Open science in conservation: combining citizen science and remote sensing approaches for habitat monitoring

Suter, Samantha (2024) Open science in conservation: combining citizen science and remote sensing approaches for habitat monitoring. PhD thesis, University of Glasgow.

Abstract

Global biodiversity conservation efforts have not been sufficient to reverse declining biodiversity trends. These shortcomings have led to gaps in monitoring initiatives for habitats, taxa, and regions. Public disconnect from nature and a lack of policy implementation to action scientific research have exacerbated these issues. Alternative monitoring tools, such as citizen science (CS) and remote sensing (RS), have the potential to increase the spatial and temporal reach of monitoring, meeting numerous global biodiversity targets. As such, there are calls for CS and RS to be united. Therefore, this thesis aimed to pair CS and RS and a provide a low cost, low intensity open science (OS) tool in Scotland. This tool sought to map a UK priority habitat – species-rich grasslands (SRGs), which have been widely reduced in area, resulting in coupled invertebrate decline and diminished ecosystem functioning. To increase success of monitoring attempts, an OS approach was adopted, whereby collaboration with stakeholders was key to enhance research impact and scientific democratisation.

To establish an OS framework, openness in current biodiversity monitoring CS surveys was initially investigated, revealing that these surveys did not consistently adhere to OS practices. This research was vital to informing the design of the CS survey of this thesis and identifying efforts to ensure full openness along the scientific process. In determining OS practices, open access Sentinel-2 satellite imagery was acquired to create a habitat classification model, with a final accuracy of 98.6%. Other RS applications were explored, such as the Spectral Variation Hypothesis and grassland trait retrieval, to investigate open access RS data in subsequent mapping attempts of SRGs. The results found no significant relationship between spectral and species diversity, and grassland traits were mostly poorly predicted across spatial and spectral scales. The habitat prediction model was applied to satellite imagery across Scotland, predicting areas of SRGs. In exploring the model outputs, a CS survey, Ecosystem Explorers, was created with Butterfly Conservation, where participants surveyed these predicted areas. Data on previously surveyed SRGs was provided by collaborators such as NatureScot, Plantlife, and the Botanical Society of Britain and Ireland. The model predicted and citizen ground-truthed SRG locations were compared and a poor alignment of 17.65% was found. Participant identification experience and habitat assessment confidence did not affect the level of agreement between the model predictions and ground-truthed observations. However, OS was implemented to combine CS and RS in a highly accessible project, with a predicted open score of 0.92/1.

The thesis provides an example of a novel, OS biodiversity monitoring tool by combining CS and RS methods. The attempt to predict SRGs across Scotland utilising this instrument was unsuccessful. Although the cause of this poor alignment is unclear, citizen scientists appear to be equally as consistent as professionals in their observations, suggesting potential weaknesses in the national application of the model for SRG predictions. This would need further exploration and then poses the question of how an OS tool can be used to improve biodiversity monitoring. Throughout the thesis, guidance and recommendations are provided on how and where OS practices and methodologies can be used for future biodiversity monitoring.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Supported by a Social Sciences PhD Scholarship from the University of Glasgow’s College.
• Subjects: G Geography. Anthropology. Recreation > GE Environmental Sciences; H Social Sciences > H Social Sciences (General)
• Colleges/Schools: College of Social Sciences > School of Social & Environmental Sustainability
• Supervisor's Name: Welden, Dr. Natalie and Barrett, Dr. Brian
• Date of Award: 2024
• Depositing User: Theses Team
• Unique ID: glathesis:2024-84402
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 26 Jun 2024 12:58
• Last Modified: 26 Jun 2024 13:06
• Thesis DOI: 10.5525/gla.thesis.84402
• URI: https://theses.gla.ac.uk/id/eprint/84402

Related URLs

• Article DOI