Cyclostationary Empirical Orthogonal Function Analysis Research Articles

The dominant modes of recent sea level variability from 1993 to 2020 in the China Seas

A better understanding of internal climatic variability is essential for estimating and predicting regional sea level rise in the coming decades. Based on the satellite altimeter data, the cyclostationary empirical orthogonal function (CSEOF) analysis is used to extract and examine the seasonal signal, trend, El Niño-Southern Oscillation (ENSO)-related, and low-frequency mode of sea level anomaly in the China Seas from 1993 to 2020. The interannual changes in the annual cycle of sea level were significantly influenced by the mean surface pressure and the 10-m zonal wind component. Furthermore, strong El Niño events (1997/1998) can significantly affect seasonal fluctuations through precipitation and runoff. By recombining the loading vector and principal component time series of the trend mode, we estimate the sea level trends to be about 3.63 ± 0.37 mm/yr. The South China Sea, followed by the Yellow Sea, is where ENSO's effects on sea level fluctuations are most noticeable. The 2015–2016 El Niño, which was a combination of the Central Pacific (CP) ENSO Eastern Pacific (EP) ENSO, had average sea levels that were much lower than those of the 1997–1998 (CP ENSO) and 2009–2010 (CP ENSO). The Pacific Decadal Oscillation (PDO) can significantly affect the low frequency mode, but other sources also contribute to this mode. The South China Sea, especially to the west of Luzon Island, exhibits the highest level of variability in the low-frequency mode. Finally, our work shows that, in contrast to global or oceanic scales, regional CSEOF analysis does, in fact, capture regional sea level variability more accurately and detailed.

Spatiotemporal patterns of groundwater over South Korea

Understanding the large-scale spatiotemporal pattern of multi-depth groundwater levels is critical to develop water management plans and policies for sustainable ecological and social prosperity, which are still lacking. Here, we investigate three major spatiotemporal modes of groundwater levels from ∼200 groundwater monitoring stations over the southern Korean Peninsula (2009–2020), using the Cyclostationary empirical orthogonal function analysis. The first two major modes are associated with the seasonality of recharge and discharge and groundwater use during the 2016/17 drought, which explained half of the total variance. The third major mode indicated a decreasing trend of deep groundwater levels over the western Korean Peninsula, where key administrative and authority offices have been relocated via balanced national land development policies. Furthermore, at least three million Koreans over this region likely experience groundwater depletion by the 2080s. Observational evidence of emerging groundwater depletion suggests a window of opportunity for pre-emptive groundwater management plans.

Impacts of Climate Oscillation on Offshore Wind Resources in China Seas

The long-term stability and sustainability of offshore wind energy resources are very important for wind energy exploration. In this study, the Cyclostationary Empirical Orthogonal Function (CSEOF) method, which can determine the time varying spatial distributions and long-term fluctuations in the cyclostationary geophysical process, was adopted to investigate the geographical and temporal variability of offshore wind resources in China Seas. The CSEOF analysis was performed on wind speeds at 70 m height above the sea surface from a validated combined Quick Scatterometer (QuikSCAT) and Advanced Scatterometer (ASCAT) wind product (2000–2016) with high spatial resolution of 12.5 km, and Climate Forecast System Reanalysis (CFSR) wind data (1979–2016) with a grid size of 0.5° × 0.5°. The decomposition results of the two datasets indicate that the first CSEOF mode represents the variability of wind annual cycle signal and contributes 77.7% and 76.5% to the wind energy variability, respectively. The principal component time series (PCTS) shows an interannual variability of annual wind cycle with a period of 3–4 years. The second mode accounts for 4.3% and 4.7% of total wind speed variability, respectively, and captures the spatiotemporal contribution of El Niño Southern Oscillation (ENSO) on regional wind energy variability. The correlations between the mode-2 PCTS of scatterometer or CFSR winds and the Southern Oscillation Index (SOI) are greater than 0.7, illustrating that ENSO has a significant impact on China’s offshore wind resources. Moreover, the mode-1 or mode-2 spatial pattern of CFSR winds is basically consistent with that of scatterometer data, but CFSR underestimates the temporal variability of annual wind speed cycle and the spatial changes of wind speed related to ENSO. Compared with reanalysis data, scatterometer winds always demonstrate a finer structure of wind energy variability due to their higher spatial resolution. For ENSO events with different intensities, the impact of ENSO on regional wind resources varies with time and space. In general, El Niño has reduced wind energy in most regions of China Seas except for the Bohai Sea and Beibu Bay, while La Niña has strengthened the winds in most areas except for the Bohai Sea and southern South China Sea.

