To support our upskilling objectives across different contexts, we have developed a core portfolio of training materials. These materials focus on teaching fundamental competencies and are organized into themed modules (e.g., wildlife protection, human-wildlife conflict). Depending on the local context, we select the most relevant modules and training topics. Our locally recruited mentors and trainers are then encouraged to adapt these materials based on their specific expertise and background.

Our goal is that our core portfolio of standardized training materials are delivered by female experts recruited from the local region, who we further engage in mentoring and leadership activities. By centering these role models throughout our programming, we provide our participants with a vision of their future careers. We strive to foster an inclusive environment for honest dialogue and encourage ongoing mentorship even after the program concludes. However, the very gender gap we aim to address often presents a challenge when it comes to recruiting female educators and role models for our programs. This situation has helped us to differentiate three leadership roles: “mentors” (female role models, who participate in training and mentorship), “allies” (male trainers and facilitators), and “trainers” (support from international organizing team). Participation of each to these types of individuals is critical to develop and support our participants.

This program aims to equip women with practical skills that are actionable within their local context, enabling them to seize opportunities such as funding and career advancement within their specific regions. To achieve this, we collaborate closely with local partners and host institutions to adapt our core training materials, ensuring they align with local challenges, processes, and institutions. By tailoring our trainings to address the unique needs and contexts of the women we support, we maximize the relevance and impact of our programming. 

Strong partnerships with local NGOs and educational departments have been critical to the success of the Arribada Clubs. These partnerships enable the customization of the curriculum to reflect community-specific conservation priorities, such as sea turtle protection in Príncipe or biodiversity monitoring in Kenya. Collaborative planning ensures that the clubs address local needs and have a lasting impact.

The Arribada Club provides hands-on STEM education tailored to conservation needs. Delivered through after-school programs in underserved communities, the curriculum incorporates local conservation challenges into lessons, fostering a deep connection between students and their environment. Students gain practical experience with tools like GPS, microcomputers, and bioacoustic monitoring, learning how these technologies support biodiversity conservation. This education empowers local youth with technical skills essential for both personal and community growth while fostering future conservation leaders.

Drivers like marine environmental factors (e.g., salinity, wave height) and anthropogenic influences (e.g., aquaculture) were analyzed using a Generalized Additive Model (GAM) to explore their relationships with vegetation evolution.

• Spartina alterniflora was influenced by marine factors (e.g., salinity and wave height).
• Phragmites australis and Suaeda salsa were driven by rainfall, anthropogenic activities, and interspecific competition.

Understanding these factors supports better management of invasive species and promotes biodiversity conservation.

Using spatial-temporal data, the long-term distribution characteristics of wetland vegetation were analyzed. The results of the study show the evolution of vegetation in the protected area both spatially and temporally. Figure 1A shows the spatiotemporal patterns of vegetation distribution from 1990 to 2022. Figure 1b shows the observed percent vegetation cover along the sea–land gradient from 1990 to 2022. Tools such as the landscape pattern index, migration modeling, and expansion/decline dynamics identified distinct trends:

• Spartina alterniflora patches showed high aggregation and a decreasing trend.
• Phragmites australis and Suaeda salsa patches displayed lower aggregation, higher fragmentation, and increasing trends.
• Vegetation migration patterns revealed significant spatial and temporal heterogeneity, with vegetation distributed in bands along the terrestrial gradient.

A comprehensive database was created, integrating remote sensing-derived vegetation cover with key environmental, climate, and human activity data. This includes metrics such as soil salinity, sea surface temperature, seawater salinity, and aquaculture pond locations, forming a robust foundation for further analyses.

Using a Gaussian function, vegetation index time series were fitted to reduce noise and extract key features. A random forest deep learning algorithm classified wetland vegetation into three types (Spartina alterniflora, Phragmites australis, Suaeda salsa). Field validation confirmed classification accuracy from 1990–2022.

The purpose of this building block is to automate the traditionally manual process of wildlife monitoring. The model works by analyzing visual data, detecting the presence of vultures, and classifying them into species with high accuracy. This reduces human effort, accelerates data analysis, and ensures consistency in monitoring.