If I understood correctly, I wouldn`t use parametric tests, it would seem that they are not spacially independent.
One way would be to model it thru geostatistical analyses, the other is to use a gradient analyses procedure, such as CANOCO and have either an artifical variable composed of altitud + hydrological position or a subjective dummy such as a 4 value, position on the landscape, the "field book forf describing and sampling soils" (it`s on the web somewhere), has a 5 value hillslope definition: summit, shoulder, backslope, footslope and toeslope. Hope it helps, --- Alexandre Souza <[EMAIL PROTECTED]> wrote: > Dear friends, > > I am dealing with a particular relativization > issue in a microhabitat analysis and would like to > hear your opinion about it, would it be possible? > > Here follows a resume: > > Hypothesis: > > The Southern dominant conifer Araucaria angustifolia > occurs more often than not on higher > microtopographic sites, and avoids the lower ones. > > Data: > > XY positions of 600 young individuals in 10 separate > plots. Each plot has 1 ha (100 m x 100 m) and is > subdivided in 100 10 x 10 m subplots. The area is a > mountainous terrain at ca. 850 m elevation. > > Altitude has been measured on the corners of each > subplot, and at the locations of each individual. > > Analysis: > > Simply, the comparison of the background altitude > distribution of the subplot corners with the > individuals altitude distribution, through a > Kolmogorov-Smirnov test. > > The expectation is that the distribution of > individuals will be different from that of the local > microtopography because they lack individuals at the > lower altitudes > > The focus is not at the altitude in itself, but in > the lower parts of the plots, which correspond to > local depressions, which are more humid. > > Problem: > > Each one of the 10 areas is located at a distinct > altitude. Although these are not too much different > (amplitude ~100 m), it precludes me to pool the > altitude data of the areas, that form multimodal > distributions. > > My Current Doubt: > > I am thinking of considering the relative altitude > of each corner and each individual, instead of the > real altitude. This relativization would be carried > out by transforming each altitude datum in its > relative "distance" from the average altitude of its > plot (xi - averageX). > > I guess this would allow me to pool all altitude > measures and focus on what is the question of the > research. > > What do you think? Should I standardize the data, > dividing the relativized measures by the standard > deviation? Why? > > Any thoughts are well come. > > Sincerely, > > Alexandre > > Dr. Alexandre F. Souza > Programa de Pós-Graduação em Biologia: Diversidade e > Manejo da Vida Silvestre > Universidade do Vale do Rio dos Sinos (UNISINOS) > Av. UNISINOS 950 - C.P. 275, São Leopoldo 93022-000, > RS - Brasil > Telefone: (051)3590-8477 ramal 1263 > Skype: alexfadigas > [EMAIL PROTECTED] > http://www.unisinos.br/laboratorios/lecopop > > -- > Esta mensagem foi verificada pelo sistema de > antivírus e > acredita-se estar livre de perigo. > Abraham de Alba Avila Terrestrial Plant Ecology INIFAP-Ags Ap. postal 20, Pabellón Arteaga, 20660 Aguascalientes, MEXICO Tel: (465) 95-801-67, & 801-86 ext. 118, FAX ext 102 alternate: [EMAIL PROTECTED] cel: 449-157-7070 ___________________________________________________________________________________ You snooze, you lose. Get messages ASAP with AutoCheck in the all-new Yahoo! Mail Beta. http://advision.webevents.yahoo.com/mailbeta/newmail_html.html
