Academia.eduAcademia.edu

Outline

Title
Abstract
Figures
Introduction
Theory: The Wavelet Analysis
Assessment of Statistical Significance
Real Life Data: Analysis of Ecological Time Series
Discussion
References
All Topics
Environmental Science
Ecology

Wavelet analysis of ecological time series

Profile image of Jon VikJon VikProfile image of Bernard CazellesBernard Cazelles

2008

https://doi.org/10.1007/S00442-008-0993-2
visibility

…

description

18 pages

link

1 file

Abstract

Wavelet analysis is a powerful tool that is already in use throughout science and engineering. The versatility and attractiveness of the wavelet approach lie in its decomposition properties, principally its time-scale localization. It is especially relevant to the analysis of nonstationary systems, i.e., systems with short-lived transient components, like those observed in ecological systems. Here, we review the basic properties of the wavelet approach for time-series analysis from an ecological perspective. Wavelet decomposition offers several advantages that are discussed in this paper and illustrated by appropriate synthetic and ecological examples. Wavelet analysis is notably free from the assumption of stationarity that makes most methods unsuitable for many ecological time series. Wavelet analysis also permits analysis of the relationships between two signals, and it is especially appropriate for following gradual change in forcing by exogenous variables.

Figures (9)
Fig. 1 Time-—frequency resolution of the wavelet approach. Left- hand panel Examples of wavelets and their time-frequency boxes representing the corresponding variance distribution. When the scale (a) decreases, the time resolution improves but the frequency resolution becomes poorer and is shifted towards high frequencies. Conversely, if a increases, the boxes shift toward the region of low frequencies and the height of the boxes decreases (with a better frequency resolution) but they become wider (with a poor time resolution). Right-hand panel In contrast to wavelets, all the boxes of the windowed Fourier transform are obtained by a time or frequency shift of a unique function, which yields the same variance spread over the entire time-frequency plane
Fig. 1 Time-—frequency resolution of the wavelet approach. Left- hand panel Examples of wavelets and their time-frequency boxes representing the corresponding variance distribution. When the scale (a) decreases, the time resolution improves but the frequency resolution becomes poorer and is shifted towards high frequencies. Conversely, if a increases, the boxes shift toward the region of low frequencies and the height of the boxes decreases (with a better frequency resolution) but they become wider (with a poor time resolution). Right-hand panel In contrast to wavelets, all the boxes of the windowed Fourier transform are obtained by a time or frequency shift of a unique function, which yields the same variance spread over the entire time-frequency plane
and the value of #(W,(a, t2)) will be low. Finally in 73, the signal and the Morlet wavelet are in a perfect opposite phase with similar period, and there will be a high negative value for R(W,(a,73)). e Wavelet transforms [or quantities derived from W, (a,t)] are plotted on a two- dimensional (2D) graph. As an example, e shows the value of |W,(a,t)|° for a = a> and for the three time positions t, t2 and t; defined in d. The complete 2D plot is obtained simply by computing wavelet transforms for a given range of a and t values Fig. 2 Wavelet analysis. a-c Form of the Morlet wavelet as a function of the scale parameter a for t = 0: a = a; = 0.5 (a), a= a = 2 (b), a = a; = 4 (c). Graphs display both the real part (solid line) and the imaginary part (dashed line) of the wavelet. d Morlet wavelet with a =a» is superimposed on a given signal at different time positions (1), T2,73). In t,, the matching between the signal and the wavelet is height; this will give a high positive value of the real part of the wavelet transform, 3t(W,(a,t,)). In t2, the matching is weak Figure 2 attempts to visualize the signification of the wavelet transform. In Fig. 2d a mother wavelet of scale a centered at location t is shown superimposed on an arbi- trary time series. In the case of good matching between the signal (x) and the wavelet (@), the integral of the product of the signal with the wavelet of scale a produces a large positive value for the real part of wavelet transform, R(W,(a, T)), at the position t (Fig. 2d). When the matching is low, R(W,(a, t)) takes low values. But an association in the opposite phase results in a high negative value for the real part of wavelet transform (Fig. 2d). By moving the wavelet along the signal (by increasing the t parameter), structures relating to a specific scale a can be identified. This process is repeated over continuous ranges of a and Tt to identify all the coherent structures within the signal The wavelet transform is basically a linear filter whose response function is given by the wavelet function. By means of the inverse transform, the original signal can be recovered by integrating over all scales and locations, a
and the value of #(W,(a, t2)) will be low. Finally in 73, the signal and the Morlet wavelet are in a perfect opposite phase with similar period, and there will be a high negative value for R(W,(a,73)). e Wavelet transforms [or quantities derived from W, (a,t)] are plotted on a two- dimensional (2D) graph. As an example, e shows the value of |W,(a,t)|° for a = a> and for the three time positions t, t2 and t; defined in d. The complete 2D plot is obtained simply by computing wavelet transforms for a given range of a and t values Fig. 2 Wavelet analysis. a-c Form of the Morlet wavelet as a function of the scale parameter a for t = 0: a = a; = 0.5 (a), a= a = 2 (b), a = a; = 4 (c). Graphs display both the real part (solid line) and the imaginary part (dashed line) of the wavelet. d Morlet wavelet with a =a» is superimposed on a given signal at different time positions (1), T2,73). In t,, the matching between the signal and the wavelet is height; this will give a high positive value of the real part of the wavelet transform, 3t(W,(a,t,)). In t2, the matching is weak Figure 2 attempts to visualize the signification of the wavelet transform. In Fig. 2d a mother wavelet of scale a centered at location t is shown superimposed on an arbi- trary time series. In the case of good matching between the signal (x) and the wavelet (@), the integral of the product of the signal with the wavelet of scale a produces a large positive value for the real part of wavelet transform, R(W,(a, T)), at the position t (Fig. 2d). When the matching is low, R(W,(a, t)) takes low values. But an association in the opposite phase results in a high negative value for the real part of wavelet transform (Fig. 2d). By moving the wavelet along the signal (by increasing the t parameter), structures relating to a specific scale a can be identified. This process is repeated over continuous ranges of a and Tt to identify all the coherent structures within the signal The wavelet transform is basically a linear filter whose response function is given by the wavelet function. By means of the inverse transform, the original signal can be recovered by integrating over all scales and locations, a
the signals displayed in c. e Wavelet power spectrum of the signal displayed in a. The colors code for power values from dark blue (low values), to dark red (high values). Dotted white lines materialize the maxima of the undulations of the wavelet power spectrum and the black line indicates the cone of influence that delimits the region not influenced by edge effects. f Average wavelet power spectrum (S;(f)) of the signal in a. g, h As in e, f but for the signal displayed in ¢ Fig. 3 Wavelet analysis of sinusoidal signals with two periodic components. a A signal with a periodic component present during the whole series [ = 5, p, = 0.25 unit of time (u.t.) and pz = 1 ut.]. b Fourier spectrum of the signals displayed in a. The periodograms have been smoothed with a Parzen window (Chatfield 1989). ec A series with the same periodic components as those of a but these two periodic components are localized on the first and on the second half- part of the signal, the period shift is at t, = 5. d Fourier spectrum of
