Influences Of Environmental Factors On Macroinvertebrate Community Structure And Function Diversity Characteristics Between Mountain And Lowland Ecoregions,Wei River

The southern Wei River watershed located across Qingba mountain evergreen-deciduous broad-leaved forest ecoregion(i.e.,mountain ecoregion)and Fenwei river valley agricultural ecoregion(i.e.,agricultural ecoregion).In the past 30 years,the condition of streams and the diversity of stream macroinvertebrate decreased in both ecoregions because of the increase of intensed human activities such as land use changes and point and non-point pollutions.In recent,researches on the comparison of the effects of environmental variables at different spatial scales on community structure and functional diversity between ecoregions are rare in China.Using data collected from southern Wei River watershed,the research aimed to(1)compare the physicochemical environmental variables,community structure,biological traits and trait and functional diversity indices of macroinvertebrate between mountain and lowland ecoregions,and(2)explore the main environmental variables influencing macroinvertebrate community structure,biological traits and trait and functional diversity indices.I expect that the results would provide new insights on and contribute to Chinese stream condition.The innovative points of this research are:1)I compared the effects of environmental variables at different spatial scales on macroinvertebrate community structure and functional diversity,and 2)I identified the dominant environmental variables affected the macroinvertebrate functional diversity indices at each mountain and lowland ecoregion in Wei River watershed.The objectives of this study were to:1)describe and compare the difference of land use variables,water physicochemical characteristics and biological habitat variables between mountain and lowland ecoregions,2)compare the most commonly used macroinvertebrate community structure metrics,composition metrics of biological traits,and trait and functional diversity indices between mountain and lowland ecoregions,and 3)identify key environmental variables affected macroinvertebrate community structure,composition of biological traits and trait and functional diversity indices in each ecoregion by applying generalized linear model(GLM)and RLQ(R:environment-site table,L:species-site table,Q:species-traits table)and fourth-corner statistical models.Our results found that 21 physicochemical variables were significantly different(P<0.05)between mountain and lowland ecoregions.Streams in mountain ecoregion were higher elevation,dissolved oxygen,max current velocity,and the percentage of cobble and boulder.In contrast,streams in lowland ecoregion had higher water temperature,calcium,magnesium,the percentage of silt,total nitrogen(TN),nitrite nitrogen,ammonia nitrogen,total phosphorus(TP),phosphate and silicon dioxide concentrations.The percentage of forest at reach and catchment scales in mountain was higher than lowland ecoregion.Whereas,the percentage of agriculture and urban land use were higher in lowland ecoregion.The results showed that higher degrees of physical habitats and water chemical degradation in lowland than in mountain ecoregion.We compared macroinvertebrate community structure metrics,traits composition and traits and functional diversity indices between mountain and lowland ecoregions.The 5 macroinvertebrate community structure metrics are taxa richness,Ephemeroptera,Plecoptera and Trichoptera(EPT)taxa,percentages of EPT,intolerant individuals,and Shannon-Wiener diversity index.The 29 trait compositions belonged to 8 trait categories including refuge,exoskeleton,respiration,body size,body shape,rheophily,habit,and functional feeding groups.The 8 biological traits and functional indices are trait richness,trait diversity,functional richness,functional evenness,functional dispersion,functional divergence,and Rao's Quadratic Entropy.The results showed that macroinvertebrate community structure,biological traits and trait and functional diversity metrics significantly differed between mountain and lowland ecoregions.Mountain streams had two times greater total taxa richness than lowland streams.All 5 community structure metrics were significantly difference between ecoregions.18 of 29 trait compositions and 6 of 7 trait and functional diversity indices were significantly different between ecoregions.Mountain ecoregions were dominant by macroinvertebrates with traits of none and shelters of leaf parts or wood refuge,lightly sclerotized,well protected exoskeleton or external protection,herbivores,and predators.In contrast,macroinvertebrates with traits of integumentary respiration,soft body,and collector-gatherers tended to frequently appearence in lowland ecoregion.All traits and functional diversity indices except functional evenness were significantly higher in mountain than in the lowland ecoregion,which demonstrated that environmental stressors significantly decreased the biological traits and functional diversity in lowland area.We identified key environmental viriables affecting macroinvertebrate community structure,biological trait composition and trait and functional diversity indices by using the combination of RLQ and fourth-corner methods and GLM models.RLQ and fourth-corner methods results revealed body shape was significantly positive related to environmental variables representing serious human disturbance(the percentage of silt,TP and the percentage of catchmental urban land use etc.).Refuge and body size were significantly positive related to TP.Exoskeleton or external protection,respiration,rheophily and functional feeding groups were significantly positive related to reach and catchment forest land use,but were significantly negative related to water temperature,the percentage of sediment,TP and the percentage of catchment urban land use which represented human disturbance gradient of environmental variables.The GLMs results demonstrated that land uses are key drivers affected physical habitat in the mountain ecoregion,and are main drivers affected both physical and chemical conditions in the lowland ecoregion.Macroinvertebrates assemblage composition metrics are predominantly influenced by land uses and physical habitat in the mountain ecoregion,but strongly influenced by land use,physical habitat,and chemical variables in the lowland ecoregion.The GLM revealed that catchment urban land uses,water temperature,and sediment concentration explained most variations within the biological traits and functional diversity in the mountain ecoregion;and reach scale agriculture land use,TN,silicon dioxide,average bed width,and Calcium mainly decreased the functional diversity in the low land ecoregion.We concluded that environmental stressors at both catchment and reach scales significantly affected functional diversity of biological assemblages in the mountain ecoregion;in contrast,reach scale environmental stressors were the main reasons decreased functional diversity of macroinvertebrates in the lowland ecoregion.Our results provide new insights on how key environmental drivers constrain physicochemical habitat and biological communities among ecoregions with different natural conditions.Variations within macroinvertebrate community structure metrics,biological traits and functional diversity indices explained by environmental variables at different spatial scales were different between ecoregions.Both macro invertebrate assemblage and functional composition metrics could be used in water quality assessment in the natural habitat preservation good environment.While the macroinvertebrate assemblage composition metrics are more suitable for water quality assessment at severely disturbed streams.