Paul Muñoz

Dr. Paul Muñoz

Universidad de Cuenca
orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 August 2021 in Science of The Total Environment
Verified
Journal Article
Closed access
David F. MuñozDavid F. Muñoz
David F. Muñoz
Paul MuñozPaul Muñoz
Paul Muñoz
HMHamed Moftakhari
Hamed Moftakhari
Hamid MoradkhaniHamid Moradkhani
Hamid Moradkhani

26 October 2018 in Water
Verified
Journal Article
Open access
PDF file
Paul MuñozPaul Muñoz
Paul Muñoz
JOJohanna Orellana-Alvear
Johanna Orellana-Alvear
PWPatrick Willems
Patrick Willems
RCRolando Célleri
Rolando Célleri
Flash-flood forecasting has emerged worldwide due to the catastrophic socio-economic impacts this hazard might cause and the expected increase of its frequency in the future. In mountain catchments, precipitation-runoff forecasts are limited by the intrinsic complexity of the processes involved, particularly its high rainfall variability. While process-based models are hard to implement, there is a potential to use the random forest algorithm due to its simplicity, robustness and capacity to deal with complex data structures. Here a step-wise methodology is proposed to derive parsimonious models accounting for both hydrological functioning of the catchment (e.g., input data, representation of antecedent moisture conditions) and random forest procedures (e.g., sensitivity analyses, dimension reduction, optimal input composition). The methodology was applied to develop short-term prediction models of varying time duration (4, 8, 12, 18 and 24 h) for a catchment representative of the Ecuadorian Andes. Results show that the derived parsimonious models can reach validation efficiencies (Nash-Sutcliffe coefficient) from 0.761 (4-h) to 0.384 (24-h) for optimal inputs composed only by features accounting for 80% of the model’s outcome variance. Improvement in the prediction of extreme peak flows was demonstrated (extreme value analysis) by including precipitation information in contrast to the use of pure autoregressive models.

10 February 2021 in Atmosphere
Verified
Journal Article
Open access
PDF file
PCPablo Contreras
Pablo Contreras
JOJohanna Orellana-Alvear
Johanna Orellana-Alvear
Paul MuñozPaul Muñoz
Paul Muñoz
JBJörg Bendix
Jörg Bendix
RCRolando Célleri
Rolando Célleri
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 the development of operational early warning systems.

16 December 2021 in Hydrology
Verified
Journal Article
Open access
PDF file
Paul MuñozPaul Muñoz
Paul Muñoz
JOJohanna Orellana-Alvear
Johanna Orellana-Alvear
JBJörg Bendix
Jörg Bendix
JFJan Feyen
Jan Feyen
RCRolando Célleri
Rolando Célleri
Worldwide, machine learning (ML) is increasingly being used for developing flood early warning systems (FEWSs). However, previous studies have not focused on establishing a methodology for determining the most efficient ML technique. We assessed FEWSs with three river states, No-alert, Pre-alert and Alert for flooding, for lead times between 1 to 12 h using the most common ML techniques, such as multi-layer perceptron (MLP), logistic regression (LR), K-nearest neighbors (KNN), naive Bayes (NB), and random forest (RF). The Tomebamba catchment in the tropical Andes of Ecuador was selected as a case study. For all lead times, MLP models achieve the highest performance followed by LR, with f1-macro (log-loss) scores of 0.82 (0.09) and 0.46 (0.20) for the 1 h and 12 h cases, respectively. The ranking was highly variable for the remaining ML techniques. According to the g-mean, LR models correctly forecast and show more stability at all states, while the MLP models perform better in the Pre-alert and Alert states. The proposed methodology for selecting the optimal ML technique for a FEWS can be extrapolated to other case studies. Future efforts are recommended to enhance the input data representation and develop communication applications to boost the awareness of society of floods.

16 November 2016 in Water
Verified
Journal Article
Open access
PDF file
Paul MuñozPaul Muñoz
Paul Muñoz
RCRolando Célleri
Rolando Célleri
JFJan Feyen
Jan Feyen
A laser-optical disdrometer served as reference to assess the absolute percent bias of calculated rainfall intensity using the data of different-resolution tipping-bucket rain gauges classically applied by climatologists and hydrologists in the Andean region. Additionally, the impact of the calculation method (tip counting versus cubic spline interpolation) was examined. The combined effect was assessed for different rainfall intensity classes (0–0.99, 1–1.99, 2–4.99, and 5–10 mm·h−1) and timescales (5, 10, 30, and 60 min). Additionally, the variation in percent absolute bias of the Davis rain gauge, the collector most widely used in the study region, was defined with respect to the Texas rain gauge along an elevation gradient between 3300 and 4000 m a.s.l. Results reveal that the value of the percent absolute bias is largest for small rainfall intensities (≤2 mm·h−1) and short timescales (≤10 min), and decreases when the cubic spline interpolation is used. No relation was found between the error, the elevation, and rainfall depth along the gradient. Based on the research findings, it is recommended to measure precipitation in the high Andean mountain region with a high-resolution sensor and to consider cubic spline for the computation of intensities.

26 January 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
YYYi Yao
Yi Yao
ADAgnès Ducharne
Agnès Ducharne
BCBenjamin I. Cook
Benjamin I. Cook
SHSteven J. De Hertog
Steven J. De Hertog
KAKjetil Schanke Aas
Kjetil Schanke Aas
WTWim Thiery
Wim Thiery
Irrigation rapidly expanded during the 20th century, affecting climate via water, energy, and biogeochemical changes. Previous assessments of these effects predominantly relied on a single Earth System Model, and therefore suffered from structural model uncertainties. Here we quantify the impacts of historical irrigation expansion on climate by analysing simulation results from six Earth system models participating in the Irrigation Model Intercomparison Project (IRRMIP). Results show that irrigation expansion causes a rapid increase in irrigation water withdrawal, which leads to less frequent 2-meter air temperature heat extremes across heavily irrigated areas (≥4 times less likely). However, due to the irrigation-induced increase in air humidity, the cooling effect of irrigation expansion on moist-heat stress is less pronounced or even reversed, depending on the heat stress metric. In summary, this study indicates that irrigation deployment is not an efficient adaptation measure to escalating human heat stress under climate change, calling for carefully dealing with the increased exposure of local people to moist-heat stress.