FLORA: A Software Package for Statistical Analysis of Ecological Data

 Branko Karadžić¹

 

 

 

¹ Institute for Biological Research „Siniša Stanković”, University of Belgrade, Bulevar Despota Stefana 142,  Belgrade, Serbia, E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

 

Abstract

The computer software package FLORA (for Microsoft® Windows®) was developed as a tool for statistical data processing. The package may be used in different scientific disciplines (taxonomy, evolutionary research, geology, soil science, social sciences such as economy, psychology, sociology etc.). Despite such generality, the FLORA was primarily designed for use in ecology related studies. The package enables the storage, selection and analysis of ecological data, and creation of concise reports in forms of graphical outputs and summary tables of statistical analyses.

Keywords: Numerical classification, Ecological data, Multivariate analyses, Ordination, Canonical analyses, Permutation tests.

 

Introduction

 

Complex processing of ecological data is necessary for understanding structure and functioning of ecosystems, for the conservation of biodiversity and for ecosystem management. Ecological studies that involve inventory, monitoring, modelling and predicting of changes in populations, ecological communities and ecosystems require both georeferenced databases and computational tools for application of statistical methods. Mathematical basis of uni- and multi-variate statistical methods, that are used in ecological studies is described in numerous  monographs (Pielou, 1969, 1975, 1984;  Poole 1974;  Orloci, 1978;  Gauch ,1982;  Greig-Smith, 1983; Legendre &  Legendre, 1983; Digby & Kempton, 1987;  Jongman, ter Braak & van Tongeren, 1987; Legendre & Legendre 1998; Karadžić and Marinković, 2009; Greenacre 2010).

However, implementation of these methods is impossible without powerful computational tools. Rapid development of information technologies resulted with proliferation of software packages (Orloci and Wildi, 1971; Hill, 1979, 1979a; ter Braak, 1987, 1995, McCune & Mefford, 1995, Hennekens 1995; Podani, 1997; Bruelheide 2000; ter Braak & Šmilauer 2002; Tichy & Holt 2006; Oksanen et al., 2012). These software packages differ significantly with respect to analytic abilities and flexibility in data manipulation (i.e. data editing and data exchange with other data banks).

The FLORA package integrates abilities of all existing packages, but also offers some general purpose routines that enable application of both uni- and multi-variate analyses in ecological research. Of course, due to its great flexibility, the package may be used in other scientific disciplines (i.e. taxonomy, evolutionary studies and social sciences such as economy, sociology, psychology, etc).

 

General properties of the package

FLORA is a software application for storage, retrieval, editing and analysing large ecological data sets. The package can be run on all Microsoft® Windows® operating systems. The minimum system requirement are a 486-computer with 16 Mb RAM. The newest version of the package is the culmination of a programming project running continuously since 1999 (Karadžić et al., 1999). The package is written in Visual Basic 6 language. It does not include neither ActiveX controls, nor object linking and embedding containers, produced by other authors. Instead of that, it represents an integrated system of subroutines and functions that are created by the author of this article. Any subroutine or function is available free, upon request.

 

Data input, import and export

Data retrieval and storage are facilitated by numerous commands. These commands are grouped within the "File" menu. There are three available methods for running a new data set: input of separate relevés or tables 'by hand', import of data by opening stored files and using the "copy-paste" function which enables fast and efficient communication of FLORA with any Microsoft ® format (i. e. MS-DOS text (*.txt), Rich text (*.rtf) and document (*.doc) formats for Microsoft® WORD®, spreadsheet format (*.xsl) for Microsoft® EXCEL® or database format (*.mdb) for Microsoft® ACCESS® ).

The option "New" displays a dialog box that specifies dimensions of new data matrix. The dialog box asks for the number of species and sites in the new file. Users may alter these parameters at any time by using the Edit/Columns (Rows)/Add (Delete) options. New parameters of the data matrix are automatically updated.

The command "Open" offers two submenus: Files and Graphs. The Files submenu displays a dialog box for searching stored files. Flora is able to open files written in ASCII formats such as MS-DOS text (*.txt) files and coma-separated values (*.csv) files.

The Graphs submenu offers opening pictures that are stored in different raster or vector formats. Considering raster formats, FLORA is able to open bitmaps (files with extension ".bmp") and JPEGs (files with extension ".jpg"). The vector formats recognizable by FLORA are Windows Enhanced Meta Files (".emf).

The command "Close" closes the active window with its documents. The command "Save As" displays a dialog box for saving either pictures or data files with specified file name and extension for specific file types.

The command "Print Preview" displays a toolbar, which shows the results of a project file without printing it. The command Print (Keyboard shortcut: CTRL+P) displays a dialog box for printing the current file, report, or graph.

The command Exit (Keyboard shortcut: ALT+F4) ends your FLORA session and returns control to the operating system.

Ecological data are usually organized in a matrix form. FLORA operates using two forms of data matrices (the 'response' data matrix X(nxm), which describes the distribution of n species in m sites, and the 'explanatory' matrix Z(qxm), which contains values of q environmental variables in m sites). Both matrices are displayed on the screen as scrollable tables (Fig 1).

Within the explanatory matrix Z, users can select on or more particular variables for data analysis. Moreover, we may divide variables into two different groups. The first group is called the environmental variables and refers to the variables which are the primary subject of interest in our particular analysis. The other group represents the so-called covariables (often referred to as covariates). The covariables are associated variables whose effects on both "response" matrix and a set of selected environmental variables should be removed.

 

 

Fig01

Figure 1: FLORA displays two data matrices. The ‘response’ data matrix describes the distribution of an n species in m sites. The ‘explanatory’ matrix (shaded table) contains values of q environmental variables in m sites