the signals displayed in c. e Wavelet power spectrum of the signal displayed in a. The colors code for power values from dark blue (low values), to dark red (high values). Dotted white lines materialize the maxima of the undulations of the wavelet power spectrum and the black line indicates the cone of influence that delimits the region not influenced by edge effects. f Average wavelet power spectrum (S;(f)) of the signal in a. g, h As in e, f but for the signal displayed in ¢ Fig. 3 Wavelet analysis of sinusoidal signals with two periodic components. a A signal with a periodic component present during the whole series [ = 5, p, = 0.25 unit of time (u.t.) and pz = 1 ut.]. b Fourier spectrum of the signals displayed in a. The periodograms have been smoothed with a Parzen window (Chatfield 1989). ec A series with the same periodic components as those of a but these two periodic components are localized on the first and on the second half- part of the signal, the period shift is at t, = 5. d Fourier spectrum of
delimits the region not influenced by edge effects. d Average wavelet power spectrum (S,(f)) of the signal. On the graphs ec, d dashed lines show the «= 5% significance levels computed based on 1,000 Markov bootstrapped series. ¢ P-values associated with the values within the area delineated by the dashed line are less than 5%. e Frequency-average wavelet power for the signal displayed in a, over the 0.1-5 u.t. band (black line), around the 1 u.t. band (dashed line) and around the 0.25 u.t. band (dotted—dashed line) Fig. 4 Synthetic transient signal studied with wavelet approach. a A transient time series with modification of both the amplitude and the period described by Eq. 10 with uw = 5, t,) =3 and t,. =5.25. b Fourier spectrum of the signal displayed in a. The periodogram has been smoothed with a Parzen window (Chatfield 1989). ¢ Wavelet power spectrum of the signal. The colors code for power values from dark blue (low values) to dark red (high values). Dotted white lines materialize the maxima of the undulations of the wavelet power spectrum and the black line indicates the cone of influence that
delimits the region not influenced by edge effects. d Average wavelet power spectrum (S,(f)) of the signal. On the graphs ec, d dashed lines show the «= 5% significance levels computed based on 1,000 Markov bootstrapped series. ¢ P-values associated with the values within the area delineated by the dashed line are less than 5%. e Frequency-average wavelet power for the signal displayed in a, over the 0.1-5 u.t. band (black line), around the 1 u.t. band (dashed line) and around the 0.25 u.t. band (dotted—dashed line) Fig. 4 Synthetic transient signal studied with wavelet approach. a A transient time series with modification of both the amplitude and the period described by Eq. 10 with uw = 5, t,) =3 and t,. =5.25. b Fourier spectrum of the signal displayed in a. The periodogram has been smoothed with a Parzen window (Chatfield 1989). ¢ Wavelet power spectrum of the signal. The colors code for power values from dark blue (low values) to dark red (high values). Dotted white lines materialize the maxima of the undulations of the wavelet power spectrum and the black line indicates the cone of influence that
On the wavelet power spectrum graphs, the dashed lines show the a= 5% significance levels computed based on 1,000 Markov bootstrapped series. On these 2-D graphs (b—h), the colors code for power values from dark blue (low values), to dark red (high values), the dotted white lines materialize the maxima of the undulations of the wavelet power spectrum, the black lines indicate the cone of influence that delimits the region not influenced by edge effects and the P-values associated with the values within the area delineated by the dashed line are less than 5%. The parameters used for the AR(2) models are: a; = 1.5, b; = 1.2, c) = —0.9, 07 = 0.15, ao = 4, b> = 0.1, Cy = —1, 03 = 0.05 and t, = 60 Fig. 5 Influence of the number of data point and the observational noise on wavelet analysis. a The time series is generated with the lag- 2 autoregressive [AR(2)] models (Eq. 11). Triangles indicate the odd data point, stars even data points and the dashed line noisy even data points. b Wavelet power spectrum of the full data set (n = 100 and the sampling time step Ar = 1). d Wavelet power spectrum of the odd data points (n = 50 and At = 2). f Wavelet power spectrum of the even data points (n = 50 and Ar = 2). h Wavelet power spectrum of the even data points (x = 50 and At = 2) when observational noise is added (¢; = 0.80 o,). c-i Average wavelet power spectrum corre- sponding to the wavelet power spectrum plotted in b-h, respectively.
On the wavelet power spectrum graphs, the dashed lines show the a= 5% significance levels computed based on 1,000 Markov bootstrapped series. On these 2-D graphs (b—h), the colors code for power values from dark blue (low values), to dark red (high values), the dotted white lines materialize the maxima of the undulations of the wavelet power spectrum, the black lines indicate the cone of influence that delimits the region not influenced by edge effects and the P-values associated with the values within the area delineated by the dashed line are less than 5%. The parameters used for the AR(2) models are: a; = 1.5, b; = 1.2, c) = —0.9, 07 = 0.15, ao = 4, b> = 0.1, Cy = —1, 03 = 0.05 and t, = 60 Fig. 5 Influence of the number of data point and the observational noise on wavelet analysis. a The time series is generated with the lag- 2 autoregressive [AR(2)] models (Eq. 11). Triangles indicate the odd data point, stars even data points and the dashed line noisy even data points. b Wavelet power spectrum of the full data set (n = 100 and the sampling time step Ar = 1). d Wavelet power spectrum of the odd data points (n = 50 and At = 2). f Wavelet power spectrum of the even data points (n = 50 and Ar = 2). h Wavelet power spectrum of the even data points (x = 50 and At = 2) when observational noise is added (¢; = 0.80 o,). c-i Average wavelet power spectrum corre- sponding to the wavelet power spectrum plotted in b-h, respectively.
Fig. 6 Quantification of the association between two synthetic signals with the wavelet approach. a Time series, x(t) and y(t), generated with two different AR(2) models with correlated noise components (n = 100). These series have been normalized for the analyses. b Wavelet cross-spectrum. ¢ Average wavelet cross-spectrum. d Coherence between the two time series. Colors code from dark blue (low values) to dark red (high values) and the black lines indicate the cone of influence that delimits the region not influenced by edge effects. On the graphs b-d dashed lines show the « = 5% significance levels computed based on 1,000 Markov bootstrapped series. P-values associated with the values within the region delineated by the dashed line are less than 5%. e Phases of the two time series computed in the 7-9 periodic band; dotted line represents the phase difference. f Distribution of the phase difference of the two time series. The parameters used for the AR model are: a, = 1.5, Db, = 1.2, c=-0.9, 0? =0.025, a =1,by=1.35, cy = —0.75, a = 0.025 and pe, <, = 0.50
Fig. 6 Quantification of the association between two synthetic signals with the wavelet approach. a Time series, x(t) and y(t), generated with two different AR(2) models with correlated noise components (n = 100). These series have been normalized for the analyses. b Wavelet cross-spectrum. ¢ Average wavelet cross-spectrum. d Coherence between the two time series. Colors code from dark blue (low values) to dark red (high values) and the black lines indicate the cone of influence that delimits the region not influenced by edge effects. On the graphs b-d dashed lines show the « = 5% significance levels computed based on 1,000 Markov bootstrapped series. P-values associated with the values within the region delineated by the dashed line are less than 5%. e Phases of the two time series computed in the 7-9 periodic band; dotted line represents the phase difference. f Distribution of the phase difference of the two time series. The parameters used for the AR model are: a, = 1.5, Db, = 1.2, c=-0.9, 0? =0.025, a =1,by=1.35, cy = —0.75, a = 0.025 and pe, <, = 0.50
