Atmospheric Pressure Time Series Research Articles

Brief Report of Possible Anticipation of Earthquakes from Geomagnetic Studies Water Level and Atmospheric Pressure in the South Atlantic

This paper presents a brief summary of studies carried out in the southwestern Atlantic, based on the analysis of disturbances in the earth’s Geomagnetic Field (GMF), Water Level (WL) and sea-level Atmospheric Pressure (AP) associated with seismic events. GMF records from observatories of the International Real-time Magnetic Observatory Network (INTERMAGNET), WL records from tide stations of the Intergovernmental Oceanographic Commission (IOC), and hourly AP records from databases of the Global Modeling and Assimilation Office (GMAO) of the National Aeronautics and Space Administration (NASA) have been analyzed in the area under study. A comparative table is presented between the frequencies of the GMF wavelet transform analysis, filtered WL records and variation in the AP grids at sea level before and after the earthquakes. It is possible to observe different degrees of approximation in each of the variables regarding the occurrence of earthquakes. For example, in the case of the CMF, some high-intensity seismic events show anomalous frequencies some hours before the earthquake. On the other hand, the WL filtered by astronomical tide, account for the seismic event along the seacoast with a correspondence between wave amplitude and distance to the nearest ports. AP grids account for a dipole type configuration in the distribution of the pressure difference around the earthquake. On the other hand, the time series of atmospheric pressure at ocean level show a local minimum in a range of hours around the occurrence of the event.

Predictive Analysis of Machine Learning Schemes in Forecasting Offshore Wind Speed

High variability of wind in the farm areas causes a drastic instability in the energy markets. Therefore, precise forecast of wind speed plays a key role in the optimal prediction of offshore wind power. In this study, we apply two deep learning models, i.e. Long Short-Term Memory (LSTM) and Nonlinear Autoregressive EXogenous input (NARX), for predicting wind speed over long-range of dependencies. We use a four-month-long wind speed/direction, air temperature, and atmospheric pressure time series (all recorded at 10 m height) from a meteorological mast (Vigra station) in the close vicinity of the Havsul-I offshore area near Ålesund, Norway. While both predictive methods could efficiently predict the wind speed, the LSTM with update generally outperforms the NARX. The NARX suffers from vanishing gradient issue and its performance declines by abrupt variability inherited in the input data during training phase. It is observed that this sensitivity will significantly decrease by integrating, for example, the wind direction at low frequencies in the learning process. Generally, the results showed that the predictive models are robust and accurate in short-term and somewhat long-term forecasting of wind.

Analysis of long-term observations of the groundwater level In an aseismic region

The results of the analysis of continuous precise time series of atmospheric pressure and groundwater level fluctuations in a well drilled to a depth of 400 m in the territory of Moscow are presented. The observations are remarkable in terms of their duration of more than 22 years (from February 2, 1993 to April 4, 2015) and by the sampling interval of 10 min. These long observations are suitable for exploring the stationarity of the properties of hydrogeological time series in a seismically quiet region, which is important from the methodological standpoint for interpreting the similar observations in seismically active regions aimed at earthquake prediction. Factor and cluster analysis applied to the sequence of multivariate vectors ofthe statistical properties of groundwater level time series in the successive 10-day windows after adaptive compensation for atmospheric pressure effects distinguish five different statistically significant states of the time series with the transitions between them. An attempt to geophysically interpret the revealed states is made. Two significant periods – 46 and 275 days – are established by spectral analysis of the sequence of the transitions times between the clusters.

Wavelet-Based Clustering of Sea Level Records

The classification of multivariate time series in terms of their corresponding temporal dependence patterns is a common problem in geosciences, particularly for large datasets resulting from environmental monitoring networks. Here a wavelet-based clustering approach is applied to sea level and atmospheric pressure time series at tide gauge locations in the Baltic Sea. The resulting dendrogram discriminates three spatially-coherent groups of stations separating the southernmost tide gauges, reflecting mainly high-frequency variability driven by zonal wind, from the middle-basin stations and the northernmost stations dominated by lower-frequency variability and the response to atmospheric pressure.

