Abstract
Recruitment is a fundamental process that influences coral population dynamics and reef community structure. However, recruitment success rates remain poorly quantified due to labor-intensive manual survey methods prone to human error. This study introduces a non-invasive methodology using the Fluorescence Imaging System (FluorIS) to identify coral recruits as small as 1 mm in diameter during the day. By monitoring quadrats in the Gulf of Aqaba, Israel, over eight months, the researchers demonstrated that fluorescence imaging reduces surveyor error and improves precision, detecting 24% more recruits than traditional visual surveys. The findings suggest relative stability in the local coral population, providing a standardized tool for coral reef monitoring and management.
The Challenge of Monitoring Coral Recruitment
Coral recruitment—the settlement and growth of larval corals—is vital for reef recovery. Quantifying this process is difficult because baby corals are tiny, cryptic, and often camouflaged against the reef substrate. Traditional methods, such as artificial settlement plates or in situ visual surveys by divers, are either invasive or logistically complex, often requiring night dives with UV lights to see the fluorescence. To address these limitations, researchers developed a repeatable, daytime photographic method to detect recruits in their natural environment with high accuracy.
The Fluorescence Imaging System (FluorIS)
The methodology leverages the natural fluorescence of corals and their symbiotic algae, which contain fluorescent proteins and chlorophyll. The hardware consists of a modified Nikon D810 camera in an underwater housing, equipped with blue excitation strobes and a yellow barrier filter to isolate the fluorescence signal.
To operate during the day, the system employs an ambient light subtraction technique. It captures two rapid-fire images: one with strobes ($I_{day}$) and one without ($I_{ambient}$). By subtracting the ambient image from the strobe image, the sunlight "noise" is removed, leaving only the fluorescence signal. Custom "RecruitTracker" software uses SURF and RANSAC algorithms to automatically align these image pairs, compensating for slight camera movements between shots.
Key Results and Population Dynamics
The study validated the system through lab tests and long-term field monitoring of 40 quadrats (20 x 20 cm) in the Gulf of Aqaba. Key findings include:
- Enhanced Detection: Fluorescence imaging identified 24% more recruits than experienced divers using blue flashlights.
- Size Sensitivity: The system successfully detected recruits as small as 1 mm in diameter.
- Recruitment Rates: Surveys revealed an average of ~2 new recruits (<2 cm) per quadrat per month.
- Survival Rates: Individual tracking showed an 83% survival rate in the first month, stabilizing at roughly 75% over the eight-month period.
- Precision: The automated registration algorithm was successful for 92% of image pairs, significantly streamlining data processing.
Discussion and Future Directions
While the FluorIS system provides a major advancement in non-invasive monitoring, it is limited by line-of-sight; recruits hidden in deep crevices remain undetectable. Additionally, while image alignment is automated, the identification of recruits still requires manual annotation. Future work aims to integrate deep learning (Convolutional Neural Networks) to fully automate the detection and counting process. Despite these challenges, the ability to conduct high-precision recruit surveys during the day using relatively low-cost equipment offers a significant step toward standardizing coral reef management tools globally.
Summary updated