(Elton and Nicholson 1942) and the porcupine time series comes from a dendrochronological reconstruction of porcupine abundance (Klvana et al. 2004). As shown in Fig. 8b—d, the association between sunspot number and lynx abundance appears very weak and significant in the 10-12 periodic band only for the time periods 1821-1840 and 1915-1934. In addition, the phase analysis shows that for the time period 1821-1840 the two series are in phase while for 1915-1934 the two series are out of phase with a delay of 32/2 or 3/4 of the quasi-period (Fig. 8c). Conversely, we obtained a strong association between porcupine abundance and the sunspot number in the 10-12 periodic band for virtually the full length of the time series (Fig. 8e-g). This relation is weaker during 1940-1980 when the oscillations are coherent around the 22-year periodic band (Fig. 8e). The coherence is also very significant around the 22-year mode as stressed in Klvana et al. (2004). Phase analysis reveals (Fig. 8f) that the two series are out of phase with an approximate delay of half a quasi-cycle (2). The sign of the associ- ation is then negative. Furthermore, this relationship between the phases of the two series is highly significant (Fig. 8g).
(Elton and Nicholson 1942) and the porcupine time series comes from a dendrochronological reconstruction of porcupine abundance (Klvana et al. 2004). As shown in Fig. 8b—d, the association between sunspot number and lynx abundance appears very weak and significant in the 10-12 periodic band only for the time periods 1821-1840 and 1915-1934. In addition, the phase analysis shows that for the time period 1821-1840 the two series are in phase while for 1915-1934 the two series are out of phase with a delay of 32/2 or 3/4 of the quasi-period (Fig. 8c). Conversely, we obtained a strong association between porcupine abundance and the sunspot number in the 10-12 periodic band for virtually the full length of the time series (Fig. 8e-g). This relation is weaker during 1940-1980 when the oscillations are coherent around the 22-year periodic band (Fig. 8e). The coherence is also very significant around the 22-year mode as stressed in Klvana et al. (2004). Phase analysis reveals (Fig. 8f) that the two series are out of phase with an approximate delay of half a quasi-cycle (2). The sign of the associ- ation is then negative. Furthermore, this relationship between the phases of the two series is highly significant (Fig. 8g).
by edge effects and dashed lines show the a = 5% and « = 10% significance levels computed based on 1,000 Markov bootstrapped series. c, f Phases of the population abundance and the sunspot numbers computed in the 10-12 year periodic band. Dotted line represents the phase difference and the colors are as in a. d, g Distribution of the phase difference of the two considered time series. d Normalized entropy of this phase difference distribution (see Cazelles and Stone 2003) is Q = 0.19 and it is not significant. g Normalized entropy of this distribution is Q =0.47 and it is significant with a P-value computed with 1,000 Markov bootstrapped series less than 0.1% Fig. 8 Comparison between the association between the sunspot numbers and the lynx population and between the sunspot numbers and the porcupine population. a Normalized time series: the sunspot numbers (red line), the lynx abundance (blue line) and the porcupine abundance (black line). Note that the lynx and porcupine time series have a different starting date and different length, which explains the blank spaces in the following graphs. b—d Lynx-sunspot analyses and e-g porcupine-sunspot analyses. b, e Wavelet coherency between the population abundance and the sunspot numbers. Colors code for coherence values from dark blue (QO) to dark red (1). Black line indicates the cone of influence that delimits the region not influenced
by edge effects and dashed lines show the a = 5% and « = 10% significance levels computed based on 1,000 Markov bootstrapped series. c, f Phases of the population abundance and the sunspot numbers computed in the 10-12 year periodic band. Dotted line represents the phase difference and the colors are as in a. d, g Distribution of the phase difference of the two considered time series. d Normalized entropy of this phase difference distribution (see Cazelles and Stone 2003) is Q = 0.19 and it is not significant. g Normalized entropy of this distribution is Q =0.47 and it is significant with a P-value computed with 1,000 Markov bootstrapped series less than 0.1% Fig. 8 Comparison between the association between the sunspot numbers and the lynx population and between the sunspot numbers and the porcupine population. a Normalized time series: the sunspot numbers (red line), the lynx abundance (blue line) and the porcupine abundance (black line). Note that the lynx and porcupine time series have a different starting date and different length, which explains the blank spaces in the following graphs. b—d Lynx-sunspot analyses and e-g porcupine-sunspot analyses. b, e Wavelet coherency between the population abundance and the sunspot numbers. Colors code for coherence values from dark blue (QO) to dark red (1). Black line indicates the cone of influence that delimits the region not influenced
signals that contain the major part of the variance are non- stationary, as are the analyzed signals (Fig. 9). For instance, the first intrinsic mode function has the same complexity as the raw signal (Fig. 9c, f) and the fourth and fifth levels of reconstruction signals show a marked shift around f, = 60 (Fig. 9b, e). These decompositions (DWT and EMD) then do not permit direct quantification of the time evolution of the main characteristics of the analyzed time series, however crucial for non-stationary signals. These results (Fig. 9b, e) also stress the high sensitivity of the DWT decomposition to the number of data points, contrary to the continuous wavelet approach. Despite its interesting features, illustrated by our work, wavelet approaches can be improved in different ways. In an ecological context two points warrant further improve- ment: the irregular or uneven sample data, and the multivariate analysis of time series. As regards awkward sampling quality, Keitt and Fischer (2006), analyzing the response of zooplankton communities to anthropogenic disturbance, employed an adaptive “second-generation” wavelet named the “lifting scheme” (Sweldens 1998). These discrete wavelets have the advantage of adapting their shape near sampling gaps and boundaries. This is an important consideration for ecological time series which signals that contain the major part of the variance are non- ifth levels of reconstruction signals show a marked shift time evolution of the main characteristics of the analyzed their shape near sampling gaps and boundaries. This is an
signals that contain the major part of the variance are non- stationary, as are the analyzed signals (Fig. 9). For instance, the first intrinsic mode function has the same complexity as the raw signal (Fig. 9c, f) and the fourth and fifth levels of reconstruction signals show a marked shift around f, = 60 (Fig. 9b, e). These decompositions (DWT and EMD) then do not permit direct quantification of the time evolution of the main characteristics of the analyzed time series, however crucial for non-stationary signals. These results (Fig. 9b, e) also stress the high sensitivity of the DWT decomposition to the number of data points, contrary to the continuous wavelet approach. Despite its interesting features, illustrated by our work, wavelet approaches can be improved in different ways. In an ecological context two points warrant further improve- ment: the irregular or uneven sample data, and the multivariate analysis of time series. As regards awkward sampling quality, Keitt and Fischer (2006), analyzing the response of zooplankton communities to anthropogenic disturbance, employed an adaptive “second-generation” wavelet named the “lifting scheme” (Sweldens 1998). These discrete wavelets have the advantage of adapting their shape near sampling gaps and boundaries. This is an important consideration for ecological time series which signals that contain the major part of the variance are non- ifth levels of reconstruction signals show a marked shift time evolution of the main characteristics of the analyzed their shape near sampling gaps and boundaries. This is an

