South American Intraseasonal Oscillation: EOF and Neural Network Approaches

The relationship between the Madden–Julian oscillation (MJO) and the seasonal cycle of the intraseasonal rainfall variability in the Amazon Basin (AB) are analysed using band‐pass‐filtered gauge‐based gridded rainfall data for the 1980–2009 period. Intraseasonal events (IE) have been defined and selected based on extreme values of the first principal component (PC1) time series, which comes from the empirical orthogonal function (EOF) analysis applied to the filtered rainfall data over the AB. A total of 132 IEs were identified with an average of approximately five events per year. About 25% of the total IEs in the Amazon region are not restricted to the eastwards propagating equatorially confined MJO and other mechanisms (e.g., through Rossby wave trains in the Southern Hemisphere) might play an important role. In addition, we find that the incomplete IEs (events that do not evolve through a complete life cycle) are associated with suppressed rainfall conditions over tropical South America. The development of the IEs over the AB, when compared with the different phases of the MJO index, shows a coherent relationship, where convective‐based indices are able to better account their evolution. On a global scale, the upper‐tropospheric patterns and the rainfall composites based on the PC1 time series show that the MJO is one of the main atmospheric modulator mechanisms of the intraseasonal rainfall variability over the Amazon region throughout the annual cycle. It is found that the intraseasonal variability is particularly important during the austral winter, when the percentage contribution with respect to the mean daily seasonal precipitation over some Amazon regions can reach 50%.

The spatial and temporal structures of the leading modes of intraseasonal atmospheric variability over Central Africa using outgoing longwave radiation (OLR) and one-degree daily Global Precipitation Climatology Project (1DD GPCP) data are compared. A spectral analysis indicates that in the two datasets, the intraseasonal variability is dominated by 20–70-day period bands with center near 40–50 days. Results from empirical orthogonal function (EOF) analysis have shown that three main spatial structures characterize the leading modes of 20–70-day intraseasonal oscillations (ISO), but they differ in the two data by the amount of variance explained by each mode. For both time series, the power spectra of all the three principal components peak around 40–50 days, indicating Madden–Julian oscillation (MJO) signal. Moreover, the cross correlation computed among the principal components indicates that there exists a relatively high positive correlation between the EOFs of similar spatial loadings. For the two datasets, an index of MJO strength was built by averaging the 30–50-day power for each day and each EOF mode. A plot of the two indices revealed a coherent onset, peak, and decay of ISO activity but with different amplitudes. The correlation coefficients computed between the ISO indices corresponding to each mode revealed that they are highly correlated, especially for the annual mean time series where the correlation coefficient reaches up to 0.88 for the Tanzanian mode. Overall, the analysis showed that the leading modes are similar, but the 1DD GPCP loadings have better spatial localization, when compared with those of OLR datasets.

The influence of Madden–Julian oscillation (MJO) is examined on intraseasonal rainfall variability during the wet season (November–April) by using the real-time multivariate (RMM) MJO index, ERA5 reanalysis, and daily observed rainfall dataset from 26 stations in Saudi Arabia for the period 1985–2021. The MJO 8 phases are categorized into wet (phases 1, 2, 7 and 8) and dry (phases 3, 4, 5, and 6) based on the Saudi Arabian intraseasonal rainfall characteristics associated with MJO phases. It is observed that 41% (46%) of total (extreme) rainfall events occur during the MJO wet phases, while only 23% (18%) of such events occur during MJO dry phases. The intraseasonal variability signals are isolated from daily dataset by applying a 30- to 90-day period bandpass filter. The analyses are validated by constructing composites of daily filtered precipitation anomalies during MJO 8 phases. The physical mechanism indicates that the significant intraseasonal wetter conditions are linked with enhanced easterly and southeasterly moisture convergence over Saudi Arabia from the Arabian Sea. The atmospheric cyclonic circulation anomalies during the wet phases favor more moisture convergence and vertical moisture advection, which may lead to enhanced convection and rainfall. However, during the dry phases, anticyclonic circulation anomalies enhance moisture divergence and reduce vertical moisture advection and consequently suppress the convection and rainfall activity over Saudi Arabia. The analyses show that the intraseasonal rainfall variability over Saudi Arabia is significantly influenced by the MJO during the wet season. These findings have important implications for sub-seasonal rainfall forecasting in Saudi Arabia.

