Selection of indicative species for evaluation of ecological state: Case study on benthic macroinvertebrates focused on Random forest method

KUBOŠOVÁ, Klára, Jiří JARKOVSKÝ, Karel BRABEC and Světlana ZAHRÁDKOVÁ. Selection of indicative species for evaluation of ecological state: Case study on benthic macroinvertebrates focused on Random forest method. In Program and Abstracts, The Seventeenth International Conference on Qualitative Methods for the Environmental Sciences, TIES 2007. 2007. ISBN 978-80-210-4333-6.

Basic information

• Original name: Selection of indicative species for evaluation of ecological state: Case study on benthic macroinvertebrates focused on Random forest method
• Name in Czech: Výběr indikačních taxonů pro hodnocení ekologického stavu pomocí metody Random Forests
• Authors: KUBOŠOVÁ, Klára (203 Czech Republic, guarantor, belonging to the institution), Jiří JARKOVSKÝ (203 Czech Republic, belonging to the institution), Karel BRABEC (203 Czech Republic, belonging to the institution) and Světlana ZAHRÁDKOVÁ (203 Czech Republic, belonging to the institution).
• Edition: Program and Abstracts, The Seventeenth International Conference on Qualitative Methods for the Environmental Sciences, TIES 2007, 2007.

Other information

• Original language: English
• Type of outcome: Conference abstract
• Field of Study: 10600 1.6 Biological sciences
• Country of publisher: Czech Republic
• Confidentiality degree: is not subject to a state or trade secret
• RIV identification code: RIV/00216224:14310/07:00023788
• Organization unit: Faculty of Science
• ISBN: 978-80-210-4333-6
• Keywords in English: indicative species; random forest; indicator species analysis; logistic regression
• Tags: indicative species, indicator species analysis, logistic regression, random forest
• Tags: International impact, Reviewed

Abstract

Definition of species with significant relationship to environmental parameters is indispensable part of systems for the evaluation of environmental state. Nevertheless this is not considered as easy task and several methods with various advantages and disadvantages are available. In this paper two main methods were applied for identification of indicative species of river habitats; relatively new method Random Forest and classical Indicator Species Analysis; the results were supplemented also by logistic regression of species response curves which can produce additional information on taxa tolerance and optimum. The basic advantages and disadvantages of methods can be summarized according to our experience from this case study; these conclusions can be utilized during the development of the ARROW system for evaluation of ecological state of surface waters. The results of the study also support simultaneous usage of at least two different methods for indicative species identification to reduce known weak points of individual methods.

Abstract (in Czech)

Identifikace indikačního potenciálu taxonů je velmi důležitou součástí všech modelů pro hodnocení ekologického stavu. Na správném výběru závisí kvalita detekce znečištění dané lokality či prostředí. Poměrně nová metoda Random Forests byla srovnávána s metodou indikačních druhů (Indicator Species Analysis). Výsledky byly doplněny o valenční křivky určující optima a toleranci jednotlivých taxonů.

Links

• MSM0021622412, plan (intention): Name: Interakce mezi chemickými látkami, prostředím a biologickými systémy a jejich důsledky na globální, regionální a lokální úrovni (INCHEMBIOL) (Acronym: INCHEMBIOL)

Investor: Ministry of Education, Youth and Sports of the CR, Interactions among the chemicals, environment and biological systems and their consequences on the global, regional and local scales (INCHEMBIOL)