Abrupt Changes In Time Series Research Articles

Detecting Regime Shifts in the Portuguese Continental Shelf Ecosystem Within the Last Three Decades

Marine ecosystems are affected by diverse pressures and consequently may undergo significant changes that can be interpreted as regime shifts. In this study we used integrated trend analysis (ITA) that combines multivariate statistics and methodologies to identify abrupt changes in time-series, in order to test a hypothesis about the occurrence of regime shifts in the Portuguese continental shelf ecosystem (PCSE). We used two types of data describing ecosystem drivers (fishing mortality and environmental/climatic indices) and ecosystem state (observed and modelled biomass and ecosystem indices). Modelled biomass and ecosystem indices were outputs of Ecopath with Ecosim temporal model parametrised for PSCE between 1986 and 2017. The analyses indicated that the regime shifts in the PCSE have occurred during three periods in the last three decades: “early regime” until the mid-1990s, followed by “transition regime” in-between and “late regime” since the mid-2010s. The detected regime shifts are characterised by changes in the pelagic community that became more dominant when compared to the demersal community and shifted from sardine, the main fishing resource, abundant in the “early regime”, to other less valuable pelagic fishes such as chub mackerel that dominated the “late regime”. The “early regime” was characterised by high catch, a larger proportion of demersal species, and higher diversity while, the “late regime” was represented by lower catch, an increase in higher trophic level (TL) predatory fish and lower diversity. Moreover, the “late regime” showed lower resilience and reduced maturity when compared to the “early regime”. Changes described in the ecosystem were probably related to (1) the shift in the north Atlantic environmental conditions that affected small pelagic fish (SPF) and lower TLs groups, (2) reduction in fishing pressure, and (3) internal triggers, related to the indirect trophic interactions that might have benefited higher TL fish and impacted the pelagic community. In the context of PCSE management, this study highlighted a need to consider the possibility of regime shifts in the management process. For example, regime specific harvest rates and environmental reference points should be considered when an indication of abrupt change in the ecosystem exists.

Change-point detection: application of Cusum method to real life data

Several procedures have been developed for the detection of abrupt changes in time series. Among these procedures, it can be mentioned the Cumulative Sum (Cusum) type method. It is in such a perspective that Katchekpele et al. (2017) proposed a method using a Cusum type test to detect a change-point in the unconditional variance of the generalised autoregressive conditional heteroskedasticity(GARCH) models. The aim of this paper is to present an application of their technique. After briefly recalling how the test statistic was constructed, the change-point detection algorithm is given and it is shown how it is applied to some real life data.

Fuzzy Estimations for Detecting Abrupt Changes: Cases on Tourism Series

This article deals with problems of detecting abrupt changes in time series ba Change Point Model (CPM) framework. We propose a fuzzification in a Fuzzy Time Series (FTS) model to eliminate a trend in a contaminated dependent series. The independent residuals are then inputed on the CPM method. In simulating an abrupt change, an ARIMA(1,1,1) and variance of the model are considered. The abrupt change is modelled as an AO (Additive Outlier) type of outliers. The minimum weight or breaksize of the abrupt change is defined based on the ARIMA variance formulated in this article. The percentage of uncorrelated residuals obtained by the FTS model and the percentage of correct detection of the proposed procedure are shown by simulation. The proposed detecting algorithm is implemented to detect abrupt changes in monthly tourism series in literature, i.e., in Taiwan and in Bali. The first series shows a slowly increasing trend with one abrupt change while the second series exhibits not only a slowly increasing trend but also a strong seasonal pattern with two abrupt changes. For comparison, we detect the changes in the empirical examples on an existing automatic detection procedure using tso package in R. For the first example, the results show that both detecting procedures give exactly a similar location of one change point where the package recognises it as an AO type of outliers. The abrupt change is related to the period of SARS outbreak in Taiwan. On the second example, the proposed procedure locates 4 change points which form two locations of changes, i.e., the first two change points are within 2 time points so do the last two change points. The locations are closed to times of Bali Bombing events. Meanwhile, the automatic procedure recognizes only one AO outlier on the series.