An EOF and SVD analysis of interannual variability of GPS coordinates, environmental parameters and space gravity data

Eleven years, 1999–2010, of GPS positions of stations located in Europe and the Mediterranean were analyzed together with time series of atmospheric pressure, terrestrial water storage, and surface mass anomalies provided by the GRACE gravity mission. The Empirical Orthogonal Functions (EOF) technique was applied to single variables data series detrended and de-seasoned to identify the main modes of interannual variability in both time and space. For the GPS coordinates, the analysis has revealed the presence of significant interannual fluctuations which were also identified in the time series of the environmental parameters and of the space gravity data. A prominent feature common to the main modes of the time components of all variables is a slope inversion occurring around 2003–2004. During 2003 most of the European continent was affected by a severe drought. A second type of analysis was performed by applying the Singular Value Decomposition (SVD) approach to a few series of variable pairs to study the coupled space/time variability. This mathematical tool is used to examine possible cause-effect physical relationships between variables pairs. The analysis shows, for example, for the atmospheric pressure and the up coordinate that the first two SVDs account for about 97% of the total covariance. The clear anticorrelation observed between the two SVD1 spatial patterns represents the vertical crustal displacements associated with variations in atmospheric pressure. East–West (EW) and North–South (NS) gradients of the environmental parameters and of the GRACE surface mass anomalies were computed and compared to the EOFs of the east and north GPS coordinates respectively. Significant correlations were found, in particular, between the EW atmospheric pressure gradient and the east EOFs and between the GRACE EW and NS gradients and the east and north EOFs respectively. Significant correlations were also identified between the Multivariate ENSO Index (MEI) and the EOFs of GPS coordinates.

Application of Artificial Neural Network (ANN) to improve forecasting of sea level

To improve on forecasting of tidal water level beyond harmonic analysis requires the incorporation of meteorological variables in the analysis. This suggests the use of Artificial Neural Networks (ANN) as an optimum tool to train and translate the combined influence of meteorological and astronomical forcing to predict sea level variations and reduce the margin of error of close to 50% (from 26% to 12%). To accomplish this, the ANN was trained by using hourly time series of atmospheric pressure, wind, and harmonically derived tides for 1982 as input data and hourly time series of measured tides as output data. The meteorological data were obtained from São Sebastião (SP) and Ponta da Armação (RJ), and the sea level data from Cananéia (SP) and Ilha Fiscal (RJ). Data gaps in the time series were interpolated based on FFT analysis. To forecast water levels, the 1983 meteorological time series was used as input data, and compared the resulting water level outputs to the water level measurements for the same period. The ANN served as a very good forecasting tool for sea level variability. In the case of Cananéia, with several meteorological data gaps, the comparison was less successful as compared to the Ilha Fiscal results, besides this, there is a local influence of the estuary flows, a variable not considered that could answers for the remaining 12% of the correlation. The coefficient of correlation between predicted and measured water level time series at Cananéia was 0.88 and at Ilha Fiscal 0.98. This kind of improvement can be used for port terminals and marinas, for handling incoming and outgoing ships and boats more safely through the navigation channels in the estuaries. It is applicable and useful information for decision makers in management activities in the coastal area.

Statistical pre-processing and analyses of hydro-meteorological time series in a geologic clay site (methodology and first results for Mont Terri’s PP experiment)

This article presents a set of statistical methods for pre-processing (or pre-conditioning) and analyzing multivariate hydro-geologic time series, such as pore pressures and atmospheric pressure (temporal structures). The pre-processing methods are necessary to enhance or complete the signals due to defects in the observed time series (data gaps, spurious values, variable time steps). The statistical analyses aim, in fine, at characterizing the hydraulic behaviour of a clayey formation in the context of deep geologic disposal of radioactive waste. Pore water and atmospheric pressure time series from the Mont Terri rock laboratory (BPP-1 borehole) are taken as an example.

Radon 222 Emanometry: A relevant methodology for water well siting in hard rock aquifers

To improve the understanding of soil gas emanations with the aim of groundwater prospecting in fractured substrates, radon anomalies were mapped in a tropical region (French Guyana) where the structural features of a potential gas source are masked owing to the thick regolith cover. Two anomalies were located. Two probes were then installed, one (Bl) on the border of the north anomaly and the second (B2) at the center of the south anomaly, as well as a meteorological station, and monitoring was carried out for 6 months extending from the dry to the rainy season. Signal processing of radon signal–atmospheric pressure time series allowed us to determine the physical processes causing migration of the gases in the soil and the subsoil. The radon anomalies are very probably associated with the presence, in the hard rock substratum, of a permeable structure that permits the transit of “deep” origin gas to the surface. This transport by advection disturbs the radon diffusion profile from the first meters of soil into the atmosphere. The radon thus behaves as a marker of deeper gases percolating through the regolith cover. Indeed, in such a context with low‐permeability weathered formations, because of its short half‐life (3.8 days) the radon measured in the first decimeters of the soil displays an exclusively superficial origin. Hence the measurement of the radon content in soil, at an identical depth from one sampling site to another, helps locate the zones affected by advective gas transfers which explains why the radon anomalies in soil gases, identified at the scale of a hydrogeological prospecting site located in a hard rock context, usually coincide vertically in hydraulically active fracture zones. This fully justifies the use of this method to site water wells, provided some precautions are observed, particularly pertaining to the spatial and temporal variability of the signal associated with different types of soils and changes in the meteorological parameters and soil moisture contents. The methodology is simple since only robust equipment is required. The estimation of gas transport parameters also helped identify the extent of advection areas. During the dry season they are characterized by low gas velocity. These zones shrink with the early rains, and the advection area is confined vertically up the fault where a channeling effect occurs: Soil gas preferably follows the less constricted pores, and the increased pressure gradient offsets the decrease in permeability.

