Defining the location and condition of coastal habitats, including coral reef, mangroves, and seagrass, is essential for effective protection and management of these invaluable ecosystems. Producing detailed habitat maps over large spatial scales, however, has traditionally required large capital investments and the deployment of highly skilled local practitioners to remote geographies. We are bridging this critical knowledge gap by leveraging recent advances in remote sensing technologies and partnering with the organizations at the forefront of this rapidly developing field. In collaboration with Planet, whose fleet of around 200 satellites collects global data on a daily time step, and the Carnegie Airborne Observatory, whose high-fidelity imaging spectrometers and multiple laser scanners provide high fidelity habitat differentiation, we are pioneering coastal monitoring approaches to provide spatial, temporal, and taxonomic resolutions never before possible. eCognition, an object-oriented approach, is being used to delineate benthic habitats throughout the Caribbean using the Planet satellite imagery. In contrast to pixel-based classification methods, object-oriented image analysis segments satellite data into seascape objects that have ecologically-meaningful shapes, and classifies the objects across spatial, spectral, and textural scales. These objects represent distinct patches of uniform benthic habitat and are classified based on both spectral and non-spectral attributes of the imagery which include bathymetry, geomorphic zones, and corresponding texture, spatial, and contextual information. We integrated thousands of GPS-referenced field video transects, drone, and scuba diver data collected throughout the area to train the classification algorithm. This innovative, scalable approach to coastal ecosystem mapping and monitoring will provide the data needed to target coastal conservation actions to provide maximum human benefit, including ecosystem valuation, targeted conservation and habitat restoration, and effective marine spatial planning.

Papahānaumokuākea Marine National Monument (PMNM) is one of the largest protected marine conservation areas in the world. During NOAA’s 2017 Reef Assessment and Monitoring Program expedition to PMNM, Rapid Ecological Assessments (REAs) were conducted to survey fish, coral, and invertebrate abundance at 7 remote islands, shoals, and atolls. In addition to the REAs, the benthos was photographed in order to be reconstructed into 3D reef models. The goal of this research is to analyze REA and benthos characteristics in an effort to study reef complexity and ecosystem function. Utilizing Structure-from-Motion photogrammetry techniques, the photos taken from each transect were aligned using Agisoft software, generated into a point cloud of overlapping data, to ultimately produce a 3D and Digital Elevation Model (DEM) of the reef. The 3D model was flattened to a 2D orthograph, and then analyzed together with the DEM using ArcGIS software. Along the orthograph, arbitrary survey plots were drawn and processed for many different complexity parameters using the Benthic Terrain Modeler tool, such as slope, curvature, roughness, and rugosity of the reef. GIS processing of the 86 reefs surveyed across PMNM has been completed at 1 centimeter (cm) and 5cm resolutions. Additionally, orthographs were analyzed using CoralNet, an online annotation tool for classifying percent coral coverage by genus and morphology (ex. Branching Acropora), as well as percent algal, macro-fauna, and abiotic benthos coverage. Future goals of this ongoing research include GIS processing up to the 100cm resolution, as well as statistical analysis to investigate correlations between reef complexity and REA data.

In the Caribbean, there are numerous organizations propagating corals in nurseries for use in restoration and population enhancement. The most common species is Acropora cervicornis followed by Acropora palmata, both of which are listed as threatened under the US Endangered Species Act (ESA). The ESA Recovery Plan for these two species promotes scaling up of nursery efforts as one of the actions to promote species recovery. As coral propagation is becoming more common and widespread, there is a need to track outplanting efforts, particularly for ESA-listed species, to gauge their distribution and cumulative impact on population status. A preliminary database was developed to collect information, including species, number of corals outplanted, and outplant location, from nursery operators. The information was initially collected only for Acropora species in US jurisdiction, but through the work of the Coral Restoration Consortium, information collection has expanded to include additional species and locations throughout the Caribbean. This information can be spatially depicted in GIS to show information such as nursery locations, species distribution, genotype locations, and outplants through time. The database can be used for a number of purposes such as to inform best management practices, increase partner collaboration, identify priority areas or gaps, or help with scaling up of restoration. We envision an online portal for easy data entry and searchable content that would allow users to download data for their own studies and management purposes. We will present data, maps, and capabilities we have developed to date.

Photomosaics can be a useful scientific tool for efficiently and accurately analyzing large numbers of coral outplants. Coral Restoration Foundation, Inc. (CRF) has developed cost-effective, in-house procedures for creating large-scale (~2,000 sq meters) reef photomosaics of reefs for the purposes of long-term monitoring of Acropora spp. outplants. Here, we compare the monitoring data obtained using standard, in-water data recordings of coral survivorship and health, with that obtained using photomosaics, for transects of outplanted corals located at Carysfort Reef located off of north Key Largo, Florida. We expect to show that (1) similar information can be obtained from each method with comparable accuracy and (2) that photomosaics offer the potential to capture additional, more complex information for calculating sophisticated metrics, such as percent coral cover. Already, photomosaics created by CRF immediately before and after Hurricane Irma allowed for analysis of large areas of reef- not just individual outplanted corals- and provide a clear snap-shot of coral cover. As restoration efforts increase in the future, photomosaics can also offer a way for monitoring emphases to shift from individual corals to entire reef communities, thus allowing natural dynamics such as asexual fragmentation to be captured, deepening our understanding of best restoration practices and coral reef ecosystems.