Rainfall over Africa varies across timescales of a few days to several weeks due to several tropical and extratropical modes of variability. Excessive rains or prolonged drought regularly result in natural disasters and have thus a severe impact on the local economy, agriculture, spread of diseases, and entire ecosystems. The dynamical nature of the atmosphere allows the existence of planetary balanced modes, which are called Rossby waves, and smaller-scale unbalanced inertio-gravity (IG) waves. The former, which are more rotational, arise from the horizontal pressure gradient force, while for the latter gravity acts as the restoring force, making their flow pattern more divergent. The main source of variability in the extratropics stems from Rossby waves. At the equator, further types of convectively coupled equatorial waves (CCEWs) exist, namely Kelvin and mixed Rossby-gravity (MRG) waves. As the slowest intraseasonal tropical mode, the Madden–Julian Oscillation (MJO), which is related to Kelvin and Rossby waves, acts on a timescale of 30 to 90 days. Although it is primarily a planetary mode, the MJO has a specific “flavor” over the African continent. On the short intraseasonal timescale of 10 to 25 days, equatorial Rossby (ER) waves and the internal modes of the West African monsoon, the quasi-biweekly zonal dipole (QBZD) and the Sahel mode, modulate rainfall. On the synoptic timescale of a few days to a week, African easterly waves (AEWs) are a dominant mode over West Africa, whereas Kelvin waves predominantly modulate rainfall over equatorial Africa. Extratropical influences on northern and southern Africa manifest themselves in Rossby wave trains, which modulate synoptic to intraseasonal rainfall through tropical rainfall plumes, cold air surges, and upper-tropospheric dry air intrusions. Furthermore, the Saharan heat low (SHL) acts as a link between the northern hemispheric extratropics and tropics. Finally, the Indian monsoon, the Atlantic, Indian, and the Pacific Oceans can remotely affect the intraseasonal variability of African rainfall. Forecasting synoptic to intraseasonal rainfall variability is an integral part of seamless prediction between the weather and climate regimes. In the early 21st century, numerical weather prediction (NWP) systems can forecast larger intraseasonal signals such as the MJO several weeks into the future, but they still struggle to forecast shorter scale features reliably. Besides NWP, statistical models can successfully forecast intraseasonal variability of rainfall. Due to the relevance of synoptic to intraseasonal rainfall variability for African societies, early warning systems (EWSs) have been developed to mitigate impacts.