Related papers

A wavelet analysis to compare environmental time series
Pedro Morettin

2012

Cities such as São Paulo, Tokyo, New York and Mexico City are on the list of the most polluted in the world. PM10 is a major component of air pollution that threatens both our health and our environment. In this paper, we are interested in comparing time series using wavelet analysis. The comparison is made using three statistical procedures, namely, the scalogram, the test given by the ratio of cumulative wavelet periodograms and the analysis of variance. These methods are applied to compare the rates of hourly PM10 in four different districts of the city of São Paulo, Brazil. For the analysis we use the discrete wavelet transformation (DWT) considering the Haar and Daubechies wavelets. In the analysis of variance for the wavelet coefficients, we tested the local effect considering the months from June to October as replications. Scalograms were constructed for each series and we note that for the two wavelet bases used they presented different behavior leading to the conclusion that the series of energies are different. The effect of location was significant in the analysis of variance considering levels 4 ≤ j for both bases Haar and Daubechies DWT. We also consider a statistical test at each level j given by the ratio of Thelma Sáfadi and Pedro A. Morettin 80 the cumulative wavelet periodograms of the series. Again we could find that the series are generated by different processes.

View PDFchevron_right
Analysing multiple time series and extending significance testing in wavelet analysis
Jean-marc Fromentin

Marine Ecology Progress Series, 2008

In nature, non-stationarity is rather typical, but the number of statistical tools allowing for non-stationarity remains rather limited. Wavelet analysis is such a tool allowing for nonstationarity but the lack of an appropriate test for statistical inference as well as the difficulty to deal with multiple time series are 2 important shortcomings that limits its use in ecology. We present 2 approaches to deal with these shortcomings. First, we used 1/ƒ β models to test cycles in the wavelet spectrum against a null hypothesis that takes into account the highly autocorrelated nature of ecological time series. To illustrate the approach, we investigated the fluctuations in bluefin tuna trap catches with a set of different null models. The 1/ƒ β models approach proved to be the most consistent to discriminate significant cycles. Second, we used the maximum covariance analysis to compare, in a quantitative way, the time-frequency patterns (i.e. the wavelet spectra) of numerous time series. This approach built cluster trees that grouped the wavelet spectra according to their time-frequency patterns. Controlled signals and time series of sea surface temperature (SST) in the Mediterranean Sea were used to test the ability and power of this approach. The results were satisfactory and clusters on the SST time series displayed a hierarchical division of the Mediterranean into a few homogeneous areas that are known to display different hydrological and oceanic patterns. We discuss the limits and potentialities of these methods to study the associations between ecological and environmental fluctuations.

View PDFchevron_right
Detecting climate-induced patterns using wavelet analysis
GA Bradshaw

Environmental Pollution, 1994

One of the difficulties encountered in the detection of ecosystem responses to climate change is distinguishing climate-induced patterns from those created by other sources. For example, changes in the trend of stream discharge records over time may reflect a composite response of changes in the climate (i.e. precipitation and temperature), land-use (e.g. timber harvesting and grazing), and local basin characteristics. Methods which quantify and relate information of temporal and spatial patterns across scales are critical to assess climatically induced changes in the forest and stream ecosystems. A methodology utilizing wavelet analysis is introduced for the purpose of identifying and isolating inferred climatic components of the hydrologic record Trends observed in stream discharge records from eastern Oregon, USA are identified and used to illustrate the utility of a new time series technique, wavelet analysis, as a complementary approach for discerning pattern. This methodology affords an informed procedure for choosing filter dimensions for the purpose of signal decomposition. The wavelet cross-covariance is applied to precipitation and discharge records to identify the climatic component in the discharge record. Reconstruction of these dominant frequencies is effected to isolate the climatic components. The discharge pattern shows two dominant scales of pattern coincident with the precipitation record A 3-year half-period pattern is found to be correlated with the Southern Oscillation Index at the same frequency.

