Oceanic-atmospheric phenomena of different time scales concurrently might affect the streamflow in several basins around the world. The Atrato River Basin (ARB) and Patía River Basin (PRB) of the Colombian Pacific region are examples of such basins. Nevertheless, the relations between the streamflows in the ARB and PRB and the oceanic-atmospheric factors have not been examined considering different temporal scales. Hence, this article studies the relations of the climate indices and the variability of the streamflows in the ARB and PRB at interannual and decadal timescales. To this, the streamflow variability modes were obtained from the principal component analysis (PCA); furthermore, their linear dependence with indices of the El Niño/Southern Oscillation (ENSO), precipitation (PRP), the Choco low-level jet (CJ), and other indices were quantified through (a) Pearson and Kendall’s tau correlations, and (b) wavelet transform. The PCA presented a single significant mode for each basin, with an explained variance of around 80%. The correlation analyses between the PC1s of the ARB and PRB, and the climate indices showed significant positive (negative) high correlations with PRP, CJ, and Southern Oscillation Index (SOI) (ENSO indices). The wavelet coherence analysis showed significant coherencies between ENSO and ARB: at interannual (2–7 years) and decadal scale (8–14), preferably with the sea surface temperature (SST) in the east and west Tropical Pacific Ocean (TPO). For PRB with the SST in the central and western regions of the TPO in the interannual (4–8 years) and decadal (8–14 years) scales, the decreases (increases) in streamflow precede the El Niño (La Niña) events. These results indicate multiscale relations between the basins’ streamflow and climate phenomena not documented in previous works, relevant to forecast the extreme flow events in the Colombian Pacific rivers and for planning and implementing strategies for the sustainable use of water resources in the basins studied.

Improving the accuracy of rainfall forecasting is relevant for adequate water resources planning and management. This research project evaluated the performance of the combination of three Artificial Neural Networks (ANN) approaches in the forecasting of the monthly rainfall anomalies for Southwestern Colombia. For this purpose, we applied the Non-linear Principal Component Analysis (NLPCA) approach to get the main modes, a Neural Network Autoregressive Moving Average with eXogenous variables (NNARMAX) as a model, and an Inverse NLPCA approach for reconstructing the monthly rainfall anomalies forecasting in the Andean Region (AR) and the Pacific Region (PR) of Southwestern Colombia, respectively. For the model, we used monthly rainfall lagged values of the eight large-scale climate indices linked to the El Niño Southern Oscillation (ENSO) phenomenon as exogenous variables. They were cross-correlated with the main modes of the rainfall variability of AR and PR obtained using NLPCA. Subsequently, both NNARMAX models were trained from 1983 to 2014 and tested for two years (2015–2016). Finally, the reconstructed outputs from the NNARMAX models were used as inputs for the Inverse NLPCA approach. The performance of the ANN approaches was measured using three different performance metrics: Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Pearson’s correlation (r). The results showed suitable forecasting performance for AR and PR, and the combination of these ANN approaches demonstrated the possibility of rainfall forecasting in these sub-regions five months in advance and provided useful information for the decision-makers in Southwestern Colombia.

Given that the analysis of past monthly rainfall variability is highly relevant for the adequate management of water resources, the relationship between the climate-oceanographic indices, and the variability of monthly rainfall in Southwestern Colombia at different time scales was chosen as the research topic. It should also be noted that little-to-no research has been carried out on this topic before. For the purpose of conducting this research, we identified homogeneous rainfall regions while using Non-Linear Principal Component Analysis (NLPCA) and Self-Organizing Maps (SOM). The rainfall variability modes were obtained from the NLPCA, while their teleconnection in relation to the climate indices was obtained from Pearson’s Correlations and Wavelet Transform. The regionalization process clarified that Nariño has two regions: the Andean Region (AR) and the Pacific Region (PR). The NLPCA showed two modes for the AR, and one for the PR, with an explained variance of 75% and 48%, respectively. The correlation analyses between the first nonlinear components of AR and PR regarding climate indices showed AR high significant positive correlations with Southern Oscillation Index (SOI) index and negative correlations with El Niño/Southern Oscillation (ENSO) indices. PR showed positive ones with Niño1 + 2, and Niño3, and negative correlations with Niño3.4 and Niño4, although their synchronous relationships were not statistically significant. The Wavelet Coherence analysis showed that the variability of the AR rainfall was influenced principally by the Niño3.4 index on the 3–7-year inter-annual scale, while PR rainfall were influenced by the Niño3 index on the 1.5–3-year inter-annual scale. The El Niño (EN) events lead to a decrease and increase in the monthly rainfall on AR and PR, respectively, while, in the La Niña (LN) events, the opposite occurred. These results that are not documented in previous studies are useful for the forecasting of monthly rainfall and the planning of water resources in the area of study.

The success of many projects linked to the management and planning of water resources depends mainly on the quality of the climatic and hydrological data that is provided. Nevertheless, the missing data are frequently found in hydroclimatic variables due to measuring instrument failures, observation recording errors, meteorological extremes, and the challenges associated with accessing measurement areas. Hence, it is necessary to apply an appropriate fill of missing data before any analysis. This paper is intended to present the filling of missing data of monthly rainfall of 45 gauge stations located in southwestern Colombia. The series analyzed covers 34 years of observations between 1983 and 2016, available from the Instituto de Hidrología, Meteorología y Estudios Ambientales (IDEAM). The estimation of missing data was done using Non-linear Principal Component Analysis (NLPCA); a non-linear generalization of the standard Principal Component Analysis Method via an Artificial Neural Networks (ANN) approach. The best result was obtained using a network with a [45−44−45] architecture. The estimated mean squared error in the imputation of missing data was approximately 9.8 mm. month−1, showing that the NLPCA approach constitutes a powerful methodology in the imputation of missing rainfall data. The estimated rainfall dataset helps reduce uncertainty for further studies related to homogeneity analyses, conglomerates, trends, multivariate statistics and meteorological forecasts in regions with information deficits such as southwestern Colombia.

No abstract