Multi-model ensemble of statistically downscaled GCMs over southeastern South America: historical evaluation and future projections of daily precipitation with focus on extremes

Olmo, M. E.; Balmaceda-Huarte, Rocío; Bettolli, Maria Laura

doi:10.21203/rs.3.rs-942877/v1

Cited by 10 publications

(14 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, the dry‐conditions seemed to be controlling the long‐term variability of these sequential CE over northern SA. Additionally, the positive changes over southeastern SA—that were found to be robust even in the RCP 2.6 scenario—were coincident with previous findings on the increasing number of ER events in the region, probably related to modifications in specific atmospheric circulation patterns (Blázquez & Solman, 2020; Olmo et al., 2022). In the same line, a decline in precipitation amounts over central and southern Chile has been detected in the literature working with both observational and model data (Balmaceda‐Huarte et al., 2021; Diaz et al., 2020; Olmo et al., 2020), which seems to be influencing the future projections of this CE.…”

Section: Summary and Final Remarkssupporting

confidence: 88%

“…Precipitation changes are more variable and season‐dependent over southeastern SA, with more heterogeneous changes in dry conditions and increases in ER events that are projected to intensify in the future (Ceron et al., 2020; Regoto et al., 2020; Olmo et al., 2022). This seasonality could have some implications upon the development of agricultural and hydrological practices in the region.…”

Section: Summary and Final Remarksmentioning

confidence: 99%

See 1 more Smart Citation

Compound Events in South America Using the CORDEX‐CORE Ensemble: Current Climate Conditions and Future Projections in a Global Warming Scenario

et al. 2022

Self Cite

View full text Add to dashboard Cite

Climate hazards associated with compound events (CEs) have lately received increasing attention over South America (SA) due to their potential risks and amplification of impacts. This work addressed the evaluation of different temperature‐ and precipitation‐based CE in SA considering the CORDEX‐CORE ensemble of regional climate models (RCMs) and their driving earth system models (ESMs) in the reference period 1981–2010 and the late 21st century (2070–2099), for the Representative Concentration Pathways (RCPs) 2.6 and 8.5 scenarios. The assessment focused on model performance for the individual events—heatwaves (HWs), Extreme rainfall (ER) days, and dry‐spells (DSs)—and their compound occurrence in terms of climatological frequency and duration. The spatial patterns of individual events were adequately reproduced by the RCMs, evidencing general overestimations in extreme precipitation intensities. In terms of CE, the frequencies of coincident HWs and DSs (sequential DSs and ER) were remarkable over central‐eastern Brazil and southern SA (southeastern SA). The main features of CE were generally well‐simulated by the RCMs, although they presented regional differences such as an underestimation of the maximum frequencies of these two CE in northeastern Brazil and southeastern SA, respectively. The high‐resolution information was generally in line with the larger‐scale driving ESMs. The climate change signal analysis generally showed robust future increases in CE frequency and duration in different areas of SA, as for coincident HWs and DSs (sequential DSs and ER) over northern SA and southern Brazil (southeastern SA). This was mostly consistent among the RCMs ensemble and notably strengthened in the worst‐case scenario (RCP 8.5).

show abstract

Section: Summary and Final Remarkssupporting

confidence: 88%

Section: Summary and Final Remarksmentioning

confidence: 99%

Compound Events in South America Using the CORDEX‐CORE Ensemble: Current Climate Conditions and Future Projections in a Global Warming Scenario

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, we apply our trained models to downscale the projections from this ensemble for the historical and RCP8.5 scenario (2006-2100) periods. We follow previous work in this field (Baño-Medina et al, 2021;Olmo et al, 2022) and select the RCP8.5 scenario, which shows the strongest climate change signal (especially for temperature) and therefore allows the generalization capability of CNNs to be optimally explored. Due to their different spatial resolutions, all GCM data have been interpolated to the reference 2 • grid (considering the nearest grid box) to match the predictor space used for ERA-Interim.…”