View PDFchevron_right
The use of the Mexican Hat and the Morlet wavelets for detection of ecological patterns
Keping Ma

Plant Ecology, 2005

In this paper, we compare the relationship between scale and period in ecological pattern analysis and wavelet analysis. We also adapt a commonly used wavelet, the Morlet, to ecological pattern analysis. Using Monte Carlo assessments, we apply methods of statistical significance test to wavelet analysis for pattern analysis. In order to understand the inherent strength and weakness of the Morlet and the Mexican Hat wavelets, we also investigate and compare the properties of two frequently used wavelets by testing with field data and four artificial transects of different typical patterns which is often encountered in ecological research. It is shown that the Mexican Hat provides better detection and localization of patch and gap events over the Morlet, whereas the Morlet offers improved detection and localization of scale over the Mexican Hat. There is always a trade-off between the detection and localization of scale versus patch and gap events. Therefore, the best composite analysis is the combination of their advantages. The properties of wavelet in dealing with ecological data may be affected by characteristics intrinsic to wavelet itself. The peaks of different scales in isograms of wavelet power spectrum from the Mexican Hat may overlap with each other. Alternatively, these peaks of different scales in isograms of wavelet power spectrum may combine with each other unless the size of the analyzed scales is significantly different. These overlapping or combining lead to combining of peaks for different scales, or the masking of trough between peaks of different scales in the scalogram. Ecologists should combine all the information in scalogram and isograms of wavelet coefficient and wavelet power spectrum from different wavelets, which can provide us a broader view and precise pattern information.

View PDFchevron_right
Extracting oscillating components from nonstationary time series: A wavelet-induced method
Samuel Nicolay

Physical Review E, 2017

This paper consists in the description and application of a method called wavelet-induced mode extraction (WIME) in the context of time-frequency analysis. WIME aims to extract the oscillating components that build amplitude modulated-frequency modulated signals. The essence of this technique relies on the successive extractions of the dominant ridges of wavelet-based time-frequency representations of the signal under consideration. Our tests on simulated examples indicate strong decomposition and reconstruction skills, trouble-free handling of crossing trajectories in the time-frequency plane, sharp performances in frequency detection in the case of mode-mixing problems, and a natural tolerance to noise. These results are compared with those obtained with empirical mode decomposition. We also show that WIME still gives meaningful results with real-life data, namely, the Oceanic Niño Index.

View PDFchevron_right
Using Wavelet to Analyze Periodicities in Hydrologic Variables
Sajjad Ahmad

World Environmental and Water Resources Congress 2017, 2017

The trend and shift in the seasonal temperature, precipitation and streamflow time series across the Midwest have been analyzed, for the period 1960-2013, using the statistical analyses (Mann-Kendall test with and without considering short term persistence (MK2 and MK1, respectively) and Pettitt test). The paper also utilizes a relatively new approach, wavelet analysis, for testing the existence of trend and shift in the time series. The method has the ability to decompose a time series in to lower (trend) and higher frequency components (noise). Discrete wavelet transform (DWT) has been employed in the present study with an aim to find which periodicities are mainly responsible for trend in the original data. The combination of MK1, MK2 and DWT along with Pettitt test hasn't been extensively used up to this time, especially in detecting trend and shift in the Midwest. The analysis of climate division temperature and precipitation data and USGS naturalized streamflow data revealed the presence of periodicity in the time series data. All the incorporated time series data were seasonal to analyze the trends and shifts for four seasons-winter, spring, summer and fall independently. D3 component of DWT were observed to be influential in detecting real trend in Temperature, precipitation and streamflow data, however unlike temperature, precipitation and streamflow showed decreasing trend as well. Shift was relatively observed more than trend in the region with dominance of D3 component in the data. The result indicate the significant warming trend which agrees with the "increasing temperature" observations in the past two decades, however a clear explanation for precipitation and streamflow is not obvious.

View PDFchevron_right
Wavelets in time-series analysis
Rainer Von Sachs

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1999

This article reviews the role of wavelets in statistical time series analysis. We survey work that emphasises scale such as estimation of variance and the scale exponent of a process with a speci c scale behaviour such as 1/f processes. We present some of our own work on locally stationary wavelet (lsw) processes which model both stationary and some kinds of non-stationary processes. Analysis of time series assuming the lsw model permits identi cation of an evolutionary wavelet spectrum (ews) that quanti es the variation in a time series over a particular scale and at a particular time. We address estimation of the ews and show how our methodology reveals phenomena of interest in an infant electrocardiogram series.

View PDFchevron_right
Can Wavelet Analysis detect low frequency periodicities in climatic time series ?
Assaf Hochman

2016

The Continuous Wavelet Transform (CWT) is a frequently used tool to study periodicity in climate and other time series. Periodicity plays a significant role in climate reconstruction and prediction. In numerous studies, the use of CWT revealed Dominant Periodicity (DP) in climatic time series. Several studies suggested that these "natural oscillations" would even reverse global warming. It is shown here that the results of wavelet analysis for detecting DPs can be miss-interpreted in the presence of local singularities that are manifested in lower frequencies. This may lead to false DPs detection. In CWT analysis of synthetic and real-data climatic time series, with local singularities, CWT indicates on a low frequency DP even if there is no true periodicity in the time series. It is argued that this is an inherent general property of CWT. Hence, applying CWT to climatic time series should be re-evaluated and more careful analysis of the entire wavelet power spectrum is required, focusing on high frequencies as well. Thus, a cone-like shape in the wavelet power spectrum most likely indicates the presence of a local singularity in the time series rather than a DP, even if the local singularity has an observational or a physical basis. It is shown that analyzing the derivatives of the time series may be helpful in interpreting the wavelet power spectrum. Nevertheless, this is only a partial remedy that does not completely neutralize the effects caused by the presence of local singularities.

