Background Remotely-sensed environmental data from earth-orbiting satellites are progressively used to

Background Remotely-sensed environmental data from earth-orbiting satellites are progressively used to magic size the distribution and abundance of both flower and animal varieties, especially those of economic or conservation importance. in the estimation of the amplitudes and phases of the Fourier harmonics. Methodology/Principal Findings We present a 112809-51-5 novel spline-based algorithm that overcomes the processing problems of composited MODIS data. The algorithm is definitely tested on artificial data generated using randomly selected ideals of both amplitudes and phases, and provides an accurate estimate of the input variables under all conditions. The algorithm was then applied to create layers that capture the seasonality in MODIS data for the period from 2001 to 2005. Conclusions/Significance Global temporal Fourier processed images of 1 1 km MODIS data for Middle Infrared Reflectance, day time- and night-time Land Surface Temp (LST), Normalised Difference Vegetation Index (NDVI), and Enhanced Vegetation Index (EVI) are offered for ecological and epidemiological applications. The finer spatial and temporal resolution, combined with the higher geolocational and spectral accuracy of the MODIS tools, compared with earlier multi-temporal data units, mean that these data may be used with greater confidence in varieties’ 112809-51-5 distribution modelling. Intro Environmental variables, such as temp and vegetation greenness, are important determinants of the distributions of many varieties [1]. The presence or absence of a varieties in any area is definitely often distinguished not only by the complete levels of weather or vegetation ideals, but also by delicate variations in the seasonality of these variables [2], which can only become captured by repeated measurements over time. Such time series may be derived from ground-based meteorological records, 112809-51-5 but acquiring spatially continuous, global records of these environmental variables is only practical using remotely sensed data from Earth-orbiting satellites. Historically, the National Oceanographic and Atmospheric Administration (NOAA) series of satellites transporting the Advanced Very High Resolution Radiometer (AVHRR) have provided time series of global imagery more or less continually since 1981 [3]C[5]. These time series have been used to produce, among others, images of Land Surface Temp (LST) [6] and of the Normalised Difference Vegetation Index (NDVI), a correlate of vegetation productivity, biomass and climatic conditions [7]. Serial correlation among successive observations taken over a period of time reduces the statistical energy of captured imagery. Data reduction (ordination) methods are usually employed to remove these correlations and provide one or more transformed images without such correlation, which can then be used in further analyses or applications. One ordination approach generally applied to multi-temporal imagery is definitely principal parts analysis (PCA, e.g. [8]), but explicit Rabbit Polyclonal to ACBD6 actions of seasonality are misplaced in the ordination process. PCA therefore achieves data reduction at the expense of biological descriptiveness. Alternative methods that retain information about seasonality include polynomial functions [9], [10] and temporal Fourier analysis [11]C[19]. Temporal Fourier analysis (TFA) transforms a series of observations taken at intervals over a period of time into a set of (uncorrelated) sine curves, or harmonics, of different frequencies, amplitudes and phases that collectively sum to the original time series. For many multi-temporal satellite data, the most important harmonics are those that correspond to the annual, bi-annual and tri-annual cycles of seasonal changes, and these harmonics often have a definite biological interpretation [13]. Both longer period cycles (variance on inter-annual scales) and shorter period cycles (high rate of recurrence intra-annual variance) can also be recognized by TFA, but tend to become less important biologically, as well as in terms of their contributions to the overall variance of the transmission [13]. Therefore TFA achieves data ordination inside a biologically transparent way. An additional advantage of TFA is definitely that it can be used to clean noisy data. Fourier analysis moves between the time and rate of recurrence domains: forward analysis produces a rate of recurrence website representation of the original time series and inverse analysis moves from your rate of recurrence website back to the time website. Filtering noisy data is easier in the rate of recurrence website because most noise is definitely associated with high frequencies which can therefore become fallen before inversion to produce a smoothed version of the original time series. Equal filtering in the time website is definitely less straightforward, because the high rate of recurrence components are combined in with all other rate of recurrence components and so cannot easily become separated from them. Different examples of smoothing happen when different rate of recurrence ranges are excluded during the filtering process. Here the primary objective is not to clean the data, but to capture their seasonality. Smoothing should be regarded as an additional advantage of the Fourier approach to capturing seasonality; an advantage that is definitely all the more important when, for numerous reasons, the satellite transmission is definitely above (e.g. sun-glint) or below (e.g. cloud contamination) its right value. Until relatively recently, global remotely sensed time series data have been available either with low spatial resolution for long time periods (e.g. 20 years of AVHRR at 8 112809-51-5 km resolution) or with higher resolution for any shorter time.