Identifying ENSO-related interannual and decadal variability on terrestrial water storage

We apply two statistical techniques to satellite measurements to identify a relationship between terrestrial water storage (TWS) and El Niño-Southern Oscillation (ENSO). First, we modified and used the least-squares regression of a previous study using longer records. Second, we applied a cyclostationary empirical orthogonal function analysis (CSEOF). Although the CSEOF technique is distinct from the least-squares regression in that it does not consider proxies, each method produces two modes (decadal and interannual), showing consistency with each technique in spatial pattern and its evolution amplitudes. We also compared the results obtained by the two methods for thirty watersheds, of which five watersheds were compared with previous studies. The combination of the two modes explains the total variance in most watersheds showing the role that interannual and decadal ENSO-related signals in understanding terrestrial water storage variability. The results show that the decadal mode, along with the interannual mode, also plays an important role in describing the local TWS.

Observation of aerosol size distribution and new particle formation under different air masses arriving at the northwesternmost South Korean island in the Yellow Sea

Particle number size distribution (PNSD) was measured with scanning mobility particle sizers from January 2014 to December 2016 at the Baengnyeong-do Comprehensive Monitoring Observatory, located on a rural island between China and the Korean Peninsula. Five different typical air mass transport pathway types were identified using the fuzzy c-means clustering method. Aerosols from the Shandong Province (Type 3) and South Korea (Seoul Metropolitan Area; Type 4) generally showed larger modal diameters than those transported from Mongolia and West Manchuria through the free troposphere (Type 1), which is attributable to the coagulation and condensational growth of preexisting particles. New particle formation (NPF) events, identified from the first mode of cyclostationary empirical orthogonal function analysis of the PNSD were observed in 42.2% (strong NPF [16.9%], weak NPF [25.3%]) of the days included in this study. Strong NPF events were frequently observed from October to March under the influence of pristine free tropospheric air masses transported along the Type 1 pathway, which can be characterized as NPF supported by meteorological conditions and a small condensation sink (3.77 × 10−3 s−1) under a limited precursor gas supply (i.e., SO2). Some occasional strong NPF events occurred under the influence of Type 3 air masses, primarily in spring (April to May), with a relatively high particle growth rate (6.95 nm h−1) and condensation sink (6.74 × 10−3 s−1), attributed to the relatively high preexisting particle concentration and abundant gaseous precursors.

Influence of Northern Hemispheric Winter Warming on the Pacific Storm Track

Effect of global warming on the sub-seasonal variability of the Northern Hemispheric winter (NDJFM) Pacific storm-track (PST) activity has been investigated. Previous studies showed that the winter-averaged PST has shifted northward and intensified, which was explained in terms of energy exchange with the mean field. Effect of global warming exhibits spatio-temporal heterogeneity with predominance over the Arctic region and in the winter season. Therefore, seasonal averaging may hide important features on sub-seasonal scales. In this study, distinct sub-seasonal response in storm track activities to winter Northern Hemispheric warming is analyzed applying cyclostationary empirical orthogonal function analysis to ERA5 data. The key findings are as follows. Change in the PST is not uniform throughout the winter; the PST shifts northward in early winter (NDJ) and intensifies in late winter (FM). In early winter, the combined effect of weakened baroclinic process to the south of the climatological PST and weakened barotropic damping to the north is responsible for the northward shift. In late winter, both processes contribute to the amplification of the PST. Further, change in baroclinic energy conversion is quantitatively dominated by eddy heat flux, whereas axial tilting of eddies is primarily responsible for change in barotropic energy conversion. A close relationship between anomalous eddy heat flux and anomalous boundary heating, which is largely determined by surface turbulent heat flux, is also demonstrated.

