CENTRE ON THE PROBLEMS OF ECOLOGY AND PRODUCTIVITY OF FORESTS
|
|
CENTRE ON THE PROBLEMS OF ECOLOGY AND PRODUCTIVITY OF FORESTS |
RUSSIAN ACADEMY OF SCIENCES |
|
|
Nedorezov L.V., Sadykova D.L. Analysis of Population Time Series using Discrete Dynamic Models from the Example of Green Oak Leaf Roller // Forest Science (Lesovedenie). 2010. No 2. P. 14-26 (in Russian). This paper presents the analysis of time series using discrete time models of population dynamics. Ways of the application of various statistical criteria that are used for the identification of model parameters are discussed. Some criteria are shown to have “double standards” with respect to elements of experimental sampling. A step procedure for the estimation of self-regulatory mechanisms is proposed in order to reveal the main mechanism responsible for the control of insect populations. Based on the data on the changes in the number of the green oak leaf roller (Tortrix viridana L.), the possibilities of the approach to the analysis of its population dynamics are considered. The known models of population dynamics with discrete time steps (in particular, Skellam model, Moran-Ricker model, Hasseli model, Maynard Smith-Slatkin model, and the discrete logistic model) that have a set of various dynamic regimes, including cycles of different lengths and chaos, were used. The conclusions of the possibility to use concrete models for the data approximation are based on analyzing the deviations of theoretical (model) data from experimental ones: The correspondence of the distribution of deviations to the normal one with the zero average was verified using the Kolmogorov-Smirnov and Shapiro-Wilk tests, as well as the existence/absence of serial correlation using the Durbin-Warson criteria. Mathematical models, assessment of parameters, analysis of time series, green oak leaf roller. |
webmaster |