Drought in the Upper Hron Region (Slovakia) between the Years 1984–2014

Climate change causes an increase in the frequency and severity of weather extremes. One of the most relevant severe and damaging phenomena in Europe is drought. However, a difference in the spatial frequency of the occurrence and drought trends is evident between southern and northern Europe. Central Europe and particularly the West Carpathian region form a transitional zone, and drought patterns are complicated because of the geomorphologically complicated landscape. Since almost half of the Slovak state territory is represented by such natural landscape, it is necessary to investigate regional drought specifics. Therefore, we decided to analyze drought occurrence and trends using the SPI (Standardised Precipitation Index) and the SPEI (Standardised Precipitation Evapotranspiration Index) at available climatological stations of the Slovak Hydrometeorological Institute (SHMI) in the upper Hron region within the 1984–2014 period. We found that (1) drought incidence decreased with increasing altitude, (2) increasing air temperature increased the difference in drought trends between lowlands and mountains during the studied period, and (3) abrupt changes in time series of drought indices, that could indicate some signals of changing atmospheric circulation patterns, were not revealed. Finally, we constructed a simplified map of drought risk as an explanation resource for local decision-makers.

ANÁLISIS DE CAMBIOS DE TENDENCIAS DE INDICADORES DE SALUD OCUPACIONAL EN ESPAÑA, EN EL PERIODO 1995 A 2017

Occupational accidents pose considerable burdens to people, businesses and society in general. The analysis of trends in the evolution of time series related to occupational health indicators is an important tool to help determining the extent to which economic or structural changes in the sectors of industry, agriculture, construction and services have influenced the behavior of these indicators. Trend analysis is defined as a process of estimating gradual changes in a series of observations over time, while breakpoint analysis allows to detect abrupt changes in time series. In the literature, different nonparametric and parametric techniques are used to estimate breakpoints and trends. The presented study focuses on the analysis of occupational accidents in Spain in the period 1995-2017. The analysis covers a period of exceptional instability in the Spanish labor market, including a period of economic growth and deep recession. These economic changes brought changes in the composition of the labor market. In the paper, the main indicators used in the statistics of work accidents (Fatal Accident Rate, Incidence Index, Gravity Index, etc.) are analyzed in order to detect trends and changes in them, emphasizing its relationship with risk metrics. The results obtained in the case of application show that the economic and structural changes have affected in a different way the trends of the indicators analyzed. Key Words: Work accidents; Risks evaluation; risk factor's; prevention of occupational hazards; occupational health; labor statistics

Impact of land degradation from mining activities on the sediment fluxes in two large rivers of French Guiana

AbstractWe analyzed two contrasting catchments located among the world's largest unspoiled tropical rainforests impacted by mining in the northeastern coastlands of South America. We used the following: (a) mining, agricultural, and urbanized areas to compare the land use evolution with suspended sediments and sediment yields; (b) field monthly river suspended sediments in the two catchments (2004–2015: n = 154); (c) MODIS remote sensing water color technique in the Maroni basin to complete (n = 387) and extend field suspended sediment sampling from 2000 to 2015; (d) hydroclimatic statistical analysis conditions and sediment concentrations to identify the long‐term trends, the abrupt changes in time series and to analyze if the environmental and anthropogenic factors control sediment yield regional variations. No significant long‐term changes were observed in precipitation or water discharge with the Mann–Kendall test. However, the mean suspended sediment concentration has increased significantly (239%) in the Maroni River with a breakpoint in 2009 and decreased (33%) in the Oyapock River (breakpoint in 2008). These differences are explained by the larger percentage of deforestation because of mining activities in the Maroni (0.37%) than in the Oyapock (0.06%) catchment. In the Maroni River, the increasing sediment yield trend (2000–2015) coincide significantly (r2 = 0.97; p < 0.0001) with the increase of 400% of mining areas, whereas no significant relationship with the runoff was found. In the Oyapock River, the runoff explains the sediment yield decreasing trend (r2 = 0.82; p < 0.0001) and no relationship with the land use change was found.