Relationship between submicron particle formation and air mass history observed in the Asian continental outflow at Gosan, Korea, during 2008–2018

We investigated the impact of air mass origin on submicron particle size distributions and new particle formation (NPF) events using 10-year measurements at Gosan, Korea, in the Asian continental outflow. NPF events were identified on approximately 27.1% of total observation days, and 123 and 381 days of them were classified as strong and weak NPF days, respectively, based on the first-mode loading vectors and corresponding principal components in the cyclostationary empirical orthogonal function analysis. NPF events occurred less than 15% of the total observation days for the South China (SC), Pacific Ocean, and stagnation (ST) air mass sectors, whereas 33.1% (243 days out of total 734 observation days) were identified as NPF events for the air mass traveling from the North China (NC) sector. Consistently, the lowest mode diameter (50.5 nm), highest formation rate, and the lowest growth rate were found in the NC air mass sector. The highest mode diameter (89.8 nm) was observed at the SC air mass sector because of the coagulation of newly formed aerosols under relatively humid conditions during the transport in summer. The condensation sink for SC (10.70 ± 2.35 × 10−3 s−1) and ST (10.80 ± 4.24 × 10−3 s−1) was approximately 25% higher than that for the NC air mass sector.

Summertime variability of the western North Pacific subtropical high and its synoptic influences on the East Asian weather

Variation of the western North Pacific subtropical high (WNPSH) is an important meteorological factor for determining summertime rainfall and temperature over East Asia. Here, three major modes of summertime WNPSH variability are identified and corresponding environmental changes are investigated using cyclostationary empirical orthogonal function analysis. The leading mode exhibits a clear reinforcement of WNPSH associated with global warming. The second and third modes are characterized by intra-seasonal variation of the WNPSH intensity related to sea surface temperature variability in the central and eastern equatorial Pacific. Although WNPSH variability is regarded as a local manifestation, it reflects much wider changes in the entire North Pacific. The three modes exert seasonally and geographically distinct impacts on the East Asian weather by setting anomalous atmospheric circulation and altering the direction of moisture and heat transport. As such, the leading WNPSH modes are an important indicator of summertime weathers in countries neighboring the western North Pacific. This study also shows that extreme weather events are likely to increase as global warming intensifies.

A preliminary study on an upper ocean heat and salt content of the western Pacific warm pool region

On the basis of Argo profile data of the temperature and salinity from January 2001 to July 2014, the spatial distributions of an upper ocean heat content (OHC) and ocean salt content (OSC) of the western Pacific warm pool (WPWP) region and their seasonal and interannual variations are studied by a cyclostationary empirical orthogonal function (CSEOF) decomposition, a maximum entropy spectral analysis, and a correlation analysis. Probable reasons for variations are discussed. The results show the following. (1) The OHC variations in the subsurface layer of the WPWP are much greater than those in the surface layer. On the contrary, the OSC variations are mainly in the surface layer, while the subsurface layer varies little. (2) Compared with the OSC, the OHC of the WPWP region is more affected by El Nino-Southern Oscillation (ENSO) events. The CSEOF analysis shows that the OHC pattern in mode 1 has strong interannual oscillation, with eastern and western parts opposite in phase. The distribution of the OSC has a positive-negative-positive tripole pattern. Time series analysis shows that the OHC has three phase adjustments with the occurrence of ENSO events after 2007, while the OSC only had one such adjustment during the same period. Further analysis indicates that the OHC variations are mainly caused by ENSO events, local winds, and zonal currents, whereas the OSC variations are caused by much more complex reasons. Two of these, the zonal current and a freshwater flux, have a positive feedback on the OSC change in the WPWP region.

Intercomparison of precipitation datasets for summer precipitation characteristics over East Asia

Precipitation data in the Global Precipitation Climatology Project (GPCP) and in four reanalysis datasets, ERA-Interim, MERRA, NCEP/NCAR, and JRA, are compared against the CPC Merged Precipitation (CMAP) in the cyclostationary empirical orthogonal function (CSEOF) space to evaluate these datasets in representing the summer precipitation characteristics over East Asia. CSEOF analysis is applied to each dataset, and regression analysis is performed in the CSEOF space with the CMAP data as the target. The regression analysis establishes one-to-one correspondence between the CSEOF loading vectors of the target variable and those of the predictors, i.e., GPCP and the four reanalysis datasets. The loading vectors of the GPCP data coincide almost exactly with those of the CMAP data, i.e., the two observation-based precipitation datasets represent practically identical summer precipitation characteristics over East Asia. The reanalysis datasets also reproduce the first five CSEOF modes reasonably; however, performance of NCEP/NCAR is notably lower than others. The re-constructed precipitation using the first five regressed CSEOF modes of the reanalysis datasets are well correlated with that of the CMAP data with reasonably large correlation coefficients, suggesting that these reanalysis precipitation products reliably simulate the major summer precipitation characteristics in East Asia. All of the four reanalysis products commonly show noticeable errors in representing the summer rainfall over the mid-latitude ocean to the south of Japan, the tropical western Pacific, tropical/subtropical regions including the Indochina Peninsula, India, the Maritime Continent, and regions of complex terrain especially those characterized by strong orographic slopes around the Tibetan Plateau. The errors over the regions of complex orography and coastal lines may be partially due to the inability of reanalysis models in simulating the effects of complex terrain and the lack of observations in these sparsely populated regions.