View PDFchevron_right
A wavelet analysis for time series
Pedro Morettin

Journal of Nonparametric Statistics, 1998

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

View PDFchevron_right
Spatiotemporal modeling of saturated dissolved oxygen through regressions after wavelet denoising of remotely and proximally sensed data. Earth Science Informatics, 1-8
Fatih Evrendilek
View PDFchevron_right

References (66)

  1. Barbraud C, Weimerskirch H (2001) Emperor penguins and climate change. Nature 411:183-186
  2. Bartlett MS (1954) Proble `mes de l'analyse spectrale des se ´ries temporelles stationnaires. Publ Inst Stat Univ Paris 3:119-134
  3. Benton TG, Plaistow SJ, Coulson TN (2006) Complex population dynamics and complex causation: devils, details and demogra- phy. Proc R Soc Lond B 273:1173-1181
  4. Bierman SM, Fairbairn JP, Petty SJ, Elston DA, Tidhar D, Lambin X (2006) Changes over time in the spatiotemporal dynamics of cyclic populations of field voles (Microtus agrestis L.). Am Nat 167:583-590
  5. Bjørnstad ON, Grenfell BT (2001) Noisy clock: time series analysis of population fluctuations in animals. Science 293:638-643
  6. Bradshaw GA, Spies TA (1992) Characterizing canopy gap structure in forest using wavelet analysis. J Ecol 80:205-215
  7. Buonaccorsi JP, Elkinton JS, Evans SR, Liebhold AM (2001) Measuring and testing spatial synchrony. Ecology 82:1628-1679
  8. Cazelles B (2001) Blowout bifurcation with non-normal parameters in population dynamics. Phys Rev E 64:032901
  9. Cazelles B (2004) Symbolic dynamics for identifying similarity between rhythms of ecological time series. Ecol Lett 7:755-763
  10. Cazelles B, Chau NP (1997) Using the Kalman filter and dynamic models to assess the changing HIV/AIDS epidemic. Math Biosci 140:131-154
  11. Cazelles B, Hales S (2006) Infectious diseases, climate influences and nonstationarity. PLoS Med 3:1212-1213 (e328)
  12. Cazelles B, Stone L (2003) Detection of imperfect population synchrony in an uncertain world. J Anim Ecol 72:953-968
  13. Cazelles B, Bottani S, Stone L (2001) Unexpected coherence and conservation. Proc Royal Soc Lond B 268:2595-2602
  14. Cazelles B, Chavez M, McMichael AJ, Hales S (2005) Nonstationary influence of El Nin ˜o on the synchronous dengue epidemics in Thailand. PLoS Med 2:313-318 (e106)
  15. Chatfield JR (1989) The analysis of time series: an introduction. Chapman & Hall, London
  16. Cummings DAT,, Irizarry RA, Huang NE, Endy TP, Nisalak A, Ungchusak K, Burke DS (2004) Travelling waves in the occurrence of dengue haemorrhagic fever in Thailand. Nature 427:344-347
  17. Cushing JM, Dennis B, Desharnais RA, Costantino RF (1998) Moving toward an unstable equilibrium: saddle nodes in population systems. J Anim Ecol 67:298-306
  18. Dale MRT, Mah M (1998) The use of wavelets for spatial pattern analysis in ecology. J Veg Sci 9:805-815
  19. Daubechies I (1992) Ten lectures on wavelets. SIAM monographs, Philadelphia
  20. Duncan CJ, Duncan SR, Scott S (1996) Whooping cough epidemics in London, 1701-1812: infection dynamics, seasonal forcing and the effects of malnutrition. Proc R Soc Lond B 263:445-450
  21. Efron B, Tibshirani RJ (1993) An introduction to the bootstrap. Chapman & Hall, London
  22. Elton CS (1924) Fluctuations in the numbers of animals, their causes and effects. Br J Exp Biol 2:119-163
  23. Elton CS, Nicholson M (1942) The ten-year cycle in numbers of the lynx in Canada. J Anim Ecol 11:215-244
  24. Forchhammer MC, Post E (2004) Using large-scale climate indices in climate change ecology studies. Popul Ecol 46:1-12
  25. Foster G (1996) Wavelets for period analysis of unevenly sampled time series. Astronom J 112:1709-1729
  26. Gabor D (1946) Theory of communication. J Inst Electr Eng 93:429- 457
  27. Grenfell BT, Bjørnstad ON, Kappey J (2001) Travelling waves and spatial hierarchies in measles epidemics. Nature 414:716-723
  28. Hare SR, Mantua NJ (2000) Empirical evidence for North Pacific regime shifts in 1997 and 1989. Prog Oceanogr 47:103-145
  29. Hastings A (2001) Transient dynamics and persistence of ecological systems. Ecol Lett 4:215-220
  30. Haydon DT, Greenwood PE, Stenseth NC, Saitoh T (2003) Spatio- temporal dynamics of the grey-sided vole in Hokkaido: identi- fying coupling using state-based Markov-chain modelling. Proc R Soc Lond B 270:435-445
  31. Huang NE, Shen Z, Long SR, Wu MC, Shih HH, Zheng Q, Yen NC, Tung CC, Liu HH (1998) The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proc R Soc Lond A 454:903-995
  32. Ives AR, Dennis B, Cottingham KL, Carpenter SR (2003) Estimating community stability and ecological interactions from time series data. Ecol Monogr 73:301-330
  33. Jenouvrier S, Weimerskirch H, Barbraud C, Park YH, Cazelles B. (2005) Evidence of a shift in the cyclicity of Antarctic seabirds dynamics linked to climate. Proc R Soc Lond B 272:887-895
  34. Johnson DM, Bjørnstad ON, Liebhold AM (2006) Landscape mosaic induces traveling waves of insect outbreaks. Oecologia 148:51- 60
  35. Jose ´MV, Bishop RF (2003) Scaling properties and symmetrical patterns in the epidemiology of rotavirus infection. Phil Trans R Soc Lond B 358:1625-1641
  36. Keitt TH, Fischer J (2006) Detection of scale-specific community dynamics using wavelets. Ecology 87:2895-2904
  37. Keitt TH, Urband DL (2005) Scale-specific inference using wavelets. Ecology 86:2497-2504
  38. Klvana I, Berteaux D, Cazelles B (2004) Porcupine feeding scars and climatic data show ecosystem effects of the solar cycle. Am Nat 164:283-297
  39. Koelle K, Pascual M (2004) Disentangling extrinsic from intrinsic factors in disease dynamics: a nonlinear time series approach with an application to cholera. Am Nat 163:901-913
  40. Lau KM, Weng H (1995) Climatic signal detection using wavelet transform: how to make a time series sing. Bull Am Meteorol Soc 76:2391-2402
  41. Liebhold AM, Koening WD, Bjørnstad ON (2004) Spatial synchrony in population dynamics. Annu Rev Ecol Evol Syst 35:467-490
  42. Liu PC (1994) Wavelet spectrum analysis and ocean windwaves. In: Foufoula-Georgiou E, Kumar P (eds) Wavelets in geophysics. Academic Press, New York, pp 151-166
  43. Mackenzie JMD (1952) Fluctuations in the number of British tetraonids.J Anim Ecol 21:128-153
  44. Mallat S (1998) A wavelet tour of signal processing. Academic Press, San Diego Me ´nard F, Marsac F, Bellier B, Cazelles B (2007) Climatic oscillations and tuna catch rates in the Indian Ocean: a wavelet approach to time series analysis. Fish Oceanogr 16:95-104
  45. Meyers SD, Kelly BG, O'Brien JJ (1993) An introduction to wavelet analysis in oceanography and meteorology with application to the dispersion of Yanai waves. Mon Weather Rev 121:2858- 2866
  46. Mi X, Ren H, Ouyang Z, Wei W, Ma K (2005) The use of the Mexican hat and the Morlet wavelets for detection of ecological patterns. Plant Ecol 179:1-19
  47. Murdoch WW, Kendall BE, Nisbet RM, Briggs CJ, McCauley E, Bolser R (2002) Single-species models for many-species food webs. Nature 423:541-543
  48. Nezlin NP, Li BL (2003) Time-series analysis of remoted-sensed chlorophyll and environmental factors in the Santa Monica-San Pedro basin off Southern California. J Mar Syst 39:185-202
  49. Pikovsky AS, Rosenblum MG, Kurths J (2001)Synchronization: a universal concept in nonlinear sciences.Cambridge University Press, Cambridge
  50. Platt T, Denman KL (1975) Spectral analysis in ecology. Annu Rev Ecol Syst 6:189-210
  51. Reid PC, Borges MF, Svendsen E (2001) A regime shift in the North Sea circa in 1988 linked to changes in the North Sea horse mackerel fishery. Fish Res 50:163-171
  52. Rodo ´X., Pascual M, Fuchs G, Faruque SG (2002) ENSO and cholera: a nonstationary link related to climate change? Proc Natl Acad Sci USA 99:12901-12906
  53. Rodriguez-Arias MA, Rodo ´X (2004) A primer on the study of transitory dynamics in ecological series using the scaledependent correlation analysis. Oecologia 138:485-504
  54. Rohani P, Earn DJD, Grenfell BT (1999) Opposite patterns of synchrony in sympatric disease metapopulations. Science 286:968-971
  55. Rohani P, Green CJ, Mantilla-Beniers NB, Grenfell BT (2003) Ecological interference between fatal diseases. Nature 422:885- 888
  56. Rosenberg M (2004) Wavelet analysis for detecting anisotropy in point patterns. J Veg Sci 15:277-284
  57. Rouyer T, Fromentin JM, Stenseth NC, Cazelles B (2008) Analysing multiple time series and extending significance testing in wavelet. Mar Ecol Progr Ser (in press)
  58. Royama T (1992) Analytical population dynamics. Chapman & Hall, London
  59. Saitoh T, Cazelles B, Vik JO, Viljugrein H, Stenseth NC (2006) Effects of the regime shift on population dynamics of the grey- sided vole in Hokkaido, Japan. Clim Res 32:109-118
  60. Sinclar ARE, Gosline JM, Holdsworth G, Krebs CJ, Boutin S, Smith JNM, Boonstra R, Dale M (1993) Can the solar cycle and climate synchronize the snowshoe hare cycle in Canada? Evidence from the tree rings and ice cores. Am Nat 141:173-198
  61. Stenseth NC, Mysterud A, Ottersen G, Hurrel JW, Chan KS, Lima M (2002) Ecological effects of climate fluctuations. Science 297:1292-1296
  62. Sweldens W (1998) The lifting scheme: a construction of second generation wavelets. SIAM J Math Anal 29:511-546
  63. Theiler J, Eubank S, Longtin A, Galdrikian B, Farmer JD (1992) Testing for nonlinearity in time series: the method of surrogate data. Physica D 58:77-94
  64. Torrence C, Compo GP (1998) A practical guide to wavelet analysis. Bull Am Meteorol Soc 79:61-78
  65. Vautard R, Yiou P, Ghil M (1992) Singular spectrum analysis: a toolkit for short, noisy chaotic signals. Physica D 58:95-126
  66. Xia Y, Bjørnstad ON, Grenfell BT (2004) Measles metapopulation dynamics: a gravity model for epidemiological coupling and dynamics. Am Nat 164:267-281