<p>The impacts of the Madden-Julian Oscillation (MJO) on the South American monsoon season (December, January, and February – DJF) and their possible changes during positive (El Niño – EN) and negative (La Niña – LN) phases of the El Niño-Southern Oscillation (ENSO) are analyzed in the UK Met Office Unified Model Global Ocean Mixed Layer configuration (MetUM-GOML3). Experiments sixty years long, with and without ENSO cycle, considering lower (200 km) and higher (90 km) spatial resolution, are performed to assess if the ENSO influences MJO characteristics such as the phase distribution, propagation, convection, and teleconnections to South America (SA). The analyzes use daily continental precipitation data, daily global outgoing longwave radiation (OLR), and zonally asymmetric streamfunction computed with daily wind data. Composites of daily filtered anomalies in the 20-90 day band are assessed. Simulations without ENSO show (1) an established MJO extratropical teleconnection triggered by enhanced convection in the central-east subtropical South Pacific (SP) (source region), and its strongest impact on precipitation over SA in phase 8, earlier than in observations (phase 1); (2) an extratropical teleconnection via Rossby wave train, triggered by suppressed convection over the same region, with strongest impact on precipitation over SA in phase 4, with opposite sign; (3) increased horizontal resolution enhances the MJO convection and the anomalous circulation-precipitation dipole over SA, mainly over subtropical SA. However, the extratropical teleconnections via Rossby wave train at upper levels are slightly shifted east at higher resolution due to an enhanced SA westerly jet with respect to the lower resolution. The ENSO affects the basic state and the MJO convective anomalies, which modulate the MJO teleconnections and their impacts on SA in simulations with ENSO cycles. The EN (LN) basic state improves (worsens) MJO eastward propagation and its convection. However, both EN and LN states produce enhanced convection over the source region in phases 8+1, while suppressed convection over the same region in phase 4 is simulated only in EN. The extratropical teleconnections via Rossby wave train (phases 8+1, 4) and their impacts are stronger under ENSO with respect to those in simulations without ENSO. Hence, both ENSO states in the model generate forcing in the central-east subtropical SP that more efficiently triggers teleconnections than simulations without ENSO, indicating nonlinear ENSO effects on MJO anomalies over SA. As the MJO and its teleconnections improve during ENSO, other coupled global climate models (CGCMs) may reproduce these features, and subseasonal to seasonal (S2S) predictions to SA may be better forecast when ENSO and MJO peak in DJF, though the MJO impacts in phase 1 remain challenging.</p><p>Keywords: Coupled global models; ENSO-MJO Interaction; South American monsoon; Teleconnections.</p>

The space–time structure of intraseasonal (10–90 day) rainfall variability in the western tropical Pacific is studied using daily 3B42 TRMM and ERA-Interim reanalysis data for the period 1998–2014. Empirical orthogonal function (EOF) analysis of 10–90-day filtered daily rainfall anomalies identifies two leading modes in both May–October and November–April; together these modes explain about 11%–12% of the total intraseasonal variance over the domain in both seasons and up to 60% over large areas of the western Pacific in both climatological periods. The two leading modes in May–October are linearly related to each other and both are well correlated with the Madden–Julian oscillation (MJO) indices. Although the two leading EOF modes in November–April are linearly independent of each other, both show statistically significant correlations with the MJO. The phase composites of 30–80-day filtered data show that the two leading modes are associated with strong eastward and northward propagation of rainfall anomalies in May–October, and eastward and southward propagation of rainfall anomalies in November–April. The eastward propagation of rainfall anomalies in both seasons and southeastward propagation related with EOF2 in November–April is linked to the development of low-level moisture flux convergence ahead of the active convection. Similarly, the northward propagation in May–October is also connected with low-level moisture flux convergence, but surface wind and evaporation variations are also important. The wind–evaporation–SST feedback mechanism drives the southeastward propagation of rainfall anomalies associated with EOF1 in November–April. The different mechanisms for southeastward propagation associated with two leading modes in November–April suggest dynamically different relations with the MJO.

<p><span>We investigate the impact of air-sea coupling on the simulation of the intraseasonal variability of rainfall over the South Pacific using the ECHAM5 atmospheric general circulation model coupled with Snow-Ice-Thermocline (SIT) ocean model. We compare the fully coupled simulation with two uncoupled simulations forced with sea surface temperature (SST) climatology and daily SST from the coupled model. The intraseasonal rainfall variability over the South Pacific Convergence Zone (SPCZ) is reduced by 17% in the uncoupled model forced with SST climatology and increased by 8% in the uncoupled simulation forced with daily SST. The coupled model best simulates the key characteristics of the two intraseasonal rainfall modes of variability in the South Pacific, as identified by an Empirical Orthogonal Function (EOF) analysis. The spatial structure of the two EOF modes in all three simulations is very similar, suggesting these modes are independent of air-sea coupling and primarily generated by the dynamics of the atmosphere. The southeastward propagation of rainfall anomalies associated with two leading rainfall modes in the South Pacific depends upon the eastward propagating </span><span>Madden-Julian Oscillation (</span><span>MJO</span><span>)</span><span> signals over the Indian Ocean and western Pacific. Air-sea interaction seems crucial for such propagation as both eastward and southeastward propagations substantially reduced in the uncoupled model forced with SST climatology. Prescribing daily SST from the coupled model improves the simulation of both eastward and southeastward propagations in the uncoupled model forced with daily SST, showing the role of SST variability on the propagation of the intraseasonal variability, but the periodicity differs from the coupled model. The change in the periodicity is attributed to a weaker SST-rainfall relationship that shifts from SST leading rainfall to a nearly in-phase relationship in the uncoupled model forced with daily SST.</span></p>

