Description: Mesoscale Convective Systems (MCS) properties during large-scale easterly and westerly regimes of the low-level winds over the tropical northeastern Pacific (TEP) are analyzed during boreal summers of years 2012 to 2014. East of 120°W, two regions located north and south of 10°N present different sensitivity of the MCS properties to the zonal wind regime. Long-lasting and large MCS are favored during the westerly regime over the northern region, while more numerous MCS, but with a steady lifetime distribution, are triggered during the easterly regime over the southern part of the ITCZ. For long-lasting MCS (lifetime larger than 10 hours), the moderate low-level wind shear conditions during the westerly regime are associated with larger MCS covering longer distances. The MCS cold cloud cover and the MCS lifetime tend to locally increase with the moisture of the mid-troposphere. Conversely, the TEP area west of 120°W becomes less favorable to MCS development during the low-level westerly regime, in accordance with the diminution of rainfall observed locally.

Description: The Rossby wave source (RWS) and the corresponding extratropical wave response to tropical convection associated with different phases of the Madden-Julian Oscillation (MJO) is investigated with the dynamical core of a climate model. The initial flow is specified to correspond to the boreal winter climatological flow and an imposed tropical heating that is derived from the observed precipitation for all 8 MJO phases. One key question addressed here is why does the extratropical Rossby wave train depart the subtropics at a longitude well to the east of the RWS. For all 8 MJO phases, it is found that the extratropical response over the North Pacific and North America is almost entirely due to the MJO convection over the western tropical Pacific. The RWS is excited within the first 24 hours after the model heating is turned on. For MJO phases 1-3 and 8, the RWS leads to the development of a cyclonic anomaly over southeast Asia via advection of the climatological absolute vorticity by the anomalous divergent wind in the subtropics and by horizontal convergence in the tropics. MJO phases 4-7 show opposite features. The resulting anomaly is then advected eastward by the climatological zonal wind toward the central Pacific, after which dispersion into the extratropics and the excitation of a Pacific/North American teleconnection pattern takes place.

Description: In common with many global models, the Met Office Unified Model (MetUM) climate simulations show large errors in Indian summer monsoon rainfall, with a wet bias over the equatorial Indian Ocean, a dry bias over India, and with too weak low-level flow into India. The representation of moist convection is a dominant source of error in global models, where convection must be parametrised, with the errors growing quickly enough to affect both weather and climate simulations. Here we use the first multi-week continental-scale MetUM simulations over India, with grid-spacings that allow explicit convection, to examine how convective parametrisation contributes to model biases in the region. Some biases are improved in the convection-permitting simulations with more intense rainfall over India, a later peak in the diurnal cycle of convective rainfall over land, and a reduced positive rainfall bias over the Indian Ocean. The simulations suggest that the reduced rainfall over the Indian Ocean leads to an enhanced monsoon circulation and transport of moisture into India. Increases in latent heating associated with increased convection over land deepen the monsoon trough and enhance water vapour transport into the continent. In addition, delayed continental convection allows greater surface insolation and, along with the same rain falling in more intense bursts, generates a drier land surface. This increases land-sea temperature contrasts, and further enhances onshore flow. Changes in the low-level water vapour advection into India are dominated by these changes to the flow, rather than to the moisture content in the flow. The results demonstrate the need to improve the representations of convection over both land and oceans to improve simulations of the monsoon.

Description: Increasing the size of the ensemble used in hybrid-variational assimilation methods has been shown to be beneficial, but is computationally expensive. This work sets out to see if similar improvements can be obtained, from the smaller ensemble, by better estimation of ensemble covariances. Methods for improving these are described and illustrated using a toy model. The optimal settings depend on the ensemble size, the criterion used to measure error, and errors in ensemble generation. A hybrid covariance, filtered by spectral localisation using wavebands and scale-dependent spatial localisation, is shown to perform well and robustly. In the cycled ensemble data-assimilation scheme used for NWP, another method of increasing the effective ensemble size in covariance estimates is the use of time-lagged and time-shifted perturbations. This is demonstrated to be effective in the Met Office's hybrid-4DEnVar system, both on its own and with the waveband and scale-dependent localisation. The best such combination performs nearly as well as an increase in ensemble size from 44 to 200.