Evaluation of the significance of abrupt changes in precipitation and runoff process in China

Abrupt changes are an important manifestation of hydrological variability. How to accurately detect the abrupt changes in hydrological time series and evaluate their significance is an important issue, but methods for dealing with them effectively are lacking. In this study, we propose an approach to evaluate the significance of abrupt changes in time series at five levels: no, weak, moderate, strong, and dramatic. The approach was based on an index of correlation coefficient calculated for the original time series and its abrupt change component. A bigger value of correlation coefficient reflects a higher significance level of abrupt change. Results of Monte-Carlo experiments verified the reliability of the proposed approach, and also indicated the great influence of statistical characteristics of time series on the significance level of abrupt change. The approach was derived from the relationship between correlation coefficient index and abrupt change, and can estimate and grade the significance levels of abrupt changes in hydrological time series.Application of the proposed approach to ten major watersheds in China showed that abrupt changes mainly occurred in five watersheds in northern China, which have arid or semi-arid climate and severe shortages of water resources. Runoff processes in northern China were more sensitive to precipitation change than those in southern China. Although annual precipitation and surface water resources amount (SWRA) exhibited a harmonious relationship in most watersheds, abrupt changes in the latter were more significant. Compared with abrupt changes in annual precipitation, human activities contributed much more to the abrupt changes in the corresponding SWRA, except for the Northwest Inland River watershed.

Marine regime shifts in ocean biogeochemical models: a case study in the Gulf of Alaska

Abstract. Regime shifts have been reported in many marine ecosystems, and are often expressed as an abrupt change occurring in multiple physical and biological components of the system. In the Gulf of Alaska, a regime shift in the late 1970s was observed, indicated by an abrupt increase in sea surface temperature and major shifts in the catch of many fish species. A thorough understanding of the extent and mechanisms leading to such regime shifts is challenged by data paucity in time and space. We investigate the ability of a suite of ocean biogeochemistry models of varying complexity to simulate regime shifts in the Gulf of Alaska by examining the presence of abrupt changes in time series of physical variables (sea surface temperature and mixed-layer depth), nutrients and biological variables (chlorophyll, primary productivity and plankton biomass) using change-point analysis. Our results show that some ocean biogeochemical models are capable of simulating the late 1970s shift, manifested as an abrupt increase in sea surface temperature followed by an abrupt decrease in nutrients and biological productivity. Models from low to intermediate complexity simulate an abrupt transition in the late 1970s (i.e. a significant shift from one year to the next) while the transition is smoother in higher complexity models. Our study demonstrates that ocean biogeochemical models can successfully simulate regime shifts in the Gulf of Alaska region. These models can therefore be considered useful tools to enhance our understanding of how changes in physical conditions are propagated from lower to upper trophic levels.

A Novel Change-Point Detection Approach for Monitoring High-Dimensional Traffics in Distributed Systems

Change-point detection is the problem of finding abrupt changes in time-series. However, the meaningful changes are usually difficult to identify from the original massive traffics, due to high dimension and strong periodicity. In this paper, we propose a novel change-point detection approach, which simultaneously detects change points from all dimensions of the traffics with three steps. We first reduce the dimensions by the classical Principal Component Analysis (PCA), then we apply an extended time-series segmentation method to detect the nontrivial change times, finally we identify the responsible applications for the changes by F-test. We demonstrate through experiments on datasets collected from four distributed systems with 44 applications that the proposed approach can effectively detect the nontrivial change points from the multivariate and periodical traffics. Our approach is more appropriate for mining the nontrivial changes in traffic data comparing with other clustering methods, such as center-based Kmeans and density-based DBSCAN.