Simple item page
 

Ecological states of watercourses regarding water quality parameters and hydromorphological parameters: deriving empirical equations by machine learning models

cris.virtual.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtual.author-orcid0000-0002-6549-9418
cris.virtual.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtualsource.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
cris.virtualsource.author-orcidd16ffead-1aea-4e76-a330-73e0f4301bfb
cris.virtualsource.author-orcid#PLACEHOLDER_PARENT_METADATA_VALUE#
dc.abstract.enEnvironmental biomonitoring techniques have been widely applied to assess the quality states of toxic chemical compounds in surface freshwater quality. The methods based on macrophytes are generally recruited to present trustworthy assessments of ecological status for surface water bodies. The chief goal of the present investigation was to improve robust machine learning models (MLMs) to establish relationships among macrophyte indices and the simultaneous effects of water quality parameters (WQPs) and hydromorphological factors. In the present research, the dataset included monthly WQPs, which have been accumulated from 200 sample points located in the bottomland of studied rivers, placed in Poland. This research utilized three ecological indices, namely the macrophyte index for rivers (MIR), the macrophyte biological index for rivers (IBMR), and river macrophyte nutrient index (RMNI) whereas species richness (N) and Simpson index (D) were considered as diversity indices. 12 WQPs and two hydromorphological indices have been considered as input variables for the MLMs namely gene-expression programming (GEP), evolutionary polynomial regression (EPR), multivariate adaptive regression spline (MARS), and model tree (MT). In terms of the best performance of MLMs’ results, RMNI predictions were obtained by EPR (correlation coefficient [R] = 0.6061 and root mean square error [RMSE] = 0.5355) whereas MIR, IBMR, species richness, and Simpson index were predicted by MARS (R = 0.4333 and RMSE = 11.5046), EPR (R = 0.6020 and RMSE = 1.6923), GEP (R = 0.6089 and RMSE = 7.7229), and MARS (R = 0.5066 and RMSE = 0.2217), respectively. The great impact of these indexes was evaluated by the statistical parameters. This study showed that the biological assessment of rivers through macrophyte indexes not only helps to broaden knowledge related to the ecological status of the river but also aids in managing natural and anthropogenic activities’ impacts on the water body.
dc.affiliationWydział Inżynierii Środowiska i Inżynierii Mechanicznej
dc.affiliation.instituteKatedra Ekologii i Ochrony Środowiska
dc.contributor.authorNajafzadeh, Mohammad
dc.contributor.authorAhmadi-Rad, Elahe Sadat
dc.contributor.authorGebler, Daniel
dc.date.access2025-01-14
dc.date.accessioned2025-01-14T10:43:57Z
dc.date.available2025-01-14T10:43:57Z
dc.date.issued2024
dc.description.bibliographyil., bibliogr.
dc.description.financepublication_nocost
dc.description.financecost0,00
dc.description.if3,9
dc.description.number2
dc.description.points100
dc.description.volume38
dc.identifier.doi10.1007/s00477-023-02593-z
dc.identifier.eissn1436-3259
dc.identifier.issn1436-3240
dc.identifier.urihttps://sciencerep.up.poznan.pl/handle/item/2335
dc.languageen
dc.pbn.affiliationenvironmental engineering, mining and energy
dc.relation.ispartofStochastic Environmental Research and Risk Assessment
dc.relation.pages665-688
dc.sciencecloudnosend
dc.subject.enmachine learning models
dc.subject.enwater quality parameters
dc.subject.enmacrophyte indices
dc.subject.enhydromorphological indices
dc.titleEcological states of watercourses regarding water quality parameters and hydromorphological parameters: deriving empirical equations by machine learning models
dc.typeJournalArticle
dspace.entity.typePublication
oaire.citation.issue2
oaire.citation.volume38