The spatial and temporal structures of the intraseasonal atmospheric variability over central Africa is investigated using 2.5° × 2.5° daily outgoing longwave radiation (OLR) and National Centers for Environmental Prediction (NCEP) Reanalysis zonal winds for the period 1980–2010. The study begins with an overview of the Central African rainfall regime, noting in particular the contrast amongst Western and Eastern parts, with different topography and surface conditions features. The annual mean rainfall and OLR over the region revealed a zone of intense convective activity centered on the equator near 30°E, which extends southward and covers almost all the Congo forest. The annual cycle of rainfall reflects the classical bi-annual shift of Inter-Tropical Convergence Zone across the equatorial belt, between 10°S and 10°N. The result of the empirical orthogonal functions (EOFs) analysis has shown that the three leading EOF modes explain about 45 % of total intraseasonal variability. The power spectra of all the three corresponding principal components (PCs) peak around 45–50 days, indicating a Madden–Julian Oscillation (MJO) signal. The first mode exhibits high positive loadings over Northern Congo, the second over Southern Ethiopia and the third over Southwestern Tanzania. The PCs time series revealed less interannual modulation of intraseasonal oscillations for the Congo mode, while Ethiopian and Tanzanian modes exhibit strong interannual variations. Hovmoller plots of OLR, 200 and 850 hPa NCEP zonal winds found the eastward propagating features to be the dominant pattern in all the three times series, but this propagation is less pronounced in the OLR than in the 850 and 200 hpa zonal wind anomalies. An index of MJO strength was built by averaging the 30–50 day power for each day. A plot of MJO indices and El Nino Southern Oscillation (ENSO) cycle confirm a strong interannual modulation of MJO over Eastern central Africa partially linked with the ENSO events (El Nino and La Nina). Strong MJO activity is observed during La Nina years or during ENSO-neutral years, while weak or absent MJO activity is typically associated with strong El Nino episodes.

Fifteen years (1987–2001) of rain gauge-based data are used to describe the intraseasonal rainfall variability over tropical Brazil and its associated dynamical structure. Wavelet analysis performed on rainfall time series showed significant peaks centered roughly in periods of 30–70 days, particularly in the eastern southeastern Amazon and northern northeast Brazil. A significant enhancement of precipitation with maximum anomalies in a northeastward oriented band over tropical Brazil is evidenced from empirical orthogonal function (EOF) analysis of 30–70-day filtered rainfall anomalies during rainy season (January to May). Lagged/lead composites revealed that, on a global scale, the Madden–Julian oscillation (MJO) is the main atmospheric-mechanism modulator of the pluviometric variations on intraseasonal timescale in the eastern Amazon and northeast Brazil. A coherent northward expansion of rainfall across tropical Brazil is evident during the passage of MJO over South America. Regionally, the establishment of a quasi-stationary deep convection band triggered by the simultaneous manifestation of south Atlantic convergence zone (SACZ) and intertropical convergence zone (ITCZ) explains the intensified rainfall over these regions. Such regional mechanisms are dynamically embedded within the eastward-propagating MJO-related large-scale convective envelope along tropical South America/the Atlantic Ocean. These features occur in association with a significant intraseasonal evolution of the lower-level wind and sea-surface temperature (SST) patterns, particularly in the Atlantic Ocean, including a coherent dynamical connection with atmospheric circulation, deep convective activity over South America and rainfall over tropical Brazil. Copyright © 2006 Royal Meteorological Society

