Effects of an invasive top predator on Ecosystem structure and function in a Graminoid Marsh food web
Data files
May 08, 2025 version files 381.01 KB
-
compartment_mass.csv
10.40 KB
-
CYPDRY.xlsx
17.01 KB
-
GRAMDRY.xlsx
20.05 KB
-
MANGDRY.xlsx
25.77 KB
-
nodes.csv
2.44 KB
-
properties.csv
36 B
-
pythondiet.csv
298.87 KB
-
README.md
6.42 KB
May 20, 2025 version files 312.34 KB
-
compartment_mass.csv
10.40 KB
-
CYPDRY.xlsx
17.01 KB
-
GRAMDRY.xlsx
20.05 KB
-
MANGDRY.xlsx
25.77 KB
-
nodes.csv
2.44 KB
-
properties.csv
36 B
-
pythondiet2.csv
230.28 KB
-
README.md
6.34 KB
Abstract
In this era of global change, understanding the effects of human-mediated dispersal of organisms has become a priority for ecological and conservation research. Non-native species introductions can result in biological invasions that have substantial impacts on native population abundances, community interactions, and ecosystem processes. Over the past twenty years, the establishment and subsequent invasion of the Burmese python (Python molurus bivittatus) in the Greater Everglades ecosystem has become a topic of concern amongst land managers and conservation practitioners in southern Florida. The objective of this study is to assess community-wide impacts of the python on the native food web as well as the function of the Greater Everglades Ecosystem by attempting to answer two central questions: 1) What functional role does the python occupy within the native food web and 2) is there a shift in overall food web structure and/or function post-invasion? We used ecological network analysis and an extensive diet dataset to quantify the python’s trophic role relative to other residents in the food web as well as compare ecosystem characteristics between pre- and post-invaded network models. Our findings demonstrate that the python functions similarly to the largemouth bass, another torrid, highly invasive predator that exhibits strong top-down impacts within its aquatic habitats. The python also behaves as a dominant predator akin to the Florida panther, primarily affecting native mammal populations through top-down predation effects, displacing other top predators within the food web, and altering patterns of carbon flow along the food chain. Finally, at the system scale, although we see increasing trends in ecosystem activity and decreasing trends in ecosystem structure and organization, functional metrics remain relatively stable pre- and post- invasion indicating functional resilience. Our findings provide a holistic assessment of the Burmese python invasion on the native community and function of the greater Everglades graminoid marsh ecosystem.
Dataset DOI: 10.5061/dryad.tqjq2bw9k
Description of the data and file structure
Python diet data was collected through gut content analysis of individual Burmese pythons collected across its invasive range across South Florida. Remaining data was acquired from the existing literature. The three food web networks were acquired from work conducted by Dr. Ulanowicz and his colleagues. Finally, biomass estimates were acquired from various existing sources which have been cited in respective file.
Files and variables
File: compartment_mass.csv
Description: Dataset of the average mass of each species compartment used in the food web networks that was found in the python's diet
Variables
- Species: Name of each species compartment used in the food web network
- avg.mass: Average mass in grams per individual of each species compartment
- Full.Reference: Reference where mass value was acquired
File: CYPDRY.xlsx
Description: A 68-component budget of the carbon exchanges measured in gC/m2/yr occurring during the dry season in the cypress wetlands of South Florida has been assembled. Formatted to be read in as a network class required for the enaR package.
Variables
- An n×n flow matrix oriented row-to-column
- a vector of inputs
- a vector of respirations
- a vector of exports
- a vector of outputs, which are respirations plus exports
- a vector of biomass or storage values
- Living = logical vector indicating if the node is living (TRUE) or non-living (FALSE). Indicated in the formatted data by the number of compartments and number of living compartments, where the n position in the flow matrix will be assigned living for n = 1: number of living compartments and the remaining nodes will be assigned non-living. For this reason, it is critical to have all non-living nodes designated at the tail end of the nxn flow matrix.
File: GRAMDRY.xlsx
Description: A 66-component budget of the carbon exchanges measured in gC/m2/yr occurring during the dry season in the graminoid ecosystem of South Florida has been assembled. Formatted to be read in as a network class required for the enaR package.
Components
- An n×n flow matrix oriented row-to-column
- a vector of inputs
- a vector of respirations
- a vector of exports
- a vector of outputs, which are respirations plus exports
- a vector of biomass or storage values
- Living = logical vector indicating if the node is living (TRUE) or non-living (FALSE). Indicated in the formatted data by the number of compartments and number of living compartments, where the n position in the flow matrix will be assigned living for n = 1: number of living compartments and the remaining nodes will be assigned non-living. For this reason, it is critical to have all non-living nodes designated at the tail end of the nxn flow matrix.
File: nodes.csv
Description: Csv that contains the list of nodes within the GRAMDRY network with the associated properties of each node's effective trophic level and mean body mass formatted to load as a community object and conduct network analyses from the cheddar package in R.
Variables
- node: species compartment name
- category: species categorized as producer, invertebrate, ectothermic vertebrate, or endothermic vertebrate. Those that are blank are either non-living or microbial compartments
- ETL: effective trophic level derived from ecological network analysis of the GRAMDRY network from the enaR package
- M: estimated average biomass (gC/m2/yr)
File: properties.csv
Description: Csv that defines items applicable to the community as a whole, must have column named title, formatted to load as a community object and conduct network analyses from the cheddar package in R.
Variables
- title: Graminoid system
- M.units: measurement unit of carbon exchange
File: MANGDRY.xlsx
Description: A 94-component budget of the carbon exchanges measured in gC/m2/yr occurring during the dry season in the mangrove ecosystem of South Florida has been assembled. Formatted to be read in as a network class required for the enaR package.
Variables
- An n×n flow matrix oriented row-to-column
- a vector of inputs
- a vector of respirations
- a vector of exports
- a vector of outputs, which are respirations plus exports
- a vector of biomass or storage values
- Living = logical vector indicating if the node is living (TRUE) or non-living (FALSE). Indicated in the formatted data by the number of compartments and number of living compartments, where the n position in the flow matrix will be assigned living for n = 1: number of living compartments and the remaining nodes will be assigned non-living. For this reason, it is critical to have all non-living nodes designated at the tail end of the nxn flow matrix.
File: pythondiet2.csv
Description: Diet data of the Burmese python (Python bivittatus) collected from gut content analysis of pythons collected within its invaded range of South Florida.
Variables
- ID: Identifier of each individual python
- Class: taxonomic class of prey item
- Order: taxonomic order of prey item
- Family: taxonomic family of prey item
- Genus: taxonomic genus of prey item
- Species: taxonomic species of prey item
- SciName: scientific name of the prey item based on its binomial nomenclature
- CommonName: common name of the prey item
- PreyGroup: prey item grouped as categories of mammal, bird, or reptile and as unidentifiable or other withen these groupings.
- PreyMass: estimated average mass in grams of prey item
Code/software
https://github.com/sleclare/ena-BurmesePython-invasion
Access information
This dataset, as well as other information necessary to reproduce the results in the associated publication, are available on GitHub. We recommend that the dataset is accessed through the following link:
https://github.com/sleclare/ena-BurmesePython-invasion
Version changes
20-May-2025: Updated pythondiet2.csv to retain only the columns used in analyses