Physical mechanism of spring and early summer drought over North America associated with the boreal warming

Drought during the early vegetation growing season (spring through early summer) is a severe natural hazard in the large cropland over North America. Given the recent increasing severity of climate change manifested as surface warming, there has been a growing interest in how warming affects drought and the prospect of drought. Here we show the impact of boreal warming on the spring and early summer drought over North America using Cyclostationary Empirical Orthogonal Function analysis. Northern Hemispheric warming, the leading mode of the surface air temperature variability, has led to a decrease in precipitation, evaporation and moisture transport over the central plain of North America. From a quantitative assessment of atmospheric water budget, precipitation has decreased more than evaporation and moisture transport, resulting in increased (decreased) moisture in the lower troposphere (land surface). Despite the increased moisture content, relative humidity has decreased due to the increased saturation specific humidity arising from the lower-tropospheric warming. The anomaly patterns of the soil moisture and Palmer Drought Severity Index resemble that of the anomalous relative humidity. Results of the present study suggest a credible insight that drought in the main cropland will intensify if the anthropogenic warming continues, exacerbating vulnerability of drought.

A new approach to the space\u2013time analysis of big data: application to subway traffic data in Seoul

A prevalent type of big data is in the form of space–time measurements. Cyclostationary empirical orthogonal function (CSEOF) analysis is introduced as an efficient and valuable technique to interpret space–time structure of variability in a big dataset. CSEOF analysis is demonstrated to be a powerful tool in understanding the space–time structure of variability, when data exhibits periodic statistics in time. As an example, CSEOF analysis is applied to the hourly passenger traffic on Subway Line #2 of Seoul, South Korea during the period of 2010–2017. The first mode represents the weekly cycle of subway passengers and captures the majority (~ 97%) of the total variability. The corresponding loading vector exhibits a typical weekly pattern of subway passengers as a function of time and the locations of subway stations. The associated principal component time series shows that there are two occasions of significant reduction in the amplitude of the weekly activity in each year; these reductions are associated with two major holidays—lunar New Year and Fall Festival (called Chuseok in Korea). The second and third modes represent daily contrasts in a week and are associated with taking extra days off before or after holidays. The fourth mode exhibits an interesting upward trend, which represents a general decrease in the number of subway passengers during weekdays except for Wednesday and an increase over the weekends.

Future trend in seasonal lengths and extreme temperature distributions over South Korea

CSEOF analysis is conducted on the daily mean, maximum, and minimum temperatures measured at 60 Korea Meteorological Administration stations in the period of 1979-2014. Each PC time series is detrended and fitted to an autoregressive (AR) model. The resulting AR models are used to generate 100 sets of synthetic PC time series for the period of 1979-2064, and the linear trends are added back to the resulting PC time series. Then, 100 sets of synthetic daily temperatures are produced by using the synthetic PC time series together with the The cyclostationary EOF (CSEOF) loading vectors. The statistics of the synthetic daily temperatures are similar to those of the original data during the observational period (1979-2064). Based on the synthetic datasets, future statistics including distribution of extreme temperatures and the length of four seasons have been analyzed. Average daily temperature in spring is expected to decrease by a small amount, whereas average temperatures in summer, fall and winter are expected to increase. Standard deviation of daily temperatures is expected to increase in all four seasons. The Generalized Extreme Value and Generalized Pareto distributions of extreme temperatures indicate that both warm and cold extremes are likely to increase in summer, while only warm extremes are predicted to increase significantly in winter. Thus, heat waves will increase and cold waves will decrease in number in future. Spring and fall will be shorter, whereas summer and winter will be longer. A statistical prediction carried out in the present study may serve as a baseline solution for numerical predictions using complex models.