Related papers

Application of wavelet analysis to ecological data
Tae-soo Chon

Ecological Informatics, 2006

Ecological data are difficult to analyze due to complexity residing in the ecological systems with the variables varying in non-linear fashion. Efficient methods are required to properly extract information out of the complex data. Wavelets have good time-frequency (timescale) localization, can represent data parsimoniously, and can be implemented with very fast algorithms. Brief backgrounds and computational aspects of wavelets were outlined for implementation to ecological data analysis. Wavelets are well suited for building mathematical models of ecological data and the statistical analysis of combined effects of complex factors in ecological network. Wavelet based analysis and synthesis may lead researchers in ecological studies to new insights and novel theories for understanding complex ecological and environmental phenomena.

View PDFchevron_right
An introduction to wavelet analysis with applications to vegetation time series
Muyin Wang

Community Ecology, 2004

Wavelets are relatively new mathematical tools that have proven to be quite useful for analyzing time series and spatial data. We provide a basic introduction to wavelet analysis that concentrates on their interpretation in the context of analyzing time series.

View PDFchevron_right
Wavelet procedures applied in ecosystem models
Vladimir Balan

2007

The Haar wavelet decomposition of the numerical solution for the Truscott-Brindley 3-dimensional SODE produces sets of coe-cients which contain essential information on the local oscillating patterns of the population evolution. The segmentation of the initial sample flnite series gives besides an increase of speed of the decomposition process ((3)), via the analysis of partial energies and Shannon entropy, information on the prevalence of wavelet-type evolution periodic behavior.

