A weather regime characterisation of winter biomass aerosol transport from southern Africa

During austral winter, a compact low cloud deck over the South Atlantic contrasts with clear sky over southern Africa, where forest fires triggered by dry conditions emit large amounts of biomass burning aerosols (BBAs) in the free troposphere. Most of the BBA burden crosses the South Atlantic embedded in the tropical easterly flow. However, midlatitude synoptic disturbances can deflect part of the aerosol from the main transport path towards southern extratropics. In this study, the first objective classification of the synoptic variability controlling the spatial distribution of BBA in southern Africa and the South Atlantic during austral winter (August to October) is presented. By analysing atmospheric circulation data from reanalysis products, a six-class weather regime (WR) classification of the region is constructed. The classification reveals that the synoptic variability is composed of four WRs, representing disturbances travelling at midlatitudes, and two WRs accounting for pressure anomalies in the South Atlantic. The WR classification is then successfully used to characterise the aerosol spatial distribution in the region in the period 2003–2017, in both reanalysis products and station data. Results show that the BBA transport towards southern extratropics is controlled by weather regimes associated with midlatitude synoptic disturbances. In particular, depending on the relative position of the pressure anomalies along the midlatitude westerly flow, the BBA transport is deflected from the main tropical route towards southern Africa or the South Atlantic. Moreover, the WRs accounting for midlatitude disturbances show organised transition sequences, which allow one to illustrate the evolution of the BBA northerly transport across the region in the context of a wave pattern. The skill in characterising the BBA transport shown by the WR classification indicates the potential for using it as a diagnostic/predictive tool for the aerosol dynamics, which is a key component for the full understanding and modelling of the complex radiation–aerosol–cloud interactions controlling the atmospheric radiative budget in the region.

Improvement of seasonal forecast correction by using weather regimes

<p>In order to develop seasonal forecast applications, raw forecast data generally need to be corrected to remove their systematic errors and drifts in time. In the climate community, methods based on quantile mapping techniques are quite common for their easy implementation. In the framework of the SECLI-FIRM project, we have tested a refinement of quantile mapping by conditioning the correction to weather regimes, in order to take large-scale circulation into account. For that purpose, we have used ADAMONT, a tool originally developed by Météo-France to correct climate projection scenarios. It was applied on four C3S seasonal forecast models over Europe, using ERA5 as a reference. Three parameters were treated at daily time-step: 2-metre temperature, precipitation and 10-metre wind-speed.</p><p>One of the main objectives of this study was to better understand the role weather regimes can play, if/when/where/for which parameter we gain in quality and predictability. For instance, a series of experiments were conducted on an idealized case of “perfect forecasts” of weather regimes, to point out the maximum benefits we could expect from the method.</p><p>Another focus of research was to test some strategies to optimize the positive impact of the introduction of weather regimes, by selecting members in one model ensemble or by using a multi-model approach. The selection was based on a sub-sampling of the best members in terms of weather regime frequency forecast, in order to determine the needed precision of weather regime forecast, for it to be useful in the correction.</p><p><span>We</span><span> will present the </span><span>main </span><span>results </span><span>of this work </span><span>and </span><span>some operational perspectives.</span></p>

B-matrix estimation in the Copernicus European Regional Re-Analysis (CERRA)