Mechanism of seasonal Arctic sea ice evolution and Arctic amplification

Abstract. Sea ice loss is proposed as a primary reason for the Arctic amplification, although the physical mechanism of the Arctic amplification and its connection with sea ice melting is still in debate. In the present study, monthly ERA-Interim reanalysis data are analyzed via cyclostationary empirical orthogonal function analysis to understand the seasonal mechanism of sea ice loss in the Arctic Ocean and the Arctic amplification. While sea ice loss is widespread over much of the perimeter of the Arctic Ocean in summer, sea ice remains thin in winter only in the Barents–Kara seas. Excessive turbulent heat flux through the sea surface exposed to air due to sea ice reduction warms the atmospheric column. Warmer air increases the downward longwave radiation and subsequently surface air temperature, which facilitates sea surface remains to be free of ice. This positive feedback mechanism is not clearly observed in the Laptev, East Siberian, Chukchi, and Beaufort seas, since sea ice refreezes in late fall (November) before excessive turbulent heat flux is available for warming the atmospheric column in winter. A detailed seasonal heat budget is presented in order to understand specific differences between the Barents–Kara seas and Laptev, East Siberian, Chukchi, and Beaufort seas.

Influence of ENSO on the variation of annual sea level cycle in the South China Sea

The variability of the annual sea level cycle in the South China Sea (SCS) has been explored based on tide gauge data, satellite altimetry data and Ocean General Circulation Model (OGCM) outputs. This variability has the potential to increase or decrease the flooding risk in the SCS coastal regions. The cyclostationary empirical orthogonal function (CSEOF) decompositions of the three data sets demonstrate consistent annual sea level fluctuations which are related to ENSO (El Niño-Southern Oscillation) variability. In particular, a significant reduction in the strength of the annual cycle is observed during the 1997–1998 El Niño event. The spatial variability of the annual cycle in the region as computed from the satellite altimetry and OGCM show excellent agreement, while the CSEOF analysis also captures the temporal and spatial distribution characteristics of the ENSO signal. As a secondary result, a drop in the strength of the annual sea level cycle is observed from tide gauge data resulting from the Sumatra-Andaman Earthquake event in 2004.

ENSO Teleconnection to the Isopycnal Depth Fluctuations of the East/Japan Sea Intermediate Water in the Ulleung Basin during 1968–2002

AbstractThis study first detects the decadal variability in the depths of the East/Japan Sea (EJS) Intermediate Water (ESIW) using in situ observations and relates it to strong El Niño–Southern Oscillations (ENSOs). Using multitaper cross-spectrum and cyclostationary empirical orthogonal function analysis, this study found significant coherences at the 99% confidence level and opposite phases between Niño-3 and the ESIW isopycnal depths in the Ulleung Basin (UB) at a period of 14.1 yr during 1968–2002. This suggests a teleconnection between strong ENSOs and the ESIW, the cause of which is explored. When a strong El Niño (EN) develops, the Pacific–North American pattern is intensified by the EN-related Rossby wave interfering constructively with the climatological stationary wave. The amplified wave propagates upward into the stratosphere and breaks, weakening the polar vortex. The EN-related geopotential height increases over the pole with poleward converging air and decreases over the Yakutsk Basin (YB), indicating a negative northern annular mode with the south-to-north gradient balancing the easterly anomaly that responds to vortex weakening. The converged air at the pole warms adiabatically and raises the height as it sinks. This height distribution, including the east-to-west gradient balancing the southward flow, induces the polar vortex split into two with the colder one in the YB where the EJS is closer than from the pole. The EN-related northwesterly wind directing toward the EJS is also strong. Thus, the coldest air with negative wind stress curls reaches the EJS quickly and forms more ESIW, which converges into the UB, causing the observed decadal isopycnal fluctuations.