Description: Extratropical cyclones experience vastly different genesis conditions at the first point of their tracks. A novel method is introduced to characterize this variability and classify genesis events by computing 30 diagnostic variables that describe the synoptic-scale environment of 16029 genesis events in the Northern Hemisphere extratropics, using ERA-Interim reanalyses from 2000-2011. These variables are referred to as precursors and include parameters characterizing upper-level forcing, low-level baroclinicity, thermodynamic stability, surface fluxes and moist processes. The genesis events spread over a large portion of the 30-dimensional precursor phase space and no obvious clusters occur, which highlights the high variability of cyclogenesis processes and indicates that they form in a continuum rather than a few distinct categories. A projection of the genesis events to the first two principle components (PC) of the precursor phase space allows reducing the dimensionality and introducing a meaningful segmentation of the genesis events in five classes. The first two PCs are characterized by upper-level forcing (e.g., the amplitude of the upper-level potential vorticity (PV) anomaly) and low-tropospheric diabatic processes (e.g., precipitation and diabatically-produced low-level PV), respectively. The first of the five identified classes constitutes the center of the PC1–PC2 phase space and represents average conditions. Composites reveal that the four classes of events characterized by large positive or negative scores of PC1 and PC2 occur in distinct and strongly differing flow regimes, characterized by the strength of the upper-level forcing, the structure of the upper-level jet, and the amplitude of low-level moist processes and baroclinicity. The four classes also have clearly differing geographical distributions. Many well-known cyclogenesis events fall within classes characterized by strong low-level moist processes with or without strong upper-level forcing. Also discussed are the robustness of the method and the linkage to classical concepts of cyclone classifications.

Description: The present article describes the sea surface temperature (SST) developments implemented in the Goddard Earth Observing System, Version 5 (GEOS-5) Atmospheric Data Assimilation System (ADAS). These are enhancements that contribute to the development of an atmosphere-ocean coupled data assimilation system using GEOS. In the current quasi-operational GEOS-ADAS, the SST is a boundary condition prescribed based on the OSTIA product, therefore SST and skin SST (Ts) are identical. This work modifies the GEOS-ADAS Ts by modeling and assimilating near sea surface sensitive satellite infrared (IR) observations. The atmosphere-ocean interface layer of the GEOS atmospheric general circulation model (AGCM) is updated to include near surface diurnal warming and cool-skin effects. The GEOS analysis system is also updated to directly assimilate SST-relevant Advanced Very High Resolution Radiometer (AVHRR) radiance observations. Data assimilation experiments designed to evaluate the Ts modification in GEOS-ADAS show improvements in the assimilation of radiance observations that extends beyond the thermal IR bands of AVHRR. In particular, many channels of hyperspectral sensors, such as those of the Atmospheric Infrared Sounder (AIRS), and Infrared Atmospheric Sounding Interferometer (IASI) are also better assimilated. We also obtained improved fit to withheld, in-situ buoy measurement of near-surface SST. Evaluation of forecast skill scores show marginal to neutral benefit from the modified Ts.

