Data from: Combining mesocosms with models to unravel the effects of global warming and ocean acidification on a temperate marine ecosystem
Data files
Feb 29, 2024 version files 158.55 MB
Abstract
Ocean warming and species exploitation have already caused large-scale reorganization of biological communities across the world. Accurate projections of future biodiversity change require a comprehensive understanding of how entire communities respond to global change. We combined a time-dynamic integrated food web modelling approach (Ecosim) with previous data from community-level mesocosm experiments to determine the independent and combined effects of ocean warming and acidification, and fisheries exploitation, on a well-managed temperate coastal ecosystem. The mesocosm parameters enabled important physiological and behavioural responses to climate stressors to be projected for trophic levels ranging from primary producers to top predators, including sharks. Through model simulations, we show that under sustainable rates of exploitation, near-future warming or ocean acidification in isolation could benefit species biomass at higher trophic levels (e.g., mammals, birds, and demersal finfish) in their current climate ranges, with the exception of small pelagic fish. However, under warming and acidification combined biomass-increases at higher trophic levels will be lower or absent, whilst in the longer term reduced productivity of prey species is unlikely to support the increased biomass at the top of the food web. We also show that increases in exploitation will suppress any positive effects of human-driven climate change, causing individual species biomass to decrease at higher trophic levels. Nevertheless, total future potential biomass of some fisheries species in temperate areas might remain high, particularly under acidification, because unharvested opportunistic species will likely benefit from decreased competition and show an increase in biomass. Ecological indicators of species composition such as the Shannon diversity index declined under all climate change scenarios, suggesting a trade-off between biomass gain and functional diversity. By coupling parameters from multi-level mesocosm food web experiments with dynamic food web models, we were able to simulate the generative mechanisms that drive complex responses of temperate marine ecosystems to global change. This approach, which blends theory with experimental data, provides new prospects for forecasting climate-driven biodiversity change and its effects on ecosystem processes.
README: Data from: Combining mesocosms with models to unravel the effects of global warming and ocean acidification on a temperate marine ecosystem
This README file was generated on 2024-02-24 by Hadayet Ullah.
GENERAL INFORMATION
Title of Dataset: Combining Mesocosms and Models to Unravel Global Warming and Ocean Acidification Effects on a Temperate Marine Ecosystem. The dataset is organized into Excel files. Each file serves a distinct role in supporting the article's content.
Author Information:
A. First Author Contact information: Hadayet Ullah
Southern Seas Ecology Laboratories, School of Biological Sciences, University of Adelaide, Adelaide, Australia
B. Corresponding Author Contact information: Ivan Nagelkerken
Southern Seas Ecology Laboratories, School of Biological Sciences, University of Adelaide, Adelaide, Australia
Date of data collection (approximate date): 2017-2018; Data type: Both primary (mesocosm experiment) and secondary
Geographic location of primary data collection: South Australian Research & Development Institute (SARDI), St Vincent and Spencer gulf, South Australia
Information about funding sources that supported the collection of the data: Australian Research Council Future Fellowship Grant to I.N. (FT120100183) and International Postgraduate Research Scholarship (IPRS) to Ullah, H.
SHARING/ACCESS INFORMATION
Licenses/restrictions placed on the data: CC0 1.0 Universal (CC0 1.0) Public Domain
Links to publications that cite or use the data:
Ullah, H., Fordham, D. A., Goldenberg, S.U., Nagelkerken, I. (2024). Combining mesocosms with models reveals effect of global warming and ocean acidification on a temperate marine ecosystem. Ecological Applications.
Links to other publicly accessible locations of the data: None
Links/relationships to ancillary data sets: None
Recommended citation for this dataset:
Ullah, H., Fordham, D. A., Goldenberg, S.U., Nagelkerken, I. (2024). Data from: Combining Mesocosms and Models to Unravel Global Warming and Ocean Acidification Effects on a Temperate Marine Ecosystem. Dryad Digital Repository. DOI:10.5061/dryad.jwstqjqgk
6.Supplemental files included in this manuscript is compatible with the CC0 license waiver required by Dryad for publication and are not included within the related manuscript.
DATA & FILE OVERVIEW
File Naming Conventions
The dataset files are named descriptively to facilitate easy identification by readers. Specific naming conventions have been followed in accordance with the guidelines provided by the journal Ecological Applications. An initial "M" indicates data supporting the main text, while an "A" is used for data supporting appendix files.
Excel Files for Main Article
Four Excel files, each starting with an initial "M," contain the data used to create the figures in the main body of the article. Each file includes a legend that clarifies the content and specifies the figure generated using that particular dataset.
