Does statistical model perform at par with computationally expensive general circulation model for decadal prediction?

Sahastrabuddhe, Rishi; Ghosh, Subimal

doi:10.1088/1748-9326/abfeed

Cited by 4 publications

(4 citation statements)

References 88 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Any accounts of skillful prediction over Europe are often limited to relatively small regions and/or time scales, and draw on the memory of the ocean (Årthun et al 2017). A few studies presented evidence that coupling a statistical model to dynamical and skillful SST predictions can yield skillful decadal prediction of European summer air temperature based on average North Atlantic SST (Wu et al 2019a), European late summer precipitation (Simpson et al 2019), and the Indian summer Monsoon (Sahastrabuddhe & Ghosh 2021). Such hybrid dynamical-statistical (henceforth dyn-stat) predictions rely on observed links between SST (the predictor) and European climate (the predictand) to translate predicted SST into predictions of continental climate, thus circumventing the deficiencies of climate prediction models concerning potential teleconnections.…”

Section: Introductionmentioning

confidence: 99%

Skillful decadal prediction of unforced southern European summer temperature variations

Borchert¹,

Koul²,

Menary³

et al. 2022

Preprint

View full text Add to dashboard Cite

We assess the capability of decadal prediction simulations from the Coupled Model Intercomparison Project phase 6 (CMIP6) archive to predict European summer temperature during the period 1970-2014. Using a multi-model ensemble average from 8 decadal prediction systems, we show that European summer temperatures are highly predictable for up to 10 years in CMIP6. Much of this predictive capability, or skill, is related to the externally forced response. Prediction skill for the unforced signal of European summer temperature is low. A link between unforced Southern European summer temperature and preceding spring Eastern North Atlantic - Mediterranean sea surface temperature (SST) observed during the period 1900-1969 motivates the application of a dynamical-statistical model to overcome the low summer temperature skill over Europe. This dynamical-statistical model uses dynamical spring SST predictions to predict European summer temperature. Our model significantly increases decadal prediction skill of unforced European summer temperature variations: Unlike purely dynamical predictions, the dynamical-statistical model shows significant prediction skill for unforced Southern European summer temperature 2-9 years ahead. Our results highlight that dynamical-statistical models can serve to benefit the decadal prediction of variables with initially limited skill beyond the forcing, such as summer temperature over Europe.

show abstract

Section: Introductionmentioning

confidence: 99%

Skillful decadal prediction of unforced southern European summer temperature variations

Borchert¹,

Koul²,

Menary³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…A few studies presented evidence that coupling a statistical model to dynamical and skillful SST predictions can yield skillful decadal prediction of European summer air temperature based on average North Atlantic SST (Wu et al 2019), European late summer precipitation (Simpson et al 2019), and the Indian summer monsoon (Sahastrabuddhe and Ghosh 2021). Such hybrid dynamical-statistical (henceforth dyn-stat) predictions rely on observed links between SST (the predictor) and European climate (the predictand) to translate predicted SST into predictions of continental climate, thus circumventing the deficiencies of climate prediction models concerning potential teleconnections.…”

Section: Introductionmentioning

confidence: 99%

Skillful decadal prediction of unforced southern European summer temperature variations

Borchert

Koul

Menary

et al. 2021

Environ. Res. Lett.

View full text Add to dashboard Cite

We assess the capability of decadal prediction simulations from the Coupled Model Intercomparison Project phase 6 (CMIP6) archive to predict European summer temperature during the period 1970–2014. Using a multi-model ensemble average, we show that Southern European (SEU) summer temperatures are highly predictable for up to ten years in CMIP6. Much of this predictive skill, is related to the externally forced response: historical simulations explain about 90% of observed SEU summer temperature variance. Prediction skill for the unforced signal of SEU summer temperature is low: initialized model simulations explain less than 10% of observed variance after removing the externally forced response. An observed link between unforced SEU summer temperature and preceding spring Eastern North Atlantic—Mediterranean sea surface temperature (SST) motivates the application of a dynamical-statistical model to overcome the low summer temperature skill over Europe. This dynamical-statistical model uses dynamical spring SST predictions to predict European summer temperature, and significantly increases decadal prediction skill of unforced European summer temperature variations, showing significant prediction skill for unforced Southern European summer temperature 2–9 years ahead. As a result, dynamical-statistical models can benefit the decadal prediction of variables with initially limited skill beyond the forcing, such as summer temperature over Europe.

show abstract

“…However, the variability of high-resolution precipitation in Germany makes the use of higher order polynomials, e.g., of the third order, necessary. This study reveals that statistical approaches can improve dynamical models which has also been shown by Sahastrabuddha and Ghosh (2021) and Smith, 2018) and skillful large-scale teleconnections to improve the skill (see text footnote 1) 9 . In the second case study based on the downscaling of the new model version initialized in November 2020 (and presented now in this paper) the 3-year mean high-resolution SPI is chosen, allowing for a long evaluation period and robust statistical recalibration due to available high-resolution precipitation observations.…”

Section: Discussionmentioning

confidence: 58%

High-Resolution Decadal Drought Predictions for German Water Boards: A Case Study for the Wupper Catchment

et al. 2022

View full text Add to dashboard Cite

Water boards in Germany require decadal predictions to develop optimized management and adaptation strategies, especially within the claims of flood protection and water distribution management. Specifically, the Wupper catchment water board in western Germany is interested in decadal predictions of drought indices, which are correlated to dam water levels. For the management of small catchments, they need multi-year means and multi-year seasonal means of the hydrological seasons for forecast years 1–3 at high spatial resolution. Thus, the MPI-ESM-LR global decadal prediction system with 16 ensemble members at 200 km resolution was statistically downscaled with EPISODES to ~11 km in Germany. Simulated precipitation was recalibrated, correcting model errors and adjusting the ensemble spread. We tested different recalibration settings to optimize the skill. The 3-year mean and 3-year seasonal mean SPI (Standardized Precipitation Index), indicating excess or deficit of precipitation, was calculated. We evaluated the prediction skill with HYRAS observations, applying skill scores and correlation coefficients, and tested the significance of the skill at a 95% levelvia1,000 bootstraps. We found that the high-resolution statistical downscaling is able to preserve the skill of the global decadal predictions and that the recalibration can clearly improve the precipitation skill in Germany. Multi-year annual and August–October mean SPI predictions are promising for several regions in Germany. Additionally, there is potential for skill improvement with increasing ensemble size for all temporal aggregations, except for November–January. A user-oriented product sheet was developed and published on the Copernicus Climate Change Service website (https://climate.copernicus.eu/decadal-predictions-infrastructure). It provides 3-year mean probabilistic SPI predictions for the Wupper catchment and north-western Germany. For 2021–2023, a high probability of negative SPI (dry conditions) is predicted in most of the area. The decadal prediction skill is higher than using the observed climatology as reference prediction in several parts of the area. This case study was developed in cooperation with the Wupper catchment water board and discussed with further German water managers: The skill of high-resolution decadal drought predictions is considered to be promising to fulfill their needs. The product sheet is understandable, well-structured and can be applied to their working routines.

show abstract

Does statistical model perform at par with computationally expensive general circulation model for decadal prediction?

Cited by 4 publications

References 88 publications

Skillful decadal prediction of unforced southern European summer temperature variations

Skillful decadal prediction of unforced southern European summer temperature variations

Skillful decadal prediction of unforced southern European summer temperature variations

High-Resolution Decadal Drought Predictions for German Water Boards: A Case Study for the Wupper Catchment

Contact Info

Product

Resources

About