Description: Aircraft measurements of two cumulus clouds were made during the Ice and Precipitation Initiation in Cumulus campaign over the British Isles. The 18 May 2006 cloud had high concentrations of ice particles and conditions were conducive for the Hallett-Mossop (HM) process of secondary ice production, but the 13 July 2005 cloud had low concentrations. A bin-resolved cloud model was used to investigate several factors that are known to control the HM process using the observations of the two clouds. For the 2006 cloud, the model results show that the fast production of graupel by directly freezing of supercooled raindrops through collisional collection with ice particles was crucial to the activation of the HM process. Switching-off raindrop freezing led to much delayed and suppressed formation of graupel particles, and hence, a negligible HM process. Sensitivity studies were performed on the concentration of primary ice particles required to kick-start the HM process. It was found that a concentration of the first ice as low as 0.01 L − 1 could be sufficient, as long as there was a large enough concentration of cloud droplets (small and large) available when a significant number of graupel particles developed in the HM temperature zone. For the modelled 2005 cloud, the HM process did not operate effectively mainly because of the low concentration of supercooled raindrops and hence graupel. The HM process was also hindered by the relatively greater number of aerosols, and higher temperatures at cloud base and top.

Description: The decomposition of the Brier score into Reliability, Resolution and Uncertainty has become a standard method in forecast verification. In this note a very simple derivation of the familiar Brier score decomposition is presented. The Reliability and Resolution terms can be calculated as average Brier score differences between the issued forecast, the recalibrated forecast and the climatological reference forecast. The result suggests a simple way to calculate similar decompositions for arbitrary verification scores, and that recalibration methods and reference forecasts can be chosen more flexibly than is generally appreciated. A new decomposition of the continuous ranked probability score (CRPS) is proposed.

Description: Observing System Experiments have been used to evaluate the forecast impact of sea level pressure observations from drifting buoys. Two seasons have been selected with different synoptic weather characteristics, but similar amount of buoy observations. Control and denial experiments were performed with and without assimilating drifting buoys sea level pressure observations. The denial experiments withdraw around 95% of the total surface pressure measurements from buoys; the remnant 5% is provided by moored buoys. Changes in the forecast performance are evaluated in terms of root mean-squared error and anomaly correlation scores. Adjoint diagnostic tools are also used to estimate the observations contribution to the analysis and forecast. The lack of drifter surface pressure observations has large and significant detrimental impact on the mean sea level pressure, temperature and wind fields. The signal is detectable not only near to the surface but throughout the troposphere up to 250 hPa. Drifter surface pressure observations contribute to decrease the total global forecast error by approximately 3%. In particular, case studies reveal that drifting buoy observations can be especially important to reduce the forecast error on complex or rapidly evolving cyclogenesis. All the diagnostics performed indicate that drifting buoys are essential ingredients of WMO’s Global Observing System.

Description: A leading Data Assimilation (DA) technique in meteorology is 4DVAR which relies on the Tangent Linear Model (TLM) of the non-linear model and its adjoint. The difficulty of building and maintaining traditional TLMs and adjoints of coupled ocean-wave-atmosphere-etc models is daunting. On the other hand, coupled model ensemble forecasts are readily available. Here, we show how an ensemble forecast can be used to construct an accurate Local Ensemble TLM (LETLM) and adjoint of the entire coupled system. The method features a local influence region containing all the variables that could possibly influence the time evolution of some target variable(s) near the center of the region. We prove that high accuracy is guaranteed provided that (i) the ensemble perturbations are governed by linear dynamics, and (ii) the number of ensemble members exceeds the number of variables in the influence region. The approach is illustrated in a simple coupled model. This idealized coupled model has some realistic features including reasonable predictability limits in the upper-atmosphere, lower-atmosphere, upper-ocean and lower-ocean of 10, 96, 160 and 335 days, respectively. In addition, the length scale of eddies in the ocean is about one fifth of those in the atmosphere. The easy manner in which the adjoint is obtained from the LETLM is also described and illustrated by demonstrating how the LETLM adjoint predicts the high sensitivity of oceanic boundary layer evolution to changes in the atmosphere. Finally, the feasibility of LETLMs for 4DVAR is demonstrated. Specifically, a case is considered with a 5 day data assimilation window in which non-linear terms play a significant role in the evolution of forecast error; it is shown that the posterior mode delivered by 4DVAR with an LETLM, its adjoint and 10 outer loops approximately recovers the true state in spite of a spatially sparse observational network.