<p>A new augmented Ensemble of Data Assimilations (EDA) technique, which estimates background error covariances (B-matrix), has been developed for the new Copernicus European Regional Re-Analysis (CERRA-EDA). CERRA-EDA has 10 members with two main pools of forecast differences: seasonal and daily. The seasonal component is pre-prepared (`offline') at reanalysis-resolution (5.5km). The new augmentation governs the time-dependent mixture of winter and summer differences of this seasonal component with respect to the time of year. The daily component is (`online') and averaged in moving succession over 2.5 days with subsequent B-matrix computation every 2 days. This daily component runs at 11km and the forecasts are interpolated to 5.5km prior to use. The seasonal-daily split is set to a fixed value of 80-20\% for CERRA production. The EDA is cycled 6-hourly while CERRA has a 3-hour analysis cycle. The B-matrix is modelled on a bi-Fourier limited area weather model, where dependence of vertical correlations on horizontal scale (non-separability), horizontal homogeneity and isotropy are assumed. The mass-wind and specific humidity fields are related via vorticity and geopotential and the relationships are estimated via multiple linear regressions enforcing simplified analogues of flow-dependence. </p><p>We demonstrate the potential of CERRA-EDA to estimate rapid changes in weather regime change over Europe by assessing B-matrix statistics and forecast skill scores in a case study. The case study assesses two like-periods bearing different weather regimes, Mar-03 (blocking regime) and Mar-18 (NAO- regime). The aptitude of the B-matrix to reflect weather regime change is shown to be mostly dependent on the observation network in a given year. We also illustrate the impact of: change in observation networks over time, and varying the seasonal-daily split. This is shown through analysing the spatio-temporal evolution of background standard deviations. Finally, analysis and forecast skill scores up to 24-hours are also shown to offer improvements worth considering.</p>

A weather regime characterisation of winter biomass aerosol transport from southern Africa

During austral winter, a compact low cloud deck over South Atlantic contrasts with clear sky over southern Africa, where forest fires triggered by dry conditions emit large amount of biomass burning aerosols (BBA) in the free troposphere. Most of the BBA burden crosses South Atlantic embedded in the tropical easterly flow. However, midlatitude synoptic disturbances can deflect part of the aerosol from the main transport path towards southern extratropics. In this study, a characterisation of the synoptic variability controlling the spatial distribution of BBA in southern Africa and South Atlantic during austral winter (August to October) is presented. By analysing atmospheric circulation data from reanalysis products, a 6-class weather regime (WR) classification of the region is constructed. The classification reveals that the synoptic variability is composed by four WRs representing disturbances travelling at midlatitudes, and two WRs accounting for pressure anomalies in the South Atlantic. The WR classification is then successfully used to characterise the aerosol spatial distribution in the region in the period 2003–2017, in both reanalysis products and station data. Results show that the BBA transport towards southern extratropics is controlled by weather regimes associated with midlatitude synoptic disturbances. In particular, depending on the relative position of the pressure anomalies along the midlatitude westerly flow, the BBA transport is deflected from the main tropical route towards southern Africa or the South Atlantic. This paper presents the first objective classification of the winter synoptic circulation over South Atlantic and southern Africa. The classification shows skills in characterising the BBA transport, indicating the potential for using it as a diagnostic/predictive tool for aerosol dynamics, which is a key component for the full understanding and modelling of the complex radiation-aerosol-cloud interactions controlling the atmospheric radiative budget in the region.

Rossby wave activity associated with Euro-Atlantic weather regimes in the PRIMAVERA historical runs.

<p><span>In this study we </span><span>aim to assess how the upper tropospheric Rossby wave activity is represented in the PRIMAVERA models. </span><span>The low and high resolution historical coupled simulations will be compared with ERA5 reanalysis </span><span>(spanning the 1979-2014 period)</span><span> to enlight</span><span>en</span><span> model deficiencies in representing the spatial distribution </span><span>and temporal evolution</span><span> of Rossby wave activity </span><span>and to emphasize the benefits of </span><span>increased resolution. </span><span>Our analysis focuses </span><span>on </span><span>the wintertime large scale circulation over</span><span> the Euro-</span><span>A</span><span>tlantic </span><span>sector</span><span>. </span></p><p><span>A</span><span> diagnostic based on Local </span><span>W</span><span>ave </span><span>A</span><span>ctivity </span><span>(LWA)</span><span> in isentropic coordinates </span><span>is used </span><span>to </span><span>identify Rossby waves and to </span><span>quantify </span><span>their amplitude</span><span>. </span><span>LWA is partitioned into its stationary and transient components, </span><span>to </span><span>distinguish</span><span> the contribution from </span><span>planetary</span><span> versus </span><span>synoptic scale waves (i.e. wave packets)</span><span>. </span><span>This diagnostic is then combined with another </span><span>one</span><span> to identify persistent and recurrent large scale circulation patterns, the so called weather regimes</span><span>. Weather regimes in the Euro-Atlantic sector are identified with the usual approach </span><span>of EOF decomposition and k-mean clustering applied to daily anomalies of Montgomery streamfunction, </span><span>in order </span><span>to have a consistent framework with LWA </span><span>(</span><span>which is defined in isentropic coordinates</span><span>)</span><span>. </span><span>A</span><span> composite of transient LWA is realised for each weather regime to obtain the spatial distribution of Rossby wave activity associated with each weather regime.</span></p><p><span>Results show a marked intermodel variability in the ability of reproducing the correct (i.e. the one observed in reanalysis data) LWA distribution. Many of the models in fact fails to reproduce the localized (in space) maxima of LWA associated with each weather regime and to distribute LWA over a larger region compared to reanalysis. High resolution helps to correct this bias in the majority of the models, in particular in those where the low-resolution LWA distribution was already close to reanalysis. Finally, the temporal behaviour of the spatially averaged LWA in the examined period is discussed.</span></p>

