Key variable relationship model fitting is one of the basic statistical approaches to analyze the mechanism of biological and environmental interaction. According to the interaction mechanism between organisms and environment, the models constructed by fitting can be divided into two types. One is the influence model of environmental change on biological dynamics. For example, such influence model can be used to analyze the regulation mechanism of water nutrient changes on phytoplankton biomass dynamics. The second is the driving model of organisms to the environment. For example, such driving model can be used to analyze the driving mechanism of phytoplankton metabolic activities to the related substances in water environment and atmosphere environment. The innovation of statistical methods and tools and the increase of in-situ observation data of field environment make the fitting model of the relationship between organism and environment more realistic.

The change point test of the relationship between key variables is an important statistical step in quantitative analysis of the relationship between biology and environment. The uncertainty of model parameters caused by the difference of change point test methods has, to a large extent, brought doubts to the statistical inference of biological and environmental relationship models in ecology, environmental science and other disciplines, and even caused disputes in the analysis of interaction mechanism. In this study, the Institute of Urban And Environmental Studies, Chinese Academy of Sciences, Pennsylvania State University, Peking University, and the U.S. Geological Survey (USGS) analyzed the key parameters of phytoplankton biomass (Chlorophyll Chlorophyll A, CHL) and its metabolity-driven cold chamber gases (Dimethyl sulfide, DMS was used as case data of independent and dependent variables to construct boundary change-point model of cold chamber gas response to phytoplankton.

The boundary change point model of cold chamber gas response has two interrelated basic characteristics. First, the relationship between independent variables and dependent variables presents obvious upper boundary, which means that the gas substance in cold chamber has the maximum potential for phytoplankton biomass response. Second, there is a change point phenomenon in the relationship between independent variables and dependent variables, which further means that the maximum response of the gas matter in the cold chamber to the phytoplankton biomass is not a continuous linear increase, but when the phytoplankton biomass increases to a specific value, the gas matter in the cold chamber begins to decrease linearly. In view of the above two basic characteristics, the construction of cold chamber gas response boundary change-point model is different from traditional quantile regression in statistical methodology, but further coupled with Bayesian inference. Related model methods and results using Bayesian Change point quantile regression approach to enhance the understanding of rat behavior Phytoplankton -dimethyl sulfide relationships in Aquatic Ecosystems, published in Water Research.

The construction of the boundary change-point model of cold chamber gas response provides another statistical perspective for analyzing the formation mechanism of cold chamber gas driven by phytoplankton in lakes and oceans, or has corresponding methodological reference value for testing the change-point relationship of key variables in aquatic ecosystems or even other types of ecosystems. However, this does not mean that the model can perfectly explain the formation mechanism of cold chamber gases in this aquatic ecosystem. This is because quantile regression and Bayesian inference are “hungry” methods in terms of data demand, and available case data are still scarce at present. Driven by major research projects such as the mechanism of hydrosphere microorganisms driving the earth element cycle, the in-situ observation data of phytoplankton biomass and cold chamber gas material in the field are likely to be greatly increased. This will enable a more comprehensive “feeding” quantile regression and bayesian inference of the data starvation coupling method, and further more quantitatively explain the mechanism of phytoplankton metabolic activity driving the formation of cold chamber gases at multiple scales.

In another hypothesis testing path, the principal investigator of the boundary change point model explored the influence of the amount of data “fed” on the boundary stability of the quantile regression model. This hypothesis testing path found that an increase in the amount of data “fed” tended to make the key variable relationship model fitted using quantile regression more stable. Quantile regression method was established in the field of quantitative economics and developed and applied in the field of ecology. However, this method often needs a large amount of data “feeding” and cyclic calculation to fit the relationship model of key variables, which makes it difficult to play its due scientific value in the research and application of ecology and environmental science for a long period of time. With the outdoor environment of ecology and environmental sciences in situ observation data sharing the arrival of the era, quantile regression and its related law system, such as coupling with bayesian inference methods, key variables on the boundary are expected by the ecological and environment model fitting and related effects, the drive and response mechanism of analytic produce greater support.