Physical characteristics of Eurasian winter temperature variability

Despite the on-going global warming, recent winters in Eurasian mid-latitudes were much colder than average. In an attempt to better understand the physical characteristics for cold Eurasian winters, major sources of variability in surface air temperature (SAT) are investigated based on cyclostationary EOF analysis. The two leading modes of SAT variability represent the effect of Arctic amplification (AA) and the Arctic oscillation (AO), respectively. These two modes are distinct in terms of the physical characteristics, including surface energy fluxes and tropospheric circulations, and result in significantly different winter SAT patterns over the Eurasian continent. The AA-related SAT anomalies are dipolar with warm Arctic, centered at the Barents–Kara Seas, and cold East Asia. In contrast, the negative AO-related SAT anomalies are characterized by widespread cold anomalies in Northern Eurasia. Relative importance of the AA and the negative AO contributions to cold Eurasian winters is sensitive to the region of interest.

The effect of the El Niño‐Southern Oscillation on U.S. regional and coastal sea level

AbstractAlthough much of the focus on future sea level rise concerns the long‐term trend associated with anthropogenic warming, on shorter time scales, internal climate variability can contribute significantly to regional sea level. Such sea level variability should be taken into consideration when planning efforts to mitigate the effects of future sea level change. In this study, we quantify the contribution to regional sea level of the El Niño‐Southern Oscillation (ENSO). Through cyclostationary empirical orthogonal function analysis (CSEOF) of the long reconstructed sea level data set and of a set of U.S. tide gauges, two global modes dominated by Pacific Ocean variability are identified and related to ENSO and, by extension, the Pacific Decadal Oscillation. By estimating the combined contribution of these two modes to regional sea level, we find that ENSO can contribute significantly on short time scales, with contributions of up to 20 cm along the west coast of the U.S. The CSEOF decomposition of the long tide gauge records around the U.S. highlights the influence of ENSO on the U.S. east coast. Tandem analyses of both the reconstructed and tide gauge records also examine the utility of the sea level reconstructions for near‐coast studies.

Decadal changes in the Southern Hemisphere sea surface temperature in association with El Niño–Southern Oscillation and Southern Annular Mode

The spatial structure of El Nino–Southern Oscillation (ENSO) and the Southern Annular Mode (SAM), which are the two most important climate modes affecting sea surface temperature (SST) variability in the Southern Hemisphere (SH), appear to have changed since 1999. The characteristic features of the ENSO- and SAM-related atmospheric and oceanic variability in the SH are compared between two sub-periods (1979–1998 and 1999–2012) using cyclostationary empirical orthogonal function analysis. During the earlier period of 1979–1998, the ENSO is characterized by conventional eastern Pacific type, in which the signals in the SH constitute the Pacific South America teleconnection pattern. In contrast, due to a shift of the active center of ENSO to the central Pacific in the later period (1999–2012), atmospheric circulation and SST variability over the SH significantly vary. Moreover, the SAM-related SST variability also shows remarkable differences before and after 1998–1999. This difference is primarily attributed to differences in the non-annular spatial component of the SAM between the two periods. Due to the changes in the spatial structure of the SAM, as well as those of the ENSO, SST variability in the SH displays a marked change between the two periods. Detailed descriptions of the decadal changes of the SH SST in terms of interaction in the oceanic and atmospheric variability are presented along with the possible implications of this change.

Climate change projection in the Northwest Pacific marginal seas through dynamic downscaling

AbstractThis study presents future climate change projections in the Northwest Pacific (NWP) marginal seas using dynamic downscaling from global climate models (GCMs). A regional climate model (RCM) for the Northwest Pacific Ocean was setup and integrated over the period from 2001 to 2100. The model used forcing fields from three different GCM simulations to downscale the effect of global climate change. MIROC, ECHAM, and HADCM were selected to provide climate change signals for the RCM. These signals were calculated from the GCMs using Cyclostationary Empirical Orthogonal Function analysis and added to the present lateral open boundary and the surface forcing. The RCM was validated by comparing hindcast result with the observation. It was able to project detailed regional climate change processes that GCMs were not able to resolve. A relatively large increases of water temperature were found in the marginal seas. However, only a marginal change was found along the Kuroshio path. Heat supply to the atmosphere decreases in most study areas due to a slower warming of the sea surface compared to the atmosphere. The RCM projection suggests that the temperature of the Yellow Sea Bottom Cold Water will gradually increase by 2100. Volume transports through major straits except the Taiwan Strait in the marginal seas are projected to increase slightly in future. Increased northeasterly wind stress in the East China Sea may also result in the transport change.