Browsing by Autor "Paula Pacheco Mollinedo"

Now showing 1 - 7 of 7

Development of Hourly Resolution Air Temperature Across Titicaca Lake on Auxiliary ERA5 Variables and Machine Learning-Based Gap-Filling
(Multidisciplinary Digital Publishing Institute, 2025) J. W. Sirpa-Poma; Juan Marcos Calle; Elvis Uscamayta-Ferrano; Jorge Molina‐Carpio; Frédéric Satgé; Osmar Cuentas Toledo; Ricardo Duran; Paula Pacheco Mollinedo; Rizuana Iqbal Hussain; Ramiro Pillco Zolá
This article presents an innovative procedure that combines advanced quality control (QC) methods with machine learning (ML) techniques to produce reliable, continuous, high-resolution meteorological data. The approach was applied to hourly air temperature records from six automatic weather stations located around Lake Titicaca in the Altiplano region of South America. The raw dataset contained time gaps, inconsistencies, and outliers. To address these, the QC stage employed Interquartile Range, Biweight, and Local Outlier Factor (LOF) statistics, resulting in a clean dataset. Two gap-filling methods were implemented: a spatial approach using time series from nearby stations and a temporal approach based on each station's time series and selected variables from the ERA5-Land reanalysis. Several ML models were also employed in this process: Random Forest (RF), Support Vector Machine (SVM), Stacking (STACK), and AdaBoost (ADA). Model performance was evaluated on a validation subset (30% of station data). The RF model achieved the best results, with R<sup>2</sup> values up to 0.9 and Root Mean Square Error (RMSE) below 1.5 °C. The spatial approach performed best when stations were strongly correlated, while the temporal approach was more suitable for locations with low inter-station correlation and high local variability. Overall, the procedure substantially improved data reliability and completeness, and it can be extended to other meteorological variables.
Evaluating the Sensitivity of Hydrological Models to Remotely Sensed Precipitation in a Transboundary Basin
(2025) Paula Pacheco Mollinedo; Frédéric Satgé; Renaud Hostache; Marie‐Paule Bonnet; Jorge Molina‐Carpio; Ramiro Pillco Zolá; Edson Ramírez; Daniel Espinoza
Accurate precipitation data is vital for hydrological modelling, particularly in transboundary basins with scarce hydro-climatic stations. This study evaluates the performance of 20 gridded precipitation products (GPPs), derived from remotely sensed data and reanalyses, in the transboundary Lake Titicaca basin. The methodology integrates two approaches: first, a spatial and temporal accuracy assessment of the GPPs, and second, their application as input data in hydrological models.For spatial accuracy, annual precipitation maps were generated for each GPP, preserving their native resolution, and compared with gauge-based maps. Temporal accuracy was assessed using Taylor diagrams. To evaluate the impact of GPPs on hydrological modelling, streamflow simulations were performed using the GR4J (lumped) and MGB-IPH (semi-distributed) models for three sub-basins, with model performance assessed through Kling-Gupta Efficiency (KGE).Results indicate that CHIRPS, IMERG, and MSWEP excel in spatial and temporal accuracy, capturing the north-to-south precipitation gradient shaped by Andean topography. Streamflow simulations showed that GPPs often outperform gauge-based precipitation in basins with uneven station distribution. In GR4J, MSWEP and CHIRPS yielded the highest KGE values across all sub-basins, while in MGB-IPH, SM2Rain_CCI and IMERG-FR performed best. Notably, the higher KGE scores observed for the GR4J model can be attributed to its lumped structure, which compensates for GPP over/under estimations and spatial distribution inconsistencies.This comprehensive evaluation demonstrates the potential of remotely sensed precipitation products to address data scarcity in transboundary basins. By improving streamflow simulations, these products support informed water resource management, climate adaptation, and transboundary collaboration.
