Advancing Sustainable Development and Protected Area Management with Social Media-Based Tourism Data

How to cite this study

Arkema, K.K., Fisher, D.M., Wyatt, K., Wood, S.A. and Payne, H.J. 2021. Advancing sustainable development and protected area management with social media-based tourism data. Sustainability 2021(13), 2427.


This study uses social media to measure tourism levels and spatial variation in The Bahamas Marine Protected Areas (MPAs) and Andros Island. Social media data from Flickr was found to be well correlated with visitor counts from entrance surveys. Annual visitation to The Bahamas MPAs was estimated to be an average of 383,000 visitor days. Combining social media with entrance surveys in Andros Island, the largest island in The Bahamas, tourists were found to spend 125,000 visitor nights and more than $45 million annually in the most visited district.


This study is relevant to those interested in using social media data to measure visitation patterns. Using social media can be especially useful in remote areas or areas lacking visitation data. Visitation estimates can inform investments in new infrastructure and accommodations to support popular tourist areas while also establishing protected spaces to prevent overuse and environmental degradation. However, social media users are not a random sample and Flickr data may be biased by the popularity of the website which can vary by year, user groups, and geography.


This study is located in The Bahamas. The study includes 43 Marine Protected Areas, focusing on Andros Island, the largest island in The Bahamas.

Trail Type

This study is located in The Bahamas, a country in the Caribbean. Visitor estimations include Andros Island and the 43 parks of the Bahamian Marine Protected Area network.


The purpose of this study was to use social media to generate visitation estimates and assess spatial patterns. These estimates could be used by The Bahamas to support the Caribbean Challenge Initiative that planned to protect and effectively manage 20% of marine and coastal ecosystems by 2020, as well as Vision 2040, a national development planning process to support economic opportunity and environmental protection. Research was funded by the Inter-American Development Bank and The Bahamas Reef Environment Educational Foundation.


  • A significant positive relationship was found between average annual visitation recorded on-site counts and the data from Flickr’s PUDs. 
  • 383,000 visitor days were estimated for the Bahamian MPAs annually. Visitor days are the total number of days that all tourists visit.
  • Tourists spent about 311,000 visitor nights on Andros Island and the surrounding area. Visitor nights are the number of nights that all tourists spend on each visit. 
    • Central Andros was estimated to have 125,000 visitor nights and $45 million USD in expenditures in 2014, with 160 staff and guides in nature-based tourism. 
    • North Andros was estimated to have 106,000 visitor nights and $39 million in expenditures, and an estimated 134 nature-based tourism staff and guides. 
    • South Andros estimated 63,000 visitor nights and $23 million in expenditures with 98 staff. 
    • Mangrove Cay estimated 17,000 visitor nights and $6.17 million in expenditures, with 38 staff. 
  •  In the linear regression that predicted the spatial patterns of visitation to Andros, development explained the greatest variation in visitation, followed by coastal development, access, and coral located near dive lodges, beaches, bonefish habitat, dredged port, and pine. This suggests that sites with these factors tended to attract more visitors. 
  • In contrast, agriculture was a significant negative predictor, suggesting that sites with agriculture tend to have lower levels of tourism activity. 
  • Results also suggested that sites that combine nature and accessibility/accommodation may attract more visitors. 


To evaluate the correlation between on-site and social media data, geotagged images from Flickr were compared to on-site visitation estimates published by the Ministry of Tourism of The Bahamas and the U.S. National Park Service. Visitation was measured in units of photo-user-days (PUD), and an open-source InVEST Visitation model was used to access the Flickr metadata. The model calculated the average annual PUD on images taken from 2005-2014 for eight islands and each of five national parks. This model was used to predict visitation to each MPA, using a simple linear regression to model the variability in on-site measures of visitations as a function of PUDs. Average annual visits were estimated using the number of PUDS per site. 

A different approach was used to estimate visitation on Andros Island. Spatial variation in visitation, expenditures, and nature-based tourism employment was estimated on Andros Island to inform sustainable development planning. The data was collected from Flickr photographs, airport surveys, and community surveys. The percentage of visitors visiting Andros was estimated as the ratio of average annual PUDs from Andros to average annual PUDS from all islands. This was then distributed across the islands’ 5 km grid cell scale to estimate spatial variation. Average annual expenditures were estimated using average annual visitor nights and an estimated $364 USD spent per night by visitors, which was from the Bahamas Ministry of Tourism Statistics. Data on tourism-related jobs were collected at the district level through local surveys of communities from 2009. Only jobs related to nature-based tourism were included in the analysis. Tourists’ preferences for different areas were estimated using a revealed preferences approach, estimating the preferences using a simple linear regression model with predictor variables like natural features, accessibility, development, and human activities to explain spatial variation.

Added to library on November 27, 2023