Non-linear and non-stationary relationships in the pelagic ecosystem of the Gulf of Riga (Baltic Sea)
Kuupäev
2021-12-20
Autorid
Ajakirja pealkiri
Ajakirja ISSN
Köite pealkiri
Kirjastaja
Abstrakt
Suhted toiduahelas või liikide ja keskkonna vahel võivad olla nii lihtsad (lineaarsed) kui keerukamad (mitte-lineaarsed), ning nii seose tüüp kui ka selle tugevus ja suund võivad ajas muutuda. Kuigi teadmine sellest, et pika-ajalised seosed mere ökosüsteemides on pigem harva stabiilsed ja lineaarsed, kasutatakse tihtipeale jätkuvalt mudeleid, mis ei suuda keerukamaid seosed tuvastada. See kehtib ka Läänemere kohta, kus keerukad ja ajas muutuvaid suhteid tuvastavate meetodite kasutamine on hoogustunud alles viimasel kümnendil. Töös kasutan ligi kuuekümne aastaseid (1958-2018) Liivi lahe andmeridu, millest mitmed on uudsed ja seega ei ole varasemalt analüüsidesse kaasatud. Uuritud ajaperioodil on lahes toimunud mitmeid väga olulisi muutusi, nii järk-järgulisi kui ka hüppelisi, mida nimetatakse ka režiiminiheteks. Käesolevas töös uurin: i) suure järvehormiku (aerjalaline, Limnocalanus macrurus) pikaajalise arvukuse muutuse põhjuseid; ii) invasiivse võõrvesikirbulise Cercopagis pengoi mõju ta saakloomade arvukusele ning saaklooma-keskkonna seoseid enne ja pärast võõrliigi invasiooni; iii) kohaliku kevadkuduräime (Clupea harengus membras) täiendi arvukust mõjutavaid tegureid. Andmeanalüüsis kasutati variatsioone ning kombinatsioone erinevatest statistilistest meetoditest, et kirjeldada ajas muutuvaid seoseid ja tuvastada põhjuslikke seoseid uuritavate muutujate vahel, näiteks üldistatud aditiivseid mudeleid, ning kahte erinevat “libiseva ajaakna” meetodit. Liivi lahe pelaagilise ökosüsteemi dünaamikat iseloomustavad nii lineaarsed kui ka mitte-lineaarsed, kuid ajas muutuvaid seosed. Mitmed tulemused on uudsed ja lükkavad ümber mõned seni teada seosed, mida põhjendan komplekse ja kohati uudse analüüsimetoodika rakendamisega. Töö üldine järeldus on, et Liivi lahe pelaagilise ökosüsteemi dünaamikat iseloomustavad nii lineaarsed kui ka mitte-lineaarsed, kuid ajas muutuvaid seosed. Üheks olulisimaks uueks teadmiseks on Liivi lahe räime täiendit mõjutavate tegurite omavaheliste suhete komplekssus kusjuures seni arvatud kliima mõju on suhteliselt marginaalne.
This thesis focuses on non-linear and non-stationary relationships in the Gulf of Riga (Baltic Sea) pelagic ecosystem. The relationships in the food webs or species-environment relationships may be linear or non-linear, while the type, strength and direction can also change over time (i.e. be non-stationary). Although the comprehension that long-term relationships in marine ecosystems are rarely stable and linear is widespread, still quite often the statistical methods that are used to identify these links, i.e. cannot detect non-linearity and non-stationarity. This is also true for the Baltic Sea, where the application of more advanced statistical methods and reporting of non-linear and non-stationary results has grown only over the last decade. The long-term environmental and biological monitoring data from the Gulf of Riga used in this thesis span almost 60 years (1958-2018), and include some previously unavailable data. Over this period, the Gulf of Riga ecosystem has undergone gradual as well as abrupt changes (so-called regime shifts). Publications of this thesis include three case studies: i) long term abundance dynamics of the arctic copepod Limnocalanus macrurus identification of the likely drivers of change, ii) changes in the abundance dynamics and alteration of the prey-environment relationships of the primary prey taxa of the non-indigenous predator cladoceran Cercopagis pengoi after its invasion, and iii) factors affecting the recruitment dynamics of the local spring spawning herring (Clupea harengus membras) population. In the statistical analyses, a suite of the analytical methods was used including the generalized additive models and two different “moving window” methods. Many of the results in this thesis provided new and previously unknown insights to the functioning of the Gulf of Riga ecosystem.
This thesis focuses on non-linear and non-stationary relationships in the Gulf of Riga (Baltic Sea) pelagic ecosystem. The relationships in the food webs or species-environment relationships may be linear or non-linear, while the type, strength and direction can also change over time (i.e. be non-stationary). Although the comprehension that long-term relationships in marine ecosystems are rarely stable and linear is widespread, still quite often the statistical methods that are used to identify these links, i.e. cannot detect non-linearity and non-stationarity. This is also true for the Baltic Sea, where the application of more advanced statistical methods and reporting of non-linear and non-stationary results has grown only over the last decade. The long-term environmental and biological monitoring data from the Gulf of Riga used in this thesis span almost 60 years (1958-2018), and include some previously unavailable data. Over this period, the Gulf of Riga ecosystem has undergone gradual as well as abrupt changes (so-called regime shifts). Publications of this thesis include three case studies: i) long term abundance dynamics of the arctic copepod Limnocalanus macrurus identification of the likely drivers of change, ii) changes in the abundance dynamics and alteration of the prey-environment relationships of the primary prey taxa of the non-indigenous predator cladoceran Cercopagis pengoi after its invasion, and iii) factors affecting the recruitment dynamics of the local spring spawning herring (Clupea harengus membras) population. In the statistical analyses, a suite of the analytical methods was used including the generalized additive models and two different “moving window” methods. Many of the results in this thesis provided new and previously unknown insights to the functioning of the Gulf of Riga ecosystem.
Kirjeldus
Väitekirja elektrooniline versioon ei sisalda publikatsioone
Märksõnad
aquatic ecosystems, pelagic nutrient chains, community dynamics, marine ecology, data analysis, Baltic Sea, Gulf of Riga