, we present a new and simple algorithm, based on nonparametric resampling, that overcomes the common limitation of lack of preservation of historical correlation between daily flows across months. The downscaling framework presented here is parsimonious in parameters and model assumptions, does not generate negative values, and produces synthetic series that are statistically indistinguishable from the observations. We present evidence showing that both including conditional information of climatic teleconnection signals and developing the disaggregation in cascades decrease significantly the mean error between simulated and observed flow traces. The modeling procedure is tested with data from the Payette River basin in Idaho.

@article{2011AGUFM.H11B1052M, abstract = {A multivariate, stochastic nonparametric framework for stepwise disaggregation of seasonal runoff volumes to daily streamflow is presented. The downscaling process is conditional on volumes of spring runoff and large-scale ocean-atmosphere teleconnections and includes a two-level cascade scheme: seasonal-to-monthly disaggregation first followed by monthly-to-daily disaggregation. The non-parametric and assumption-free character of the framework allows consideration of the random nature and nonlinearities of daily flows, which parametric models are unable to account for adequately. This paper examines statistical links between decadal/interannual climatic variations in the Pacific Ocean and hydrologic variability in US northwest region, and includes a periodicity analysis of climate patterns to detect coherences of their cyclic behavior in the frequency domain. We explore the use of such relationships and selected signals (e.g., north Pacific gyre oscillation, southern oscillation, and Pacific decadal oscillation indices, NPGO, SOI and PDO, respectively) in the proposed data-driven framework by means of a combinatorial approach with the aim of simulating improved streamflow sequences when compared with disaggregated series generated from flows alone. A nearest neighbor time series bootstrapping approach is integrated with principal component analysis to resample from the empirical multivariate distribution. A volume-dependent scaling transformation is implemented to guarantee the summability condition. In addition, we present a new and simple algorithm, based on nonparametric resampling, that overcomes the common limitation of lack of preservation of historical correlation between daily flows across months. The downscaling framework presented here is parsimonious in parameters and model assumptions, does not generate negative values, and produces synthetic series that are statistically indistinguishable from the observations. We present evidence showing that both including conditional information of climatic teleconnection signals and developing the disaggregation in cascades decrease significantly the mean error between simulated and observed flow traces. The modeling procedure is tested with data from the Payette River basin in Idaho.}, added-at = {2020-09-04T19:47:43.000+0200}, adsnote = {Provided by the SAO/NASA Astrophysics Data System}, adsurl = {https://ui.adsabs.harvard.edu/abs/2011AGUFM.H11B1052M}, author = {Molina, J. M. and Ramirez, J. A. and Raff, D. A.}, biburl = {https://www.bibsonomy.org/bibtex/2e368ae46be07381cb69964c180d4f31a/jomolina10}, booktitle = {AGU Fall Meeting Abstracts}, eid = {H11B-1052}, interhash = {6b854ed00e4f625f4fcb65ee103db3ca}, intrahash = {e368ae46be07381cb69964c180d4f31a}, journal = {AGU Fall Meeting Abstracts}, keywords = {Climate ENSO Flows HYDROLOGY Modeling NOAA Ocean/atmosphere PDO Stochastic USA generation myown streamflow}, month = dec, pages = {H11B-1052}, publisher = {American Geophysical Union}, timestamp = {2020-09-10T22:26:04.000+0200}, title = {A Multivariate Downscaling Model for Nonparametric Simulation of Daily Flows}, volume = 2011, year = 2011 }