Paper Link & NBSP;

Boundary change-point model of cold chamber gas response to phytoplankton

Key variable relationship model fitting is one of the basic statistical approaches to analyze the mechanism of biological and environmental interaction. According to the interaction mechanism between organisms and environment, the models constructed by fitting can be divided into two types. One is the influence model of environmental change on biological dynamics. For example, such influence model can be used to analyze the regulation mechanism of water nutrient changes on phytoplankton biomass dynamics. The second is the driving model of organisms to the environment. For example, such driving model can be used to analyze the driving mechanism of phytoplankton metabolic activities to the related substances in water environment and atmosphere environment. The innovation of statistical methods and tools and the increase of in-situ observation data of field environment make the fitting model of the relationship between organism and environment more realistic.

The change point test of the relationship between key variables is an important statistical step in quantitative analysis of the relationship between biology and environment. The uncertainty of model parameters caused by the difference of change point test methods has, to a large extent, brought doubts to the statistical inference of biological and environmental relationship models in ecology, environmental science and other disciplines, and even caused disputes in the analysis of interaction mechanism. In this study, the Institute of Urban And Environmental Studies, Chinese Academy of Sciences, Pennsylvania State University, Peking University, and the U.S. Geological Survey (USGS) analyzed the key parameters of phytoplankton biomass (Chlorophyll Chlorophyll A, CHL) and its metabolity-driven cold chamber gases (Dimethyl sulfide, DMS was used as case data of independent and dependent variables to construct boundary change-point model of cold chamber gas response to phytoplankton.

The boundary change point model of cold chamber gas response has two interrelated basic characteristics. First, the relationship between independent variables and dependent variables presents obvious upper boundary, which means that the gas substance in cold chamber has the maximum potential for phytoplankton biomass response. Second, there is a change point phenomenon in the relationship between independent variables and dependent variables, which further means that the maximum response of the gas matter in the cold chamber to the phytoplankton biomass is not a continuous linear increase, but when the phytoplankton biomass increases to a specific value, the gas matter in the cold chamber begins to decrease linearly. In view of the above two basic characteristics, the construction of cold chamber gas response boundary change-point model is different from traditional quantile regression in statistical methodology, but further coupled with Bayesian inference. Related model methods and results using Bayesian Change point quantile regression approach to enhance the understanding of rat behavior Phytoplankton -dimethyl sulfide relationships in Aquatic Ecosystems, published in Water Research.

The construction of the boundary change-point model of cold chamber gas response provides another statistical perspective for analyzing the formation mechanism of cold chamber gas driven by phytoplankton in lakes and oceans, or has corresponding methodological reference value for testing the change-point relationship of key variables in aquatic ecosystems or even other types of ecosystems. However, this does not mean that the model can perfectly explain the formation mechanism of cold chamber gases in this aquatic ecosystem. This is because quantile regression and Bayesian inference are “hungry” methods in terms of data demand, and available case data are still scarce at present. Driven by major research projects such as the mechanism of hydrosphere microorganisms driving the earth element cycle, the in-situ observation data of phytoplankton biomass and cold chamber gas material in the field are likely to be greatly increased. This will enable a more comprehensive “feeding” quantile regression and bayesian inference of the data starvation coupling method, and further more quantitatively explain the mechanism of phytoplankton metabolic activity driving the formation of cold chamber gases at multiple scales.

In another hypothesis testing path, the principal investigator of the boundary change point model explored the influence of the amount of data “fed” on the boundary stability of the quantile regression model. This hypothesis testing path found that an increase in the amount of data “fed” tended to make the key variable relationship model fitted using quantile regression more stable. Quantile regression method was established in the field of quantitative economics and developed and applied in the field of ecology. However, this method often needs a large amount of data “feeding” and cyclic calculation to fit the relationship model of key variables, which makes it difficult to play its due scientific value in the research and application of ecology and environmental science for a long period of time. With the outdoor environment of ecology and environmental sciences in situ observation data sharing the arrival of the era, quantile regression and its related law system, such as coupling with bayesian inference methods, key variables on the boundary are expected by the ecological and environment model fitting and related effects, the drive and response mechanism of analytic produce greater support.

Paper Link & NBSP;

Boundary change-point model of cold chamber gas response to phytoplankton