Gas transport in a confined unsaturated zone during atmospheric pressure cycles

Gas transport induced by temporal pressure fluctuations in the atmosphere can be an important mechanism for transport of atmospheric oxygen within the unsaturated zone. Moreover, the presence of oxygen in the unsaturated zone may be a factor controlling oxidation of sulphide minerals and other redox processes. A field study was carried out in a glacial aquifer with a 10–12 m thick sandy unsaturated zone to explore gas exchange between the atmosphere and the unsaturated zone. The exchange occurs through a “geological window” in a till layer which covers the sandy unsaturated zone. Observed pressure distribution and oxygen concentrations within the unsaturated zone were compared to numerical simulations with SUTRA, a finite element and fluid density dependent groundwater flow model. The simulations were carried out by modeling the gas pressure distribution within the unsaturated zone based on atmospheric pressure time series. The spatial variation in permeability observed from borehole logging was implemented in the model. The analysis demonstrated a good match between the field observations and the numerical simulations. During an atmospheric pressure cycle, atmospheric oxygen migrated more than 10 md−1 horizontally in the capped unsaturated zone. The analysis shows that both the amplitude and the length of the period of pressure variations are important for the transport of oxygen, and it shows that the combined effects determine the extent of a subsurface zone where atmospheric oxygen can reach.

Field-based determination of air diffusivity using soil air and atmospheric pressure time series : S. Rojstaczer & J. P. Tunks, Water Resources Research, 31(12), 1995, pp 3337–3343

Field-based determination of air diffusivity using soil air and atmospheric pressure time series : S. Rojstaczer & J. P. Tunks, Water Resources Research, 31(12), 1995, pp 3337–3343

Field‐based determination of air diffusivity using soil air and atmospheric pressure time series

Air diffusivity in two zones over a 4.6‐m interval of the unsaturated zone was determined through analysis of atmospheric pressure and soil air pressure time series. Regression analysis was used to calculate the ratio of amplitudes (admittance) and phase lag between these series at diurnal and semidiurnal frequencies. The admittance for each of the monitoring zones was close to unity for both frequencies. Phase lags between the two time series were statistically equivalent to zero at the diurnal frequency but were as large as 20° ± 7° at the semidiurnal frequency. The phase lag at diurnal and semidiurnal frequencies for each zone was compared to a diffusional model of soil air pressure response to atmospheric loading. From this comparison a composite air diffusivity was determined for each zone. The absence of large variability in air diffusivity and the trends in phase shift with depth suggest that macropores or fractures propagate through a clay layer at a depth of 0.5–1.2 m. The length of the time series (110 days) allowed for the examination of the magnitude of temporal changes in phase shift and air diffusivity. Temporal and spatial trends in phase lag are smaller than the error bounds in the phase estimates. Air diffusivity determined from the entire time record deviated only slightly from the average diffusivity determined from the analysis of a sequence of partial records.

Meteorological forcing and sea level variability in the Aegean Sea

The effect of atmospheric pressure and surface wind vector on sea level variability in the Aegean Sea is investigated. Daily values of sea level, atmospheric pressure, and surface wind covering the period 1984–1987 from five stations are analyzed. Principal component (PC) analysis is used to extract the spatially coherent signal from the sea level and atmospheric pressure time series, while rotary principal component analysis is used to extract the coherent part of surface wind. The first PC modes are subsequently used in multiple regression analysis in the frequency domain in order to examine the dependence of sea level fluctuations on each of the parameters. The response of sea level to atmospheric pressure is found to be frequency dependent. The response is overisostatic for lower frequencies and becomes underisostatic for higher frequencies approaching isostasy only at the limit of analysis (0.5 cycles/day). The contribution of the first principal wind component is found to be very small.

Aspects of marine environmental variability near Cape Town, 1960–1985

Time-series of atmospheric pressure, winds, sea surface temperature, sea level and the Southern Oscillation Index are analysed in order to document aspects of marine environmental variability of and to provide suitable environmental indices for the nearshore southern Benguela system. The circannual cycle dominated, but a high degree of month-to-month and interannual variability was evident. Two contrasting environmental scenarios are postulated, representing extended summer or winter conditions respectively and probably related to decadal changes in the mean position of the South Atlantic Anticyclone. Principal components analysis pointed to possible teleconnections between the Pacific El Nino and the associated Southern Oscillation, and local environmental conditions.