Supplementary Data Files
Additional Excel files, starting with an initial "A," are provided to support the figures and tables in the article's supplementary materials. These files contain data that have been both generated and analyzed to produce the supplementary figures and tables.
Reviewer File
An additional Excel file, named "Diet_PPB Model," has been included for review purposes. This file contains unique data essential for the review process
File List:
A. Ecosystem model skill assesment_TestData
A. Effect sizes and direction of impacts on biomass
A. Global sensitivity analysis_OAT
A. Higher TL biomass change under climate and effort
A. Higher TL biomass change under climate change
A. Historical trends of fishing events
A. Lower TL biomass change under climate and effort
A. Lower TL biomass change under climate change
A. Monte Carlo simulations food web groups
A. Predicted vs observed biomass fit_PPB
A. Transferability and model skill assessment
Diet_PPB Model
M. Biomass change under fishing and climate scenarios
M. Biomass changes under climate scenarios in 2100
M. Ecological indicators of change in community
M. Mesocosm transferability and model skill assessmentRelationship between files, if important: None
Are there multiple versions of the dataset? No
A. If yes, name of file(s) that was updated: NA
i. Why was the file updated? NA
ii. When was the file updated? NAThe presence of empty cells in certain Excel sheets indicates that data for those specific entries were either not reported for that year or were unavailable
DATA-SPECIFIC INFORMATION FOR: A. Ecosystem model skill assesment_TestData
- Number of variables: 5
- Number of cases/rows: 77
- Variable List: Species (Name of the species); Year (Data for the year); predictedPPB (model predicted biomass for PPB); predictedMESO observed (model predicted biomass for mesocosm)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S7
DATA-SPECIFIC INFORMATION FOR: A. Effect sizes and direction of impacts on biomass
- Number of variables: 3
- Number of cases/rows: 30
- Variable List: Communitygroups (Community groups are broader groups within the food web that are composed of associated/similar species/groups); Hedgesg (Hedges’ g is a measure of effect size); Scenario ( Different climate change scenario tested in the model)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S2
DATA-SPECIFIC INFORMATION FOR: A. Global sensitivity analysis_OAT
- Number of variables: 6
- Number of cases/rows: 5200
- Variable List: Trial (Number of simulation Ecosim model simulation); B_OAT (Biomass under combined ocean acidification and warming); PB_OAT (Production biomass ratio under combined ocean acidification and warming); QB_OAT (consumption biomass ratio under combined ocean acidification and warming); EE_OAT (Ecotrophic efficiency under combined ocean acidification and warming); FutureB_OAT (Modeled future biomass under combined ocean acidification and warming)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Table S7
- Missing data code: NA
DATA-SPECIFIC INFORMATION FOR: A. Higher TL biomass change under climate and effort
- Number of variables: 6
- Number of cases/rows: 420
- Variable List: CommunityGroup (Community froup); Functionalgroup (food web functional group); BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario); Effort (Level of fishing effort)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript: Figure S5
DATA-SPECIFIC INFORMATION FOR: A. Higher TL biomass change under climate change
- Number of variables: 4
- Number of cases/rows: 105
- Variable List: Functionalgroup (food web functional group); BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario);
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S3
DATA-SPECIFIC INFORMATION FOR: A. Historical trends of fishing events
- Number of variables: 2
- Number of cases/rows: 27
- Variable List: Fishing events
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript: Figure S8
DATA-SPECIFIC INFORMATION FOR: A. Lower TL biomass change under climate and effort
- Number of variables: 6
- Number of cases/rows: 204
- Variable List: CommunityGroup (Community froup); Functionalgroup (food web functional group); BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario); Effort (Level of fishing effort)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S6
DATA-SPECIFIC INFORMATION FOR: A. Lower TL biomass change under climate change
- Number of variables: 4
- Number of cases/rows: 51
- Variable List: Functionalgroup (food web functional group); BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario);
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S4
DATA-SPECIFIC INFORMATION FOR: A. Monte Carlo simulations food web groups
- Number of variables: 94
- Number of cases/rows: 138529
- Variable List: Communitygroup (Community groups are broader groups within the food web that are composed of associated/similar species/groups); Model (Different climate change scenario tested in the model including the no change scenario); Year; Trail 1 to trial 91 (Number of valid Monte Carlo simulations run)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S10
DATA-SPECIFIC INFORMATION FOR: A. Predicted vs observed biomass fit_PPB
- Number of variables: 4
- Number of cases/rows: 104
- Variable List: Year; Observed (fisheries biomass or catch data for Port Philip Bay); Predicted (model predicted biomass or catch data); Species (selected species for model comparison)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure 9
