| 1. |
- Sherratt, K., et al.
(författare)
-
Predictive performance of multi-model ensemble forecasts of COVID-19 across European nations
- 2023
-
Ingår i: eLIFE. - : eLife Sciences Publications Ltd. - 2050-084X. ; 12
-
Tidskriftsartikel (refereegranskat)abstract
- Methods: We used open-source tools to develop a public European COVID-19 Forecast Hub. We invited groups globally to contribute weekly forecasts for COVID-19 cases and deaths reported by a standardised source for 32 countries over the next 1–4 weeks. Teams submitted forecasts from March 2021 using standardised quantiles of the predictive distribution. Each week we created an ensemble forecast, where each predictive quantile was calculated as the equally-weighted average (initially the mean and then from 26th July the median) of all individual models’ predictive quantiles. We measured the performance of each model using the relative Weighted Interval Score (WIS), comparing models’ forecast accuracy relative to all other models. We retrospectively explored alternative methods for ensemble forecasts, including weighted averages based on models’ past predictive performance.Results: Over 52 weeks, we collected forecasts from 48 unique models. We evaluated 29 models’ forecast scores in comparison to the ensemble model. We found a weekly ensemble had a consistently strong performance across countries over time. Across all horizons and locations, the ensemble performed better on relative WIS than 83% of participating models’ forecasts of incident cases (with a total N=886 predictions from 23 unique models), and 91% of participating models’ forecasts of deaths (N=763 predictions from 20 models). Across a 1–4 week time horizon, ensemble performance declined with longer forecast periods when forecasting cases, but remained stable over 4 weeks for incident death forecasts. In every forecast across 32 countries, the ensemble outperformed most contributing models when forecasting either cases or deaths, frequently outperforming all of its individual component models. Among several choices of ensemble methods we found that the most influential and best choice was to use a median average of models instead of using the mean, regardless of methods of weighting component forecast models.Conclusions: Our results support the use of combining forecasts from individual models into an ensemble in order to improve predictive performance across epidemiological targets and populations during infectious disease epidemics. Our findings further suggest that median ensemble methods yield better predictive performance more than ones based on means. Our findings also highlight that forecast consumers should place more weight on incident death forecasts than incident case forecasts at forecast horizons greater than 2 weeks.
|
|
| 2. |
- Strong, K., et al.
(författare)
-
Validation of ACE-FTS N2O measurements
- 2008
-
Ingår i: Atmospheric Chemistry and Physics. - : Copernicus GmbH. - 1680-7316 .- 1680-7324. ; 8, s. 4759-4786
-
Tidskriftsartikel (refereegranskat)abstract
- The Atmospheric Chemistry Experiment (ACE), also known as SCISAT, was launched on 12 August 2003, carrying two instruments that measure vertical profiles of atmospheric constituents using the solar occultation technique. One of these instruments, the ACE Fourier Transform Spectrometer (ACE-FTS), is measuring volume mixing ratio (VMR) profiles of nitrous oxide (N2O) from the upper troposphere to the lower mesosphere at a vertical resolution of about 3–4 km. In this study, the quality of the ACE-FTS version 2.2 N2O data is assessed through comparisons with coincident measurements made by other satellite, balloon-borne, aircraft, and ground-based instruments. These consist of vertical profile comparisons with the SMR, MLS, and MIPAS satellite instruments, multiple aircraft flights of ASUR, and single balloon flights of SPIRALE and FIRS-2, and partial column comparisons with a network of ground-based Fourier Transform InfraRed spectrometers (FTIRs). Between 6 and 30 km, the mean absolute differences for the satellite comparisons lie between −42 ppbv and +17 ppbv, with most within ±20 ppbv. This corresponds to relative deviations from the mean that are within ±15%, except for comparisons with MIPAS near 30 km, for which they are as large as 22.5%. Between 18 and 30 km, the mean absolute differences for the satellite comparisons are generally within ±10 ppbv. From 30 to 60 km, the mean absolute differences are within ±4 ppbv, and are mostly between −2 and +1 ppbv. Given the small N2O VMR in this region, the relative deviations from the mean are therefore large at these altitudes, with most suggesting a negative bias in the ACE-FTS data between 30 and 50 km. In the comparisons with the FTIRs, the mean relative differences between the ACE-FTS and FTIR partial columns (which cover a mean altitude range of 14 to 27 km) are within ±5.6% for eleven of the twelve contributing stations. This mean relative difference is negative at ten stations, suggesting a small negative bias in the ACE-FTS partial columns over the altitude regions compared. Excellent correlation (R=0.964) is observed between the ACE-FTS and FTIR partial columns, with a slope of 1.01 and an intercept of −0.20 on the line fitted to the data.
|
|
| 3. |
|
|
| 4. |
- Semeniuk, M., et al.
(författare)
-
Design and fabrication of low-cost renewable carbon electrode materials and their thermo-kinetics for sustainable energy storage applications
- 2022
-
Ingår i: Materials Today Sustainability. - : Elsevier. - 2589-2347. ; 20
-
Tidskriftsartikel (refereegranskat)abstract
- To date, there have been few studies performed on the kinetics of thermal decomposition of renewable carbon and its use in energy storage devices. Novel renewable graphitic carbon from Acer saccharum, sugar maple (SM) heartwood, and Arachis hypogaea, peanut outer shell (PS), are used as anodes in lithium-ion coin cell batteries, with steady specific capacities of 180 mAh/g and 220 mAh/g, respectively, and both retain 100% columbic efficiency for over 350 cycles. Cyclic voltammetry reveals the different charge storage kinetic mechanisms of the two. The voltammogram of SM contains an oxidation peak indicating Li ion intercalation suitable for battery application, while PS lacks a peak; thus, showing potential for function as a supercapacitor. A kinetic study is undertaken on six carbon sources to improve the understanding of the thermal degradation process leading to renewable graphitic carbon. Consistent results are shown across the various model-free methods, indicating that they can accurately describe the devolatilization process. The renewable graphitic carbons morphological features were studied by SEM, XRD, Raman and nitrogen adsorption isotherms. As SM and PS compare favorably to the electrodes they replace, renewable graphitic carbon has the potential for use in a wide variety of novel applications, such as organic thin film transistors, fuel cells, organic batteries, supercapacitors, and other bioelectronics.
|
|
