Search

Diurnal warm layers develop in the upper ocean on sunny days with low surface wind speeds. They rectify intraseasonal sea‐surface temperatures (SSTs), potentially impacting intraseasonal weather patterns such as the Madden–Julian Oscillation (MJO). Here we analyse 15‐lead‐day forecast composites of coupled ocean–atmosphere and atmosphere‐only numerical weather prediction (NWP) models of the UK Met Office to reveal that the presence of diurnal warming of SST (dSST) leads to a faster MJO propagation in the coupled model compared with the atmosphere‐only model. To test the feedback between the MJO and the dSST, we designed a set of experiments with instantaneous vertical mixing over the top 5 or 10 m$$ 10\;\mathrm{m} $$ of the ocean component of the coupled model. Weaker dSST in the mixing experiments leads to a slower MJO over 15 lead days. The dSST produces a 3%$$ 3\% $$ increase in the MJO phase speed between the coupled and the atmosphere‐only model. An additional 5%$$ 5\% $$ increase is found for other coupling effects, unrelated to the dSST. A two‐way feedback manifests in the coupled model over the 15 lead days of the forecast between the MJO and the dSST. The MJO regime dictates the strength of the dSST and the dSST rectifies the intraseasonal anomalies of SST in the coupled model. Stronger dSST in the coupled model leads to stronger intraseasonal anomalies of SST. The MJO convection responds to these SSTs on a seven‐lead‐day timescale, and feeds back into the SST anomalies within the next three lead days. Overall, this study demonstrates the importance of high vertical resolution in the upper ocean for predicting the eastward propagation of the MJO in an NWP setting, which is potentially impactful for seasonal predictions and climate projections, should this feedback be unrepresented in the models. [ABSTRACT FROM AUTHOR]

The diurnal warm layer in the upper ocean develops during low surface winds and high incoming solar radiation conditions, often increasing sea‐surface temperatures (SSTs) by up to 1°C. The suppressed phase of the Madden–Julian Oscillation (MJO) favours the formation of such a layer. Here, we analyse the coupled ocean–atmosphere and atmosphere‐only numerical weather prediction systems of the UK Met Office to reveal that important differences arise from the representation of the diurnal warm layer in the coupled model. Though both models are skilful in predicting the MJO to at least a 7‐day lead time, the coupled model predicts approximately 12% faster MJO RMM phase speed propagation than the atmosphere‐only model due to the ability to resolve diurnal warming in the upper ocean that rectifies onto MJO‐associated SST anomalies. The diurnal warming of SST (dSST) in the coupled model leads to an increase in daily mean SST compared with the atmosphere‐only model persisted foundation SST. The strength of the dSST in the coupled model is modulated by MJO conditions. During suppressed MJO conditions on lead day 1, the dSST is enhanced, leading to 0.2°C warmer daily mean MJO‐associated SST anomalies and increased convection in the coupled model by lead day 7. During active MJO convection, the dSST is suppressed, leading to 0.1°C colder MJO‐associated SST anomalies in the coupled model and reduced convection by lead day 7. This variability in dSST further amplifies the MJO propagation speed, underlining the importance of the two‐way feedback between the MJO and the diurnal cycle of SST and the need to accurately represent this process in coupled models. [ABSTRACT FROM AUTHOR]

A multi-season convection permitting regional climate simulation of the Maritime Continent using the Met Office Unified Model with 2.2-km grid spacing is presented and evaluated. The simulations pioneer the use of atmosphere-ocean coupling with the multi-column K profile parametrisation (KPP) mixed layer ocean model in atmospheric convection permitting climate simulations. Comparisons are made against a convection parametrised simulation in which it is nested, and which in turn derives boundary conditions from ERA5 reanalysis. This paper describes the configuration, performance of the mean state and variability of the two simulations compared against observational datasets. The models both have minor sea surface temperature (SST) and wet precipitation biases. The diurnal cycle, representation of equatorial waves and relationship between SST and precipitation are all improved in the convection permitting model compared to the convection parametrised model. The MJO is present in both models with a faster than observed propagation speed. However, it is unclear whether fidelity of the MJO simulation is inherent to the model or whether it predominantly arises from the forcing at the boundaries. [ABSTRACT FROM AUTHOR]