Observational evidences linking tropical and extratropical wave disturbances in organizing precipitation over the Amazon basin (AB) are presented. The time scale of the wave disturbances considered in this thesis are from intraseasonal (low-and high-frequency intraseasonal) to subseasonal. The results are based on observed, satellite, and reanalysis datasets. The results suggest that on 30-70-day low-frequency, the Madden-Julian oscillation (MJO) is one of the main atmospheric mechanisms modulating intraseasonal rainfall variability over the Amazon region, playing an important role, particularly during the austral winter. Despite of the fact that the AB region extends over tropical region, the 30-70-day low-frequency intraseasonal are not strictly associated with the forcing produced by the equatorially propagating MJO. In addition, it is suggested that convective-based MJO indices are able to better account for the influence of the MJO on intraseasonal precipitation. The 10-30-day high-frequency intraseasonal (HFIS) activity account for about 22% of the total Amazon low-frequency intraseasonal rainfall events and it is part of large-scale pattern observed over South America during the onset of austral rainfall season (October-November) and during the dry season (May-September). During the peak of the wet season (December-April), the Northwest-Southeast orientation of the precipitation seems to be influenced by Rossby wave trains different to previously documented, with a southeastern South America (SESA)-South America convergence zone (SACZ) pattern.

Large intraseasonal rainfall variations are identified over the southern South China Sea (SSCS), tropical southeastern Indian Ocean (SEIO), and east coast of the Philippines (EPHI) in boreal winter. The present study contrasts origins and propagations and investigates interrelations of intraseasonal rainfall variations on the 10–20- and 30–60-day time scales in these regions. Different origins are identified for intraseasonal rainfall anomalies over the SSCS, SEIO, and EPHI on both time scales. On the 10–20-day time scale, strong northerly or northeasterly wind anomalies related to the East Asian winter monsoon (EAWM) play a major role in intraseasonal rainfall variations over the SSCS and EPHI. On the 30–60-day time scale, both the intraseasonal signal from the tropical Indian Ocean and the EAWM-related wind anomalies contribute to intraseasonal rainfall variations over the SSCS, whereas the EAWM-related wind anomalies have a major contribution to the intraseasonal rainfall variations over the EPHI. No relation is detected between the intraseasonal rainfall variations over the SEIO and the EAWM on both the 10–20-day and 30–60-day time scales. The anomalies associated with intraseasonal rainfall variations over the SSCS and EPHI propagate northwestward and northeastward, respectively, on the 10–20- and 30–60-day time scales. The intraseasonal rainfall anomalies display northwestward and northward propagation over the Bay of Bengal, respectively, on the 10–20- and 30–60-day time scales.

The Madden–Julian Oscillation (MJO) is the major mode of intraseasonal variability (30–60 days) in the tropics, having large rainfall impacts globally, and possibly on southern Africa. However, the latter impact is not well understood and needs to be further explored. The life cycle of the MJO, known to be asymmetric, has been nevertheless analyzed usually through methods constrained by both linearity and orthogonality, such as empirical orthogonal function analysis. Here we explore a non-linear classification method, the self-organizing map (SOM), a type of artificial neural network used to produce a low-dimensional representation of high-dimensional datasets, to capture more accurately the life cycle of the MJO and its global impacts. The classification is applied on intraseasonal anomalies of outgoing longwave radiation within the tropical region over the 1980–2009 period. Using the SOM to describe the MJO is a new approach, complimentary to the usual real-time multivariate MJO index. It efficiently captures this propagative phenomenon and its seasonality, and is shown to provide additional temporal and spatial information on MJO activity. For each node, the subtropical convection is analyzed, with a particular focus on the southern Africa region. Results show that the convection activity over the central tropical Indian Ocean is a key factor influencing the intraseasonal convective activity over the southern African region. Enhanced (suppressed) convection over the central Indian Ocean tends to suppress (enhance) convection over the southern African region with a 10-day lag by modulating the moisture transport.

