The impact of varying spatiotemporal scales on different joint species distribution models: A case study of pelagic fish species in the northwest Pacific Ocean
Data files
Apr 28, 2025 version files 125.16 MB
-
Data_and_code.zip
125.16 MB
-
README.md
4.61 KB
Abstract
Joint Species Distribution Models (JSDMs) have become a critical tool in community ecology research, with a wide scope of application that is continuously expanding. However, inferring interspecies relationships from co-occurrence data remains a challenge. This study examined the impact of varying spatiotemporal scales on JSDMs, with a focus on model stability and the evaluation of interspecies relationships. We compared the performance of the models across 32 different spatiotemporal scales. Our results indicate that the spatiotemporal scale significantly affects the performance of JSDMs, with notable differences among the models. As both temporal and spatial scales increased, model simulation and prediction performance improved, and stability increased. Moreover, spatial scale has a substantial impact on the evaluation of interspecies relationships, with finer spatial scales identifying weaker positive relationships and stronger negative relationships. Among the models, HMSC demonstrated better balancing performance, while the Boral model showed the least stability. Overall, the optimal JSDM identified was the HMSC model with an annual temporal and 0.25° spatial scale.
========================================================================================
"model_code"
Data and commented code for the reproduction of moldeling. The distribution data of the three species is not publicly available due to privacy or ethical restrictions. All environmental data was provided in the ".xlsx" files.
variable definitions
Sea Surface Temperature (SST): This variable indicates the temperature of the ocean’s surface water. It is an essential factor for studying species distribution and marine ecosystems.
Chlorophyll-a Concentration (Chl-a): This variable represents the amount of chlorophyll-a pigment in the water, which is an indicator of phytoplankton biomass. Phytoplankton are microscopic plants that form the base of the marine food web, making this measurement crucial for understanding marine productivity and ecosystem health.
Sea Surface Salinity (SSS): This variable indicates the salt concentration in seawater at the surface. Salinity affects water density and circulation patterns in the ocean, influencing marine species' life and distribution.
Sea Level Anomaly (SLA): This variable represents variations from the average sea level in a given area. It can result from factors like ocean currents, temperature changes, and atmospheric pressure.
Eddy Kinetic Energy (EKE): This variable quantifies the energy in ocean eddies, which are swirling water masses. EKE is important for studying ocean mixing processes, nutrient distributions, and the dynamics of marine ecosystems.
Mixed Layer Depth (MLD): This variable refers to the depth of the upper layer of the ocean where water is well-mixed due to wind and wave action. The MLD influences the distribution of heat, nutrients, and marine organisms.
Dissolved Molecular Oxygen (O2): This variable indicates the concentration of oxygen dissolved in seawater. Oxygen is vital for the survival of marine organisms, making its measurement important for marine species distribution and ecosystem health.
Dissolved Molecular Nitrate (NO3): This variable represents the concentration of nitrate ions in seawater. Nitrate is a key nutrient for phytoplankton growth, and monitoring its levels helps understand primary production and nutrient cycling in marine ecosystems.
Dissolved Molecular Phosphate (PO4): This variable indicates the concentration of phosphate ions in seawater. Similar to nitrate, phosphate is an essential nutrient for marine plants, and its levels influence primary productivity.
Net Primary Production (NPP): This variable signifies the amount of organic matter produced by phytoplankton through photosynthesis minus the amount consumed through respiration. NPP is a critical measure of the productivity of marine ecosystems and overall ocean health.
Units for all variables
SST: Measured in degrees Celsius—°C
Chl-a: Measured in milligrams per cubic meter —mg/m^3
SSS: Measured in practical salinity units—psu
SLA: Measured in meters—m
EKE: Measured in square meters per second squared—m2/s2
MLD: Measured in meters —m
O2: Measured in millimoles per cubic meter—mmol/m^3
NO3: Measured in millimoles per cubic meter—mmol/m^3
PO4: Measured in millimoles per cubic meter—mmol/m^3
NPP: Measured in milligrams per cubic meter per day—mg/m^3day
All files details
Y"2×2data.xlsx
The data for the time scale is in years, and the spatial scale is 2°.
Y1×1data.xlsx
The data for the time scale is in years, and the spatial scale is 1°.
Y0.5×0.5data.xlsx
The data for the time scale is in years, and the spatial scale is 0.5°.
Y0.25×0.25data.xlsx
The data for the time scale is in years, and the spatial scale is 0.25°.
M2×2data.xlsx
The data for the time scale is in months, and the spatial scale is 2°.
M1×1data.xlsx
The data for the time scale is in months, and the spatial scale is 1°.
M0.5×0.5data.xlsx
The data for the time scale is in months, and the spatial scale is 0.5°.
M0.25×0.25data.xlsx
The data for the time scale is in months, and the spatial scale is 0.25°.
Software versions for models
Bayesian Community Ecology Analysis: ‘BayesComm’ version 0.1-2
Generalized Joint Attribute Modeling: ‘gjam’ version 2.6.2
Bayesian Ordination and Regression Analysis: ‘boral’ version 2.0
Hierarchical Model of Species Communities: ‘Hmsc’ version 3.0-13
Software version for R
the script was created using R version 4.2.0.
To comprehensively evaluate the impact of varying spatiotemporal scales on JSDMs, this study was designed using two temporal scales (monthly and annual), four spatial scales (0.25°, 0.5°, 1°, and 2°), and four different JSDMs (Bayescomm, HMSC, Boral, and Gjam). Using three economically important pelagic fish species from the northwest Pacific Ocean as examples—Japanese sardine (Sardinops melanostictus), chub mackerel (Scomber japonicus), and neon flying squid (Ommastrephes bartramii)—we compared the performance of the models across 32 different spatiotemporal scales.