Section: Methodsmentioning

confidence: 99%

Downscaling multi-model climate projection ensembles with deep learning (DeepESD): contribution to CORDEX EUR-44

et al. 2022

View full text Add to dashboard Cite

Abstract. Deep learning (DL) has recently emerged as an innovative tool to downscale climate variables from large-scale atmospheric fields under the perfect-prognosis (PP) approach. Different convolutional neural networks (CNNs) have been applied under present-day conditions with promising results, but little is known about their suitability for extrapolating future climate change conditions. Here, we analyze this problem from a multi-model perspective, developing and evaluating an ensemble of CNN-based downscaled projections (hereafter DeepESD) for temperature and precipitation over the European EUR-44i (0.5∘) domain, based on eight global circulation models (GCMs) from the Coupled Model Intercomparison Project Phase 5 (CMIP5). To our knowledge, this is the first time that CNNs have been used to produce downscaled multi-model ensembles based on the perfect-prognosis approach, allowing us to quantify inter-model uncertainty in climate change signals. The results are compared with those corresponding to an EUR-44 ensemble of regional climate models (RCMs) showing that DeepESD reduces distributional biases in the historical period. Moreover, the resulting climate change signals are broadly comparable to those obtained with the RCMs, with similar spatial structures. As for the uncertainty of the climate change signal (measured on the basis of inter-model spread), DeepESD preserves the uncertainty for temperature and results in a reduced uncertainty for precipitation. To facilitate further studies of this downscaling approach, we follow FAIR principles and make publicly available the code (a Jupyter notebook) and the DeepESD dataset. In particular, DeepESD is published at the Earth System Grid Federation (ESGF), as the first continental-wide PP dataset contributing to CORDEX (EUR-44).

show abstract

“…The obtained CPs were able to differentiate multiple precipitation patterns related to extreme events over SESA -that is a remarkable hotspot for extreme rainfall events. These CPs were then used in the design of empirical statistical downscaling models (ESD) for precipitation extremes in the region (Olmo et al 2022). The authors detected an increasing model uncertainty in the ESD precipitation projections over SESA for the late 21st century, partially caused by GCMs spread.…”

Section: Introductionmentioning

confidence: 99%

Model uncertainty in synoptic circulation patterns and precipitation changes in Southern South America using CMIP5 and CMIP6 models

Olmo,

Bettolli,

Balmaceda-Huarte

2023

Preprint

Self Cite

View full text Add to dashboard Cite

The effects of global warming on the regional climate of southern South America (SSA) during the recent decades have been exhaustively documented with consistency throughout the literature. However, the projected changes on temperature- and precipitation-related climate hazards depict an important uncertainty, mostly in the intensity of the changes. This work assessed a set of CMIP5 and CMIP6 global climate models (GCMs) in reproducing the observed atmospheric circulation patterns (CPs) over SSA and their expected changes for the 21st century. Furthermore, the attribution of the seasonal precipitation changes to changes in the CPs was explored for the late future (2070–2100). GCMs were generally able to represent the variety of CPs and their seasonal frequencies, although presenting more deficiencies in capturing their respective precipitation patterns over SSA. Larger model agreement was found in the increasing and/or decreasing frequency of specific CPs for the near future (2040–2070) than for the late future, when model spread became more noticeable. Particularly, CPs associated with larger positive rainfall anomalies over southeastern South America -a hotspot for precipitation extremes- are expected to become more frequent in the near future, whereas their changes in the longer-term are more uncertain. When performing an attributional study, precipitation changes showed important differences between GCMs and were often associated with changes in the intra-pattern variability rather than in the CPs changing frequency. In this way, the modification of the precipitation regime of SSA may not be explained only by changes in the large-scale circulation but probably also by other regional-to-local features.

show abstract

Multi-model ensemble of statistically downscaled GCMs over southeastern South America: historical evaluation and future projections of daily precipitation with focus on extremes

Cited by 10 publications

References 22 publications

Compound Events in South America Using the CORDEX‐CORE Ensemble: Current Climate Conditions and Future Projections in a Global Warming Scenario

Compound Events in South America Using the CORDEX‐CORE Ensemble: Current Climate Conditions and Future Projections in a Global Warming Scenario

Downscaling multi-model climate projection ensembles with deep learning (DeepESD): contribution to CORDEX EUR-44

Model uncertainty in synoptic circulation patterns and precipitation changes in Southern South America using CMIP5 and CMIP6 models

Contact Info

Product

Resources

About