This study presents the simulation of the Madden–Julian oscillation (MJO) in the NCAR CCM3 using a modified Zhang–McFarlane convection parameterization scheme. It is shown that, with the modified scheme, the intraseasonal (20–80 day) variability in precipitation, zonal wind, and outgoing longwave radiation (OLR) is enhanced substantially compared to the standard CCM3 simulation. Using a composite technique based on the empirical orthogonal function (EOF) analysis, the paper demonstrates that the simulated MJOs are in better agreement with the observations than the standard model in many important aspects. The amplitudes of the MJOs in 850-mb zonal wind, precipitation, and OLR are comparable to those of the observations, and the MJOs show clearly eastward propagation from the Indian Ocean to the Pacific. In contrast, the simulated MJOs in the standard CCM3 simulation are weak and have a tendency to propagate westward in the Indian Ocean. Nevertheless, there remain several deficiencies that are yet to be addressed. The time period of the MJOs is shorter, about 30 days, compared to the observed time period of 40 days. The spatial scale of the precipitation signal is smaller than observed. Examination of convective heating from both deep and shallow convection and its relationship with moisture anomalies indicates that near the mature phase of the MJO, regions of shallow convection developing ahead of the deep convection coincide with regions of positive moisture anomalies in the lower troposphere. This is consistent with the recent observations and theoretical development that shallow convection helps to precondition the atmosphere for MJO by moistening the lower troposphere. Sensitivity tests are performed on the individual changes in the modified convection scheme. They show that both change of closure and use of a relative humidity threshold for the convection trigger play important roles in improving the MJO simulation. Use of the new closure leads to the eastward propagation of the MJO and increases the intensity of the MJO signal in the wind field, while imposing a relative humidity threshold enhances the MJO variability in precipitation.

Sea surface temperature (SST) variability at intraseasonal time scales across the Indonesian Seas during January 1998–mid-2012 is examined. The intraseasonal variability is most energetic in the Banda and Timor Seas, with a standard deviation of 0.4°–0.5°C, representing 55%–60% of total nonseasonal SST variance. A slab ocean model demonstrates that intraseasonal air–sea heat flux variability, largely attributed to the Madden–Julian oscillation (MJO), accounts for 69%–78% intraseasonal SST variability in the Banda and Timor Seas. While the slab ocean model accurately reproduces the observed intraseasonal SST variations during the northern winter months, it underestimates the summer variability. The authors posit that this is a consequence of a more vigorous cooling effect induced by ocean processes during the summer. Two strong MJO cycles occurred in late 2007–early 2008, and their imprints were clearly evident in the SST of the Banda and Timor Seas. The passive phase of the MJO [enhanced outgoing longwave radiation (OLR) and weak zonal wind stress) projects on SST as a warming period, while the active phase (suppressed OLR and westerly wind bursts) projects on SST as a cooling phase. SST also displays significant intraseasonal variations in the Sulawesi Sea, but these differ in characteristics from those of the Banda and Timor Seas and are attributed to ocean eddies and atmospheric processes independent from the MJO.

A gridded daily rainfall dataset prepared from observations at 3700 stations is used to analyze the intraseasonal and interannual variability of the summer monsoon rainfall over India. It is found that the major drought years are characterized by large-scale negative rainfall anomalies covering nearly all of India and persisting for the entire monsoon season. The intraseasonal variability of rainfall during a monsoon season is characterized by the occurrence of active and break phases. During the active phase, the rainfall is above normal over central India and below normal over northern India (foothills of the Himalaya) and southern India. This pattern is reversed during the break phase. It is found that the nature of the intraseasonal variability is not different during the years of major droughts or major floods. This suggests that a simple conceptual model to explain the interannual variability of the Indian monsoon rainfall should consist of a linear combination of a large-scale persistent s...