Monitoring of individual ephemeral coral species such as Acropora cervicornis is difficult because of frequent fragment or displacement, yielding individual colonies nearly impossible to track long-term. However, much of the remaining A. cervicornis exist as low density populations comprised of individual colonies, and we must understand individual colony dynamics of the species in order to develop proper monitoring guidelines and success metrics for population enhancement programs. In this study, the spatial and temporal components of A. cervicornis colony residency and retention were explored by 1) measuring changes in colony abundance, 2) evaluating two methods for colony fate tracking, 3) estimating colony residence and retention rates, and 4) determining if colony size effects colony residency. All parameters were measured within 3.5 m radial plots (n=56) established between numerous sites (n=11) in three sub-regions of the Florida Reef Tract from June 2011 to July 2015. Colony residence times were similar between methods used for fate tracking and less than 16% of colonies remaining after two years. A majority of colony loss came from complete colony dislodgement and not mortality. Mean colony abundance by sub- region did not change significantly between survey events; however, median colony residency time was less than one year, and three month retention rates were between 29-88% for all sub- regions, indicating significant and frequent colony movement within sites. The probability of a colony remaining through the end of the study was over three times greater in the Dry Tortugas (0.12) sub-region than Broward (0.03) and Middle Keys (0.04). Colony size had a positive effect on retention time although the relationship was weak (between 9 and 19%). Our data show that fate tracking is likely underestimating population growth, propagation, survival, and health of the species, ultimately suggesting the need to modify how A. cervicornis are being monitored to describe long-term success and species recovery. Furthermore, these results indicate the success of a population enhancement program should not be defined exclusively by fate tracking outplanted colonies, but should utilize a systematic site level monitoring approach to capture individual species dynamics.

Active restoration to mitigate declines in coral cover is an essential component of coral conservation and management plans. However, the availability of monitoring data and success criteria is limited and often leads to criticism of restoration projects. Expanding an evaluation tool developed by Lirman et al. (2017), immediate and long-term success metrics were developed for evaluating restoration efforts to assess individual or collaborative performance and advancement towards restoration goals. Metrics provided within this tool are designed to evaluate the strength and robustness of each restoration project, program, or regional-scale effort, while also identifying specific metrics which may require action to improve performance. This tool follows the recovery goals, objectives and criteria outlined in the Recovery Plan for Elkhorn and Staghorn Corals (NMFS 2015) which may also be applied to additional species now listed within the US Endangered Species Act or have recently suffered dramatic losses in abundance and cover due to severe bleaching and disease events. Success metrics focus on increasing the abundance of and protecting the genetic diversity of coral populations throughout their geographical ranges through restoration and active management. Therefore, metrics outlined here focus on best management practices or results from restoration-based research conducted by experts in the field of coral propagation and outplanting. The intent of this tool is to evaluate each restoration metric using a stop-light indicator framework and allow self-critique of methodologies, techniques, and protocols to promote the design of adaptive strategies to improve performance and encourage communication between restoration partners (locally, regionally, or globally) to increase success. Therefore, this tool will advance the development of science-based benchmarks to achieve population-based recovery for coral reefs by evaluating the status of restoration techniques, outlining the positive attributes of productive projects and programs, and promoting the development of successful strategies.

With the proliferation of local and regional coral propagation and reef restoration programs around the world, there is an increasing need to develop uniform, consistent guidelines to ascertain the status of these efforts that are initiated with differing levels of expertise and a wide range of programmatic goals (e.g., restoration, education, enhanced livelihoods). Here, we will describe a simple assessment tool based on the stop-light indicator framework that captures status information of a wide range of potential project components and goals, including key steps such as coral collection, nursery deployment and maintenance, coral monitoring, stakeholder involvement, funding sources, data sharing, education and outreach, and project sustainability. The scores obtained for the different desired program attributes are combined into a single score that can be easily tracked among projects and over time to see the status and progress of the restoration activities. The assessment tool proposed is tested using data from seven restoration projects from the Dominican Republic, highlighting strengths and shortcomings of each project as well as identifying steps that can be taken to improve project status. This tool was expanded by Schopmeyer et al. (2018) to evaluate the status of regional, multi-species restoration programs. Results from this regional perspective will be detailed in a separate presentation.

The ability to monitor the growth and survival of a restored coral is vital for assessing the success of a restoration program as a whole. Information on outplant performance is useful to restoration managers to direct future operations as well as to scientists using such coral to understand ecological and evolutionary processes of the reefs. Current non-intrusive methods for monitoring coral are limited to total linear extension or indirect estimates of surface area which can be difficult to obtain in the field, especially when dealing with structurally complex corals. 3D photogrammetry provides a way to digitally reconstruct a coral and quantify morphometric characteristics outside of the water. Within the timespan of a routine monitoring survey, Acropora cervicornis fragments were photographed with a standard underwater camera from an array of angles to produce 50 to 100 images; a process that required roughly 90 seconds per fragment. Images of each coral were entered into the Agisoft Photoscan software which then generated the 3D models. Using the scaling targets photographed into the models, we calculated linear extension, surface area, and volume of 270 corals outplanted by Mote Marine Laboratory’s Staghorn Restoration Project. In order to ground-truth the methodology, we compared photogrammetry-based measures to in situ measurements of total linear extension. To assess bleaching, analysis of color cards included in photographs provided quantification of fragment color with no additional effort on the part of the surveyors. Long-term data on outplant survival will enable determination of the utility of 3D photogrammetry for assessing the performance of coral outplants. By obtaining direct total linear extension, surface area, volume, and color measurements, 3D photogrammetry can provide a more comprehensive method of monitoring outplant success when compared to traditional methods without adding substantial effort to current restoration practices.