Reliability of gridded temperature datasets to monitor surface air temperature variability over Bolivia
(Wiley, 2023) Frédéric Satgé; Ramiro Pillco Zolá; Jorge Molina‐Carpio; Paula Pacheco Mollinedo; M‐P. Bonnet
Abstract Six gridded temperature datasets (T‐datasets) were evaluated for the first time over the South American continent, through the case study of Bolivia, by comparing them with temperature records acquired from 82 meteorological stations spanning the 1995–2010 period. The comparisons were carried out at the daily time step considering different seasons (annual scale, austral summer and austral winter) and regions (Amazon, La Plata and Altiplano basins). Overall, the climate hazards group infrared temperature with stations (CHIRTS) and the climate prediction centre (CPC) T‐datasets provided the most reliable mean daily temperature ( T mean ) and also described well the temporal variability of minimum and maximum daily temperature estimates ( T n and T x ). T mean , T n and T x trends were analysed over the 1983–2016 period to observe temperature temporal evolution across the three regions. Despite some general agreements between the trends ( T mean , T x and T n ), large discrepancies are also observed. It was found that CPC overestimates and CHIRTS underestimates mean temperature trends and that CPC (CHIRTS) was better than CHIRTS (CPC) to estimate T x ( T n ) trends, both in magnitude and space. Furthermore, opposing trends (i.e., warming and cooling) are described by CPC and CHIRTS for some specific regions, which call into question their reliability for such analyses. These findings highlight the need to validate gridded temperature products with reliable ground data for the regions under study, particularly if they have a wide elevation range.
Research and innovation missions to transform future water systems
(2023) D. Mark Smith; Christopher Gordon; Anoulak Kittikhoun; Jennifer Molwantwa; Paula Pacheco Mollinedo; Abir Ben Romdhane; Raunak Shrestha; Callist Tindimugay; Rachael McDonnell
Sensitivity of lumped and semi-distributed hydrological models to 20 gridded precipitation products in a transboundary basin
(Elsevier BV, 2025) Paula Pacheco Mollinedo; Frédéric Satgé; Marie‐Paule Bonnet; Jorge Molina‐Carpio; Ramiro Pillco Zolá; Edson Ramírez; Daniel Espinoza; Renaud Hostache
Towards the Improvement of Soil Salinity Mapping in a Data-Scarce Context Using Sentinel-2 Images in Machine-Learning Models
(Multidisciplinary Digital Publishing Institute, 2023) J. W. Sirpa-Poma; Frédéric Satgé; Eléonore Resongles; Ramiro Pillco Zolá; Jorge Molina‐Carpio; M. G. Flores Colque; Mauricio Ormachea; Paula Pacheco Mollinedo; Marie‐Paule Bonnet
Several recent studies have evidenced the relevance of machine-learning for soil salinity mapping using Sentinel-2 reflectance as input data and field soil salinity measurement (i.e., Electrical Conductivity-EC) as the target. As soil EC monitoring is costly and time consuming, most learning databases used for training/validation rely on a limited number of soil samples, which can affect the model consistency. Based on the low soil salinity variation at the Sentinel-2 pixel resolution, this study proposes to increase the learning database's number of observations by assigning the EC value obtained on the sampled pixel to the eight neighboring pixels. The method allowed extending the original learning database made up of 97 field EC measurements (OD) to an enhanced learning database made up of 691 observations (ED). Two classification machine-learning models (i.e., Random Forest-RF and Support Vector Machine-SVM) were trained with both OD and ED to assess the efficiency of the proposed method by comparing the models' outcomes with EC observations not used in the models´ training. The use of ED led to a significant increase in both models' consistency with the overall accuracy of the RF (SVM) model increasing from 0.25 (0.26) when using the OD to 0.77 (0.55) when using ED. This corresponds to an improvement of approximately 208% and 111%, respectively. Besides the improved accuracy reached with the ED database, the results showed that the RF model provided better soil salinity estimations than the SVM model and that feature selection (i.e., Variance Inflation Factor-VIF and/or Genetic Algorithm-GA) increase both models´ reliability, with GA being the most efficient. This study highlights the potential of machine-learning and Sentinel-2 image combination for soil salinity monitoring in a data-scarce context, and shows the importance of both model and features selection for an optimum machine-learning set-up.
Visceral Leishmaniasis in Bolivia: Current Status
(Brazilian Society of Tropical Medicine, 2020) Juan Sergio Mollinedo; Zoraida Aymara Mollinedo; Wilson J. Gironda; René Mollinedo; Paula Pacheco Mollinedo; Oscar Daniel Salomón
We confirmed that dogs are its primary reservoir, and Lutzomyia longipalpis is its main vector (currently dispersed in six departments). The primary vectors in areas where Lutzomyia longipalpis is absent are Migonemyia migonei and Lutzomyia cruzi.