Detecting the order of population dynamics from time series: nonlinearity causes spurious diagnosis

Partial autocorrelation and partial rate correlation functions are frequently used to detect the order of the endogenous process generating an observed population time series. Here we uncover a problem with this approach: the diagnosis of spurious second order autocorrelation due to strong nonlinearity in a first order endogenous process, as exemplified by time series data from a population of Soay sheep. Causes and a possible solution are discussed.     

Nonylphenolic compounds in drinking and surface waters downstream of treated textile and pulp and paper effluents: a survey and preliminary assessment of their potential effects on public health and aquatic life

Eleven drinking water treatment plants, located downstream of textile plants or pulp and paper mills, have been sampled monthly during a year for the analysis of 17 nonylphenol ethoxylates (NP1-17EO) and two nonylphenoxycarboxylic acids (NP1-2EC). At all but one plant, results in the drinking water, for the sum of these 19 substances, range between below detection levels and 6.7 microg/l. Annual means are between 0.02 and 2.8 microg/l. At the other plant, the yearly average concentration is 10.4 microg/l and the monthly maximum is 43.3 microg/l. In the surface (pre-treatment) water, the annual mean concentrations of the 11 plants range between 0.14 and 17.8 microg/l and the recorded instantaneous maximum is 55.3 microg/l. According to Canadian health authorities, drinking water is a negligible route of human exposure to nonylphenolic compounds, even at the highest concentrations found in this study. After transformation of the data into nonylphenol equivalents, about 20% of the surface water samples exceed the Canadian 1 microg/l nonylphenol water quality guideline for the protection of aquatic life. Some results also exceed Québec's 6 microg/l nonylphenol guideline. The efficiency of the plants in removing nonylphenolic compounds from drinking water is highly variable, ranging from 11% to 99%.