Hawaiʻi Coral Disease database (HICORDIS): species-specific coral health data from across the Hawaiian archipelago

The Hawaiʻi Coral Disease database (HICORDIS) houses data on colony-level coral health condition observed across the Hawaiian archipelago, providing information to conduct future analyses on coral reef health in an era of changing environmental conditions. Colonies were identified to the lowest taxonomic classification possible (species or genera), measured and assessed for visual signs of health condition. Data were recorded for 286,071 coral colonies surveyed on 1819 transects at 660 sites between 2005 and 2015. The database contains observations for 60 species from 22 genera with 21 different health conditions. The goals of the HICORDIS database are to: i) provide open access, quality controlled and validated coral health data assembled from disparate surveys conducted across Hawaiʻi; ii) facilitate appropriate crediting of data; and iii) encourage future analyses of coral reef health. In this article, we describe and provide data from the HICORDIS database. The data presented in this paper were used in the research article “Satellite SST-based Coral Disease Outbreak Predictions for the Hawaiian Archipelago” (Caldwell et al., 2016) [1].


a b s t r a c t
The Hawaiʻi Coral Disease database (HICORDIS) houses data on colony-level coral health condition observed across the Hawaiian archipelago, providing information to conduct future analyses on coral reef health in an era of changing environmental conditions. Colonies were identified to the lowest taxonomic classification possible (species or genera), measured and assessed for visual signs of health condition. Data were recorded for 286,071 coral colonies surveyed on 1819 transects at 660 sites between 2005 and 2015. The database contains observations for 60 species from 22 genera with 21 different health conditions. The goals of the HICORDIS database are to: i) provide open access, quality controlled and validated coral health data assembled from disparate  (Table S1)

Value of the data
The data can be used to analyze coral community composition and relationships between community composition and environment; Compare spatial and temporal trends in coral disease severity and prevalence in Hawaiʻi; Investigate the role of trait-based and environmental factors contributing to disease presence and/ or severity; Compare disease patterns in Hawaiʻi with observations from different regions around the world; Create new and more accurate forecasting models of disease outbreaks.

Data
The Hawaiʻi Coral Disease database (HICORDIS) consists of observational surveys of coral health conducted across the Hawaiian archipelago between 2005 and 2015 (Table S1). Ten research groups from academic institutions (University of Hawaiʻi at Mānoa, University of Hawaiʻi, Hilo, Cornell University, University of Wellington), state and federal agencies (Hawaiʻi/Maui Division of Aquatic Resources, National Oceanic and Atmospheric Administration Coral Reef Ecosystem Program) collected survey data at 17 islands and atolls spanning nine degrees of latitude (Fig. 1, Table S2). Data were recorded for 286,071 coral colonies on 1819 transects at 660 sites (Fig. 1). These sites capture the variability in coral community composition and environmental conditions that occur across the $2400 km Hawaiian archipelago. The data presented in this paper were used in the research article "Satellite SST-based Coral Disease Outbreak Predictions for the Hawaiian Archipelago" [1]. method with direct measures of prevalence, divers recorded every coral colony's health condition within a specified area (average length ¼20 m, range¼8-50 m; average width ¼1 m, range¼1-6 m). In the belt transect method with indirect measures of disease prevalence, divers counted all colonies with a health condition (e.g., disease) within a large belt transect area (average area¼ 25 Â 2 m 2 , range¼25 Â 2 m 2 to 25 Â 6 m 2 ) and counted the total number of coral colonies in a subset region of the large belt transect area (average area of subset region ¼10 Â 2 m 2 , range¼10 Â 2 m 2 to 10 Â 6 m 2 ). In the line-intercept method, divers recorded coral health state for every colony directly under 25 m of transect tape. Survey depths ranged from o 1 to 26 m.

Coral colony data
All coral colonies in the HICORDIS database were classified taxonomically and visually assessed for coral health conditions. Observations were recorded for 60 coral species from 22 genera (Table S3). There were 21 possible classifications for coral health state (Tables S4 and S5). Health classifications included no visible lesions, known coral diseases, bleaching, discoloration patterns, algal and bacterial infections and predatory invertebrates. In total, 17% of coral colonies exhibited visual signs of compromised health conditions (Table S4). When data was available, disease severity measurements were also incorporated. Disease severity was quantified as the percent of live tissue affected by a health condition. We note that for coral bleaching in particular, the severity metric we provide here may not be the best measure of severity (e.g., categorizing severity as pale, mottled or stark white may be more indicative of bleaching severity than amount of surface tissue affected).
Most records in the HICORDIS database included a measurement of coral colony size. There were up to four types of measurements recorded for each colony: colony length, colony width, size classes and standardized size classes. Colony lengths and widths were measured as the two longest horizontal axes along a coral colony. Colony lengths ranged from o 1 to 410 cm; colony widths ranged from o1 to 190 cm. Size classes were used to bin colonies into size ranges; however, size classes varied among research groups. In order to facilitate comparisons among observations, a "standardized size class" was included, which grouped coral colonies into size bins based on colony length, colony width and/or size class. Standardized size classes consist of the following size bins: 0-5 cm; 6-10 cm; 11-20 cm; 21-80 cm; 81-160 cm; 161-200 cm; 201-300 cm; 301-450 cm (Fig. 2). An example of how this data can be used to compare coral health across size classes, space and time is provided in Fig. 3. All missing data in the HICRODIS database was recorded as "NA".