DATA-SPECIFIC INFORMATION FOR: A. Transferability and model skill assessment
- Number of variables: 5
- Number of cases/rows: 38
- Variable List: Year; Predicted (model predicated biomass or catch); Species; Model (Mesocosm or Port Philip Bay food web model); Observed (fisheries biomass or catch data for Port Philip Bay)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure S7
DATA-SPECIFIC INFORMATION FOR: Diet_PPB Model
- Number of variables: 2 varaibles and 47 column with different species or functional groups
- Number of cases/rows: 55
- Variable List: Functional group or species
- Specialized formats or other abbreviations used: Matrix (Diet data for food web model) 5: Figure/Table associated with the manuscript : Figure S3
DATA-SPECIFIC INFORMATION FOR: M. Biomass change under fishing and climate scenarios
- Number of variables: 5
- Number of cases/rows: 12000
- Variable List: CommunityGroup (Community froup); Functionalgroup (food web functional group); BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario); Effort (Level of fishing effort)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure 3
DATA-SPECIFIC INFORMATION FOR: M. Biomass changes under climate scenarios in 2100
- Number of variables: 5
- Number of cases/rows: 3000
- Variable List: Communitygroups (Community groups are broader groups within the food web that are composed of associated/similar species/groups; Trial (Number of valid Monte Carlo simulations run) BiomChangePercent (Change in biomass); FutureScenerio (Climate change scenario); TL= Trophic level
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure 2
DATA-SPECIFIC INFORMATION FOR: M. Ecological indicators of change in community
- Number of variables: 98
- Number of cases/rows: 42623
- Variable List: Year; Effort (level of fishing effort); Scenario (Climate change scenario); Indicator (Model derived biodiversity indicator)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure 4
DATA-SPECIFIC INFORMATION FOR: A. Higher TL biomass change under climate and skill assessment
- Number of variables: 5
- Number of cases/rows: 76
- Variable List: Year; Predicted (model predicated biomass or catch); Species; Model (Mesocosm or Port Philip Bay food web model); Observed (fisheries biomass or catch data for Port Philip Bay)
- Specialized formats or other abbreviations used: None 5: Figure/Table associated with the manuscript : Figure 5
Methods
Mesocosm Experiments
We utilized our previously published mesocosm experiments to investigate the implications of future climate change on marine food webs. These experiments were specifically designed as multilevel, species-rich mesocosms (Ullah et al., 2018; Pistevos et al., 2015).
Model Parameterization
For various functional groups, we extracted model parameters from our mesocosm study by Ullah et al. in 2018. The sole exception was for chondrichthyans, for which parameters were derived from external studies.
Consumer Composition in Mesocosms
Our focus was primarily on highly mobile omnivorous and carnivorous consumers, consisting of juvenile forms of six different fish species and two shrimp species. Initial population density in each mesocosm comprised 55 individuals, varying in size between 10-40 mm. These mesocosms were subjected to climate treatments for 4.5 months, allowing both trophic and competitive interactions to influence consumer composition. By the end of the experiment, an average of 25.1 ± 4.4 individuals per mesocosm remained.
Estimating Species Interaction Under Climate Change
Using the data from our previous mesocosms, we estimated predator-prey interactions within food web models. We assessed the vulnerability of prey to predators, the predators' effective search rate, and biomass-dependent mortality for some groups.
Indirect Approach to Estimating Prey Vulnerability
We employed an indirect approach, based on risk-tasking behavior in omnivorous and carnivorous fishes, to assess the vulnerability of prey groups at trophic levels ≥ 2 (Goldenberg et al., 2018). Three metrics were used: prey attraction, food search activity, and boldness, which were assessed through manual video tracking using Solomon Coder software.
Effective Search Rates for Sharks
The search efficiency of Port Jackson sharks was quantified by measuring the time required to locate hidden prey in the sand. Data were sourced from Pistevos et al. (2015).
Assessment of Primary Producers
Biomass data for primary producer groups, including phytoplankton, phytobenthos, mat-forming algae, and macrophytes, were also collected from mesocosms.
PPB Food Web Modeling
The PPB food web model was built using the Ecopath approach, providing a static snapshot for 1990 and extending it to a dynamic model up to 2015 (Christensen et al., 2008). The model contains 53 functional groups and requires four primary input variables.
Climate Change Effect Sizes and Forcing Functions
We calculated the mean effect sizes relative to present-day controls for multiple variables, including prey vulnerability and search activity. These were then used to derive model-forcing functions for various climate change scenarios.
Climate Change Impact Assessment
The ecological implications of climate change were analyzed by comparing observed biomasses and ecological indicators under a no-change (NC) scenario to those under climate change scenarios.