Heavy rainfall in Paraguay during the 2015-2016 austral summer: causes and sub-seasonal-to-seasonal predictive skill
Abstract: During the austral summer 2015/16, severe flooding displaced over 170 000 people on the Paraguay River system in Paraguay, Argentina, and southern Brazil. These floods were driven by repeated heavy rainfall events in the lower Paraguay River basin. Alternating sequences of enhanced moisture inflow from the South American low-level jet and local convergence associated with baroclinic systems were conducive to mesoscale convective activity and enhanced precipitation. These circulation patterns were favored by cross-time-scale interactions of a very strong El Niño event, an unusually persistent Madden–Julian oscillation in phases 4 and 5, and the presence of a dipole SST anomaly in the central southern Atlantic Ocean. The simultaneous use of seasonal and subseasonal heavy rainfall predictions could have provided decision-makers with useful information about the start of these flooding events from two to four weeks in advance. Probabilistic seasonal forecasts available at the beginning of November successfully indicated heightened probability of heavy rainfall (90th percentile) over southern Paraguay and Brazil for December–February. Raw subseasonal forecasts of heavy rainfall exhibited limited skill at lead times beyond the first two predicted weeks, but a model output statistics approach involving principal component regression substantially improved the spatial distribution of skill for week 3 relative to other methods tested, including extended logistic regressions. A continuous monitoring of climate drivers impacting rainfall in the region, and the use of statistically corrected heavy precipitation seasonal and subseasonal forecasts, may help improve flood preparedness in this and other regions.
Introduction: The Challenge of Sub-seasonal Prediction
Predicting extreme weather events often falls into a "predictability gap." While short-term weather forecasts (up to 10 days) and seasonal outlooks (months in advance) have become increasingly reliable, the sub-seasonal range (2–4 weeks) remains notoriously difficult to forecast accurately. This gap is critical for disaster preparedness, as evidenced by the 2015–16 austral summer when the Lower Paraguay River Basin (LPRB) experienced catastrophic flooding. The event displaced over 170,000 people across Paraguay, Argentina, and southern Brazil, prompting urgent questions about the predictability of such persistent extremes.
This study diagnoses the multi-scale climate drivers behind this flooding and evaluates whether statistical post-processing can extend the lead time of useful heavy rainfall predictions to bridge the sub-seasonal gap.
Study Area and Methodology
The research focuses on the Lower Paraguay River Basin (LPRB), a region with low-lying topography that is highly sensitive to sustained precipitation. To analyze the event, the researchers integrated high-resolution rainfall data (CPC Unified Gauge-based Analysis), atmospheric reanalysis (NCAR-NCEP Reanalysis II), and sea-surface temperature (SST) records.
The methodology employed a two-pronged approach:
- Weather Typing: K-means clustering was applied to the 850 hPa streamfunction to identify six distinct "Weather Types" (WTs). This simplified complex daily circulation into a sequence of states, revealing which patterns were unusually persistent during the flood period.
- Model Output Statistics (MOS): To improve sub-seasonal forecasts from the ECMWF, the authors tested four statistical correction techniques: Extended Logistic Regression (XLR/HXLR), Principal Component Regression (PCR), and Canonical Correlation Analysis (CCA). PCR and CCA were particularly important as they account for spatial shifts and biases that simple grid-point corrections often miss.
The "Perfect Storm" of Climate Drivers
The extreme rainfall was the result of a "cross-timescale interaction" of several climate drivers. While a very strong El Niño provided the broad background for wet conditions, the persistence of the event was driven by specific sub-seasonal triggers:
- The Madden-Julian Oscillation (MJO): An unusually persistent MJO in phases 4 and 5 favored enhanced convection in the region.
- South Central Atlantic Dipole (SCAD): A specific dipole in Atlantic SSTs favored "No-Chaco" jet events. Unlike the standard "Chaco" jet that pushes moisture into Argentina, the "No-Chaco" variant turns eastward, dumping heavy rain directly into the LPRB.
As shown in Figure 2, monthly anomalies revealed a clear anticyclonic circulation over central Brazil (top row), which enhanced moisture transport via the low-level jet, leading to sustained precipitation anomalies (bottom row) throughout the summer.
Enhancing Forecast Skill
The study found that seasonal forecasts were highly successful; probabilistic outlooks available in November 2015 indicated a 90% probability of heavy rainfall for the December–February period. However, sub-seasonal skill was more elusive.
Raw sub-seasonal models typically lost skill after two weeks. By applying PCR-based MOS, the researchers substantially improved the spatial distribution of skill for "Week 3" (14–19 day lead time). PCR's ability to analyze the entire spatial field allowed it to correct for the model's tendency to shift rainfall patterns geographically, providing a significant boost over standard grid-point bias correction methods.
Conclusion and Future Directions
The 2015–16 Paraguay floods demonstrate the importance of monitoring multi-scale climate drivers. While raw sub-seasonal models face limitations at longer lead times, statistical post-processing—specifically methods like PCR that account for spatial modes—can extend the window of actionable information for water managers.
Future work suggests that integrating these corrected sub-seasonal forecasts into hydrological models could provide direct discharge predictions. Such advancements could significantly improve flood preparedness and mitigation strategies in the LPRB and other vulnerable regions globally.
Summary updated