Issues of concern in environmental flows, such as flammability, malodor and toxicity, generally require a specification of the probability density function (PDF) of contaminant concentration. A very important problem is that of assessing risk and hazard when a quantity of contaminant is released in the neutral atmospheric boundary layer.One can derive the PDF from a knowledge of all of its integral moments.It is generally expected that a very good approximation for the PDF for contaminant concentrations that appear in most turbulent flows can be obtained from the four lowest order moments.
A simple model, which describes the distributed moments of the PDF in terms of two functions of downstream distance in a steady flow, has received a considerable amount of direct and indirect experimental validation. Calculation of the second moment, taking advantage of this model and using a physically plausible and greatly simplified closure scheme, has provided reasonable comparison with the limited amount of available experimental data. Experimental evidence from plume experiments in grid turbulence suggest that one of the two functions in the simple model depends slightly on moment order and this slight dependence is significant in reconstructing the PDF. In this presentation, the third moment is derived and compared with the measured wind-tunnel plume results.

FOTINI   LABROPULU
Luther College - Mathematics
University of Regina
Regina, SK
S4S 0A2, Canada
fotini.labropulu@uregina.ca
 
 

As part of clean up efforts, pollution trends may change.  One approach for evaluating change is to compute and compare trends given a known time of change in trend.  A method for testing the equivalence of two modified Kendall’s Tau non-parametric correlation coefficients, neither necessarily equal to zero is presented.  Several estimators of the variance of a modified estimator are evaluated using simulation.  A bootstrap estimate, the standard null-case variance estimate and a delta method variance estimate are shown to provide reasonable variance estimates.  The results of a power study indicate that in the presence of serial correlation, a method based on the effective sample size bootstrap allowed the hypothesis test to hold its level.  An example demonstrates how the test may be used to detect change in trend in water quality parameters.

ERIC P. SMITH
Department of Statistics
Virginia Tech
Hutcheson Hall
Blacksburg, VA 24061-0439, USA
epsmith@vt.edu
 
 

Sulphur dioxide is considered as the most important contributor to the acidification problem and its threat to the health of the environment. Therefore, monitoring programs were established to measure the concentrations of SO2 in the air and wet deposition with objective of using the data to detect and estimate both the spatial and temporal trends in SO2. This paper presents the results of performing trend analysis for SO2 Finnish air and Canadian lakes water quality data.

JARI WALDEN
Department of Air Quality Research
Finnish Meteorological Institute
Sahaajankatu 20 E
Helsinki 00810, Finland
jari.walden@fmi.fi