Dark Mode
Global Hotspots of Sharks and Longline Fishing
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Data provides a critical global assessment of hotspots for shark interactions with industrial longline fisheries. It uses machine-learning techniques to determine the spatial distribution patterns of at-risk shark species, thereby highlighting crucial risk areas for threatened populations. Through various parameters, such as catch size, environmental values, and presence or absence of species, this resource helps analysts understand which fishing activities potentially threaten sharks while protecting those that are not detrimental. This information supports maneuvering strategies toward sustainability to help conserve the oceans' fragile ecosystems.
Columns
The material contains numerous attributes derived from observation data and model outputs:
- .pred_class: The predicted classification of the species (String).
- pres_abs: Indicates the presence or absence of the species in a region (Boolean).
- catch: The total recorded catch of the species (Integer).
- rfmo: The Regional Fisheries Management Organization (String).
- year: The year of the observation (Integer).
- latitude/longitude: The location details (Float).
- species_sciname/species_commonname: Scientific and common names of the species (String).
- median_price_group/species: Median price associated with the group or species (Float).
- mean_sst/min_sst/max_sst: Mean, minimum, and maximum sea surface temperatures (Float).
- mean_chla: Mean chlorophyll-a concentration (Float).
- bycatch_total_effort_[country]_longline: Total bycatch effort attributed to longline fishing operations from specific countries, including Spain, Portugal, China, Japan, India, and the United Kingdom (Integer).
- rmse/rsq/mae: Results from machine-learning models, including Root Mean Square Error, Coefficient of Determination, and Mean Absolute Error scores (Float).
Distribution
The data is distributed across multiple CSV files, such as
IOTC_ll_untuned_final_predict.csv, WCPFC_ll_models_others_results.csv, and 1x1_count_all_rfmos_ll_effort_results.csv. The catch data has been transformed to make it simpler to use in developing predictive models. One sample file detailing model results contains 12 total values for RMSE, Rsq, and MAE, all of which are 100% valid. For instance, the mean Rsq is 0.45 and the mean MAE is 5.24.Usage
This resource provides valuable insights for researchers and conservationists. It can be used to predict future patterns of shark interactions and pinpoint specific hotspots of activity. It is useful for understanding how human activities and climate change may be impacting sharks and their environments. The data supports the development of targeted strategies and measures to protect threatened shark populations globally and can lead to the design of improved preservation policies.
Coverage
The material provides a global assessment of the interactions between sharks and industrial longline fisheries. It spans various Regional Fisheries Management Organizations and includes detailed environmental data, such as sea surface temperature, sea surface height, and chlorophyll-a concentration. It also incorporates fishing effort data from several nations, including the United Kingdom, Spain, and India. The data is not expected to receive future updates.
License
CC0 1.0 Universal (CC0 1.0) - Public Domain
Who Can Use It
The dataset is highly relevant for researchers and conservationists focused on marine ecology and fisheries management. It is also valuable for data scientists and analysts developing machine learning models to forecast species risk and identify areas requiring focused conservation efforts. The material has a high usability rating of 10.00.
Dataset Name Suggestions
- Global Hotspots of Sharks and Longline Fishing
- Machine-Learning Spatial Distribution of At-Risk Shark Species
- Shark-Fishery Interaction Risk Analysis
Attributes
Original Data Source:Global Hotspots of Sharks and Longline Fishing
QUALITYLoading...