View PDFchevron_right
Wavelet analysis in ecology and epidemiology: impact of statistical tests
Bernard Cazelles

Journal of The Royal Society Interface, 2013

Wavelet analysis is now frequently used to extract information from ecological and epidemiological time series. Statistical hypothesis tests are conducted on associated wavelet quantities to assess the likelihood that they are due to a random process. Such random processes represent null models and are generally based on synthetic data that share some statistical characteristics with the original time series. This allows the comparison of null statistics with those obtained from original time series. When creating synthetic datasets, different techniques of resampling result in different characteristics shared by the synthetic time series. Therefore, it becomes crucial to consider the impact of the resampling method on the results. We have addressed this point by comparing seven different statistical testing methods applied with different real and simulated data. Our results show that statistical assessment of periodic patterns is strongly affected by the choice of the resampling method, so two different resampling techniques could lead to two different conclusions about the same time series. Moreover, our results clearly show the inadequacy of resampling series generated by white noise and red noise that are nevertheless the methods currently used in the wide majority of wavelets applications. Our results highlight that the characteristics of a time series, namely its Fourier spectrum and autocorrelation, are important to consider when choosing the resampling technique. Results suggest that data-driven resampling methods should be used such as the hidden Markov model algorithm and the 'beta-surrogate' method.

View PDFchevron_right
Wavelet Transform Application for/in Non-Stationary Time-Series Analysis: A Review
manel rhif

Applied Sciences, 2019

Non-stationary time series (TS) analysis has gained an explosive interest over the recent decades in different applied sciences. In fact, several decomposition methods were developed in order to extract various components (e.g., seasonal, trend and abrupt components) from the non-stationary TS, which allows for an improved interpretation of the temporal variability. The wavelet transform (WT) has been successfully applied over an extraordinary range of fields in order to decompose the non-stationary TS into time-frequency domain. For this reason, the WT method is briefly introduced and reviewed in this paper. In addition, this latter includes different research and applications of the WT to non-stationary TS in seven different applied sciences fields, namely the geo-sciences and geophysics, remote sensing in vegetation analysis, engineering, hydrology, finance, medicine, and other fields, such as ecology, renewable energy, chemistry and history. Finally, five challenges and future w...

View PDFchevron_right

Related topics

  • Ecology
  • Wavelet Analysis
  • Wavelet Decomposition
  • Time Factors

    • Cited by

    Seasonality of meningitis in Africa and climate forcing: aerosols stand out
    Hélène BROUTIN

    Journal of the Royal Society, Interface / the Royal Society, 2013

    Bacterial meningitis is an ongoing threat for the population of the African Meningitis Belt, a region characterized by the highest incidence rates worldwide. The determinants of the disease dynamics are still poorly understood; nevertheless, it is often advocated that climate and mineral dust have a large impact. Over the last decade, several studies have investigated this relationship at a large scale. In this analysis, we scaled down to the district-level weekly scale (which is used for in-year response to emerging epidemics), and used wavelet and phase analysis methods to define and compare the time-varying periodicities of meningitis, climate and dust in Niger. We mostly focused on detecting time-lags between the signals that were consistent across districts. Results highlighted the special case of dust in comparison to wind, humidity or temperature: a strong similarity between districts is noticed in the evolution of the time-lags between the seasonal component of dust and meningitis. This result, together with the assumption of dust damaging the pharyngeal mucosa and easing bacterial invasion, reinforces our confidence in dust forcing on meningitis seasonality. Dust data should now be integrated in epidemiological and forecasting models to make them more realistic and usable in a public health perspective.

    View PDFchevron_right
    Reconstruction of a 1,910-y-long locust series reveals consistent associations with climate fluctuations in China
    Kyrre Kausrud

    Proceedings of The National Academy of Sciences, 2011

    View PDFchevron_right
    Periodic temperature-associated drought/flood drives locust plagues in China
    Huidong TIAN

    Proceedings of The Royal Society B: Biological Sciences, 2009

    Global warming is currently of great concern. Yet the ecological effects of low-frequency climate variations remain largely unknown. Recent analyses of interdecadal variability in population abundance of the Oriental migratory locust (Locusta migratoria manilensis) in China have revealed negative associations with temperature and positive associations with Yangtze drought and flood frequencies during the past millennium (AD 957-1956). In order to shed new light on the causal relationships between locust abundance, floods, droughts and temperature in ancient China, we used wavelet analysis to explore how the coherencies between the different variables at different frequencies have been changed during the past millennium. We find consistent in-phase coherencies between locusts and drought/flood frequencies, and out-of-phase coherencies between locusts and temperature and between drought/flood and temperature at period components of 160-170 years. Similar results are obtained when historical data of drought/flood frequencies of the Yangtze Delta region are used, despite flood data showing a weak and somewhat inconsistent association with other factors. We suggest that previously unreported periodic cooling of 160-170-year intervals dominate climatic variability in China through the past millennium, the cooling events promoting locust plagues by enhancing temperature-associated drought/flood events. Our results signify a rare example of possible benign effects of global warming on the regional risk of natural disasters such as flood/drought events and outbreaks of pest insects.

    View PDFchevron_right
    Nonlinear impacts of climatic variability on the density-dependent regulation of an insect vector of disease
    Luis Chaves

    Global Change Biology, 2012

    Aedes aegypti is one of the most common urban tropical mosquito species and an important vector of dengue, chikungunya, and yellow fever viruses. It is also an organism with a complex life history where larval stages are aquatic and adults are terrestrial. This ontogenetic niche shift could shape the density-dependent regulation of this and other mosquito species, because events that occur during the larval stages impact adult densities. Herein, we present results from simple density-dependent mathematical models fitted using maximum likelihood methods to weekly time series data from Puerto Rico and Thailand. Density-dependent regulation was strong in both populations. Analysis of climatic forcing indicated that populations were more sensitive to climatic variables with low kurtosis, i.e., climatic factors highly variable around the median, rainfall in Puerto Rico, and temperature in Thailand. Changes in environmental variability appear to drive sharp changes in the abundance of mosquitoes. The identification of density-independent (i.e., exogenous) variables forcing sharp changes in disease vector populations using the exogenous factors statistical properties, such as kurtosis, could be useful to assess the impacts of changing climate patterns on the transmission of vector-borne diseases.

    View PDFchevron_right
    Cutaneous Leishmaniasis and Sand Fly Fluctuations Are Associated with El Niño in Panamá
    Luis Chaves, José Calzada

    PLoS Neglected Tropical Diseases, 2014

    Background: Cutaneous Leishmaniasis (CL) is a neglected tropical vector-borne disease. Sand fly vectors (SF) and Leishmania spp parasites are sensitive to changes in weather conditions, rendering disease transmission susceptible to changes in local and global scale climatic patterns. Nevertheless, it is unclear how SF abundance is impacted by El Niñ o Southern Oscillation (ENSO) and how these changes might relate to changes in CL transmission.

    View PDFchevron_right
    580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved