Global Habitat Suitability for Framework-Forming Cold-Water Corals - Environmental Data

Environmental data layers used in the production of global habitat suitability models for several species of Scleractinian corals from the publication: Davies, A.J. & Guinotte, J.M. (2011) Global Habitat Suitability for Framework-Forming Cold-Water Corals. PLoS ONE 6(4): e18483. doi:10.1371/journal.pone.0018483 Files are ASCII raster layers, zipped using 7zip. Each raster has a WGS 1984 coordinate system. Brief methodology and outline of variables provided are presented below, for full details refer to the above manuscript in PLoS ONE: Terrain variables Several terrain attributes were extracted from the SRTM30 data (Table 1) following techniques and algorithms described in Wilson et al (2007). Individual approaches are described within the footnote of Table 1, however, briefly the extraction process and description of each variable is described here. Bathymetric position index (BPI) is an approach to determine topographical features based on their relative position within a neighbourhood, and can be calculated over fine or broad scales to capture smaller or larger terrain features respectively. This calculation has been developed into an ArcGIS tool by Wright et al. (2005). Slope was calculated using DEM Tools for ArcGIS developed by Jenness (2012), in particular the 4-cell method of calculating slope, which is accepted as the most accurate approach (Jones 1998). In this manuscript, slope is defined as the gradient in the direction of the maximum slope. Curvature attempts to describe terrain features and may provide an indication of how water would interact with the terrain. In this manuscript, plan and tangential curvature can describe how water would converge or diverge as it flows over relief, whilst profile curvature describes how water would accelerate or decelerate as it flows over relief (Jenness 2012). Aspect is defined as the direction of maximum slope and was converted to continuous radians following Wilson et al. (2007). Rugosity, terrain ruggedness index and roughness all generally describe the variability of the relief of the seafloor (Wilson et al. 2007). Rugosity is defined as the ratio of the surface area to the planar area across a neighbourhood of a central pixel (Jenness 2012). Terrain ruggedness index is defined as the mean difference between a central pixel and its surrounding cells and roughness which is the largest inter-cell difference of a central pixel and its surrounding cell (Wilson et al. 2007). Environmental variables Environmental variables were created using the variable up-scaling approach presented within (Davies & Guinotte 2011). This approach takes gridded layers of an environmental variable and drapes them over bathymetry to provide an indication of conditions near the seabed, it has been proven to work well over global and regional scales (Davies & Guinotte 2011, Guinotte & Davies 2012). In this manuscript, these environmental layers must be considered as representations of general conditions, as likely, the highly variable topography of the canyon will not yield a good prediction of environmental variables at such a small spatial scale. Limited CTD profiles were collected using a SeaBird 911+, collating data for turbidity (Seapoint, formazin turbidity units), dissolved oxygen (mg L-1), depth (m), conductivity (Siemens/m), temperature (°C), salinity, and pH. These casts were compared with the modelled layers (specifically dissolved oxygen, salinity and temperature) to determine their relative accuracy for certain areas of the seafloor Variable name // Filename // Units // Reference (see original paper) Terrain Variables Aspect // aspect // Degree // Jenness (2012) Aspect – Eastness // eastness // // Wilson et al. (2007) Aspect – Northness // northness // // Wilson et al. (2007) Bathymetry // srtm30 // m // Becker et al. (2009) Curvature – Profile // profilecurve // // Jenness (2012) Curvature – Plan // plancurve // // Jenness (2012) Curvature – Tangential // tangcurve // // Jenness (2012) Roughness // roughness // // Wilson et al. (2007) Rugosity // rugosity // // Jenness (2012) Slope // slope // Degrees // Jenness (2012) Terrain Ruggedness Index // tri // // Wilson et al. (2007) Topographic Position Index // tpi // // Wilson et al. (2007) Environmental variables Alkalinity // alk_stein // μmol l-1 // Steinacher et al. (2009) Apparent oxygen utilisation // woaaoxu // mol O2 m-3 // Garcia et al. (2006b) Chlorophyll a // modismin, modismean, modismax // mg m-3 // NASA Ocean Color Dissolved inorganic carbon // dic_stein // μmol l-1 // Steinacher et al. (2009) Dissolved oxygen // woadiso2 // ml l-1 // Garcia et al.(2006a) Nitrate // woanit // μmol l-1 // Garcia et al. (2006b) Omega aragonite // arag_stein // ΩARAG // Steinacher et al. (2009) Omega aragonite // arag_orr // ΩARAG // Orr et al. (2005) Omega calcite // calc_stein // ΩCALC // Steinacher et al. (2009) Omega calcite // calc_orr // ΩCALC // Orr et al. (2005) Percent oxygen saturation // woapoxs // % O2S // Garcia et al. (2006b) Phosphate // woaphos // μmol l-1 // Garcia et al. (2006b) Regional current velocity // regfl // m s-1 // Carton et al. (2005) Salinity // woasal // pss // Boyer et al. (2005) Silicate // woasil // μmol l-1 // Garcia et al. (2006b) Seasonal variation index // lutzsvi // // Lutz et al. (2007) Temperature // woatemp // °C // Boyer et al. (2005) Particulate organic carbon // poc // g Corg m-2 yr-1 // Lutz et al. (2007) Vertical current velocity // vertfl // m s-1 // Carton et al. (2005) Vertically generalised productivity model // vgpmmin, vgpmmean, vgpmmax // mg C m-2 d-1 // Behrenfeld and Falkowski (1997)

本数据集为论文《Davies, A.J. & Guinotte, J.M. (2011) Global Habitat Suitability for Framework-Forming Cold-Water Corals. PLoS ONE 6(4): e18483. doi:10.1371/journal.pone.0018483》中用于构建多种石珊瑚（Scleractinian corals）全球生境适宜性模型所使用的环境数据图层。文件为ASCII栅格图层（ASCII raster layers），通过7zip压缩，所有图层均采用WGS 1984坐标系。下文仅提供变量的简要研究方法与框架，完整细节请参阅上述发表于PLoS ONE期刊的研究论文。

## 地形变量
本研究从SRTM30数据中提取了多项地形属性（详见表1），提取方法与算法参照Wilson等人2007年的研究成果。各变量的具体处理流程详见表1脚注，下文简要介绍各变量的提取逻辑与定义：
1. 水深位置指数（Bathymetric Position Index, BPI）：基于中心像元在邻域内的相对位置识别地形特征，可通过精细或粗尺度计算分别捕捉小型或大型地形特征。该计算方法已由Wright等人2005年开发为ArcGIS工具。
2. 坡度（Slope）：采用Jenness 2012年开发的ArcGIS DEM工具计算，具体使用4像元法求解坡度，该方法被认为是目前精度最高的坡度计算方式（Jones 1998）。本研究中坡度定义为最大坡度方向的梯度。
3. 曲率（Curvature）：用于描述地形特征，可反映水流与地形的相互作用关系。本研究中，平面曲率与切向曲率可描述水流流经地形时的汇聚或发散状态，而剖面曲率则反映水流流经地形时的加速或减速情况（Jenness 2012）。
4. 坡向（Aspect）：定义为最大坡度的方向，参照Wilson等人2007年的方法转换为连续弧度值。
5. 粗糙度、地形崎岖度指数与地表粗糙度：三者均用于描述海底地形起伏的变化程度（Wilson et al. 2007）。其中，粗糙度为中心像元邻域内的地表面积与平面面积之比（Jenness 2012）；地形崎岖度指数为中心像元与周围像元的平均差值；地表粗糙度则为中心像元与周围像元的最大差值（Wilson et al. 2007）。

## 环境变量
环境变量基于Davies与Guinotte 2011年提出的变量升尺度方法构建。该方法将环境变量的网格化图层叠加至水深数据，以反映海底附近的环境条件，已被证实可在全球及区域尺度上取得良好应用效果（Davies & Guinotte 2011, Guinotte & Davies 2012）。需注意，本研究中的环境图层仅为一般环境条件的表征，由于峡谷地形高度复杂，难以在如此小的空间尺度上精准预测环境变量。
研究使用SeaBird 911+型设备采集了有限的CTD剖面数据，涵盖浊度（以Seapoint甲醛浊度单位计）、溶解氧（mg L-1）、深度（m）、电导率（Siemens/m）、温度（°C）、盐度与pH值。将这些实测站位数据与建模图层（尤其是溶解氧、盐度与温度）进行比对，以评估其在部分海底区域的相对精度。

### 变量详情表
| 变量分类 | 变量名称 | 文件名 | 单位 | 参考文献 |
|----------|------------------------|------------------------------------------|---------------------|------------------------------|
| 地形变量 | 坡向 | aspect | 度（Degree） | Jenness (2012) |
| 地形变量 | 坡向-东向分量 | eastness | - | Wilson et al. (2007) |
| 地形变量 | 坡向-北向分量 | northness | - | Wilson et al. (2007) |
| 地形变量 | 水深 | srtm30 | 米（m） | Becker et al. (2009) |
| 地形变量 | 剖面曲率 | profilecurve | - | Jenness (2012) |
| 地形变量 | 平面曲率 | plancurve | - | Jenness (2012) |
| 地形变量 | 切向曲率 | tangcurve | - | Jenness (2012) |
| 地形变量 | 地表粗糙度 | roughness | - | Wilson et al. (2007) |
| 地形变量 | 粗糙度 | rugosity | - | Jenness (2012) |
| 地形变量 | 坡度 | slope | 度（Degrees） | Jenness (2012) |
| 地形变量 | 地形崎岖度指数 | tri | - | Wilson et al. (2007) |
| 地形变量 | 地形位置指数 | tpi | - | Wilson et al. (2007) |
| 环境变量 | 碱度 | alk_stein | μmol l-1 | Steinacher et al. (2009) |
| 环境变量 | 表观氧利用率 | woaaoxu | mol O2 m-3 | Garcia et al. (2006b) |
| 环境变量 | 叶绿素a | modismin, modismean, modismax | mg m-3 | NASA Ocean Color |
| 环境变量 | 溶解无机碳 | dic_stein | μmol l-1 | Steinacher et al. (2009) |
| 环境变量 | 溶解氧 | woadiso2 | ml l-1 | Garcia et al.(2006a) |
| 环境变量 | 硝酸盐 | woanit | μmol l-1 | Garcia et al. (2006b) |
| 环境变量 | 文石饱和状态（Omega aragonite） | arag_stein | ΩARAG | Steinacher et al. (2009) |
| 环境变量 | 文石饱和状态（Omega aragonite） | arag_orr | ΩARAG | Orr et al. (2005) |
| 环境变量 | 方解石饱和状态（Omega calcite） | calc_stein | ΩCALC | Steinacher et al. (2009) |
| 环境变量 | 方解石饱和状态（Omega calcite） | calc_orr | ΩCALC | Orr et al. (2005) |
| 环境变量 | 氧饱和度百分比 | woapoxs | % O2S | Garcia et al. (2006b) |
| 环境变量 | 磷酸盐 | woaphos | μmol l-1 | Garcia et al. (2006b) |
| 环境变量 | 区域海流速度 | regfl | m s-1 | Carton et al. (2005) |
| 环境变量 | 盐度 | woasal | pss | Boyer et al. (2005) |
| 环境变量 | 硅酸盐 | woasil | μmol l-1 | Garcia et al. (2006b) |
| 环境变量 | 季节变化指数 | lutzsvi | - | Lutz et al. (2007) |
| 环境变量 | 温度 | woatemp | °C | Boyer et al. (2005) |
| 环境变量 | 颗粒有机碳 | poc | g Corg m-2 yr-1 | Lutz et al. (2007) |
| 环境变量 | 垂直海流速度 | vertfl | m s-1 | Carton et al. (2005) |
| 环境变量 | 垂直广义生产力模型 | vgpmmin, vgpmmean, vgpmmax | mg C m-2 d-1 | Behrenfeld and Falkowski (1997)