Flow-dependent sub-seasonal forecast skill for Atlantic-European weather regimes and its relation to planetary- to synoptic-scale processes

<div><span>Sub-seasonal numerical weather forecasts (10 – 60 days) primarily aim to predict the evolution of the large-scale circulation and its associated surface weather on continent- and multi-daily scales. In the extratropics, this atmospheric variability is depicted best by so-called weather regimes. Here, we assess the ability of sub-seasonal reforecasts of the European Centre for Medium-Range Weather Forecasts (ECMWF) to predict 7 year-round weather regimes in the Atlantic-European region. We first investigate how well the forecasts reproduce frequency, length, and transitions of the weather regime life cycles. We then show that the average forecast skill horizon varies by several days for different weather regimes, seasons, and initial planetary-scale flow states. In a final part, we provide first insight into how synoptic-scale processes, more specifically warm conveyor belts, and their inherent intrinsic predictability limit might affect this flow-dependent sub-seasonal weather regime forecast skill.</span></div>

Improving Sub-Seasonal Hydrological Forecasts in Switzerland: An Exploratory Study of Post-Processing Techniques by Using Machine Learning and Weather Regime Diagnostics

<p>Across the globe, there has been an increasing interest in improving the predictability of weekly to monthly (sub-seasonal) hydro-meteorological forecasts as they play a valuable role in medium- to long-term planning in many sectors such as agriculture, navigation, hydro-power production, and hazard warnings. A Precipitation-Runoff-Evapotranspiration HRU model (PREVAH) has been previously set up with raw metrological forcing of 51 ensemble members and 32 days lead time taken from the operational European Centre for Medium-Range Weather Forecasts (ECMWF) extended-range forecast. The PREVAH model is used to generate hydrological forecasts for the study area, which consists of 300 catchments covering approximately the entire area of Switzerland. The primary goal of this study is to improve the quality of the categorical forecast of weekly mean total discharge in a catchment laying in the lower, normal, or upper tercile of the climatological distribution at a monthly horizon. Therefore, we explore <span>the approach to post-process PREVAH outputs using machine learning algorithm Gaussian process</span>. Weather regime (WR) data, based on 500 hPa geopotential height in the Atlantic-European region are used as an added feature to further enhance the post-processing performance.</p><p>By comparing the overall accuracy and the ranked probability skill score of the post-processed forecasts with the ones of raw forecasts we show that the proposed post-processing techniques are able to improve the forecast skill. The degree of improvement varies by catchment, lead time and variable. The benefit of the added WR data is not consistent across the study area but most promising in high altitude catchments with steep slopes. Among the seven types of WRs, the majority of the corrections are observed when either a European blocking or a Scandinavian blocking is forecasted as the dominant weather regime. By applying a “best practice” to each individual catchment, that is the processing technique with the highest accuracy among the different proposed techniques, a median accuracy of 0.65 (improved from a value of 0.53 with no processing technique) can be achieved at 4-week lead time. Due to the small data size, the conclusions should be considered preliminary, but this study highlights the potential of improving the skill of sub-seasonal hydro-meteorological forecasts utilizing weather regime data and machine learning in a real-time deployable setup.</p>