BioSCape Invasive Alien Plant map for the Cape Floristic Region

BackgroundThis project was supported by the NASA BioSCape Project: “Impacts of Invasive Alien Species on Biodiversity and Ecosystem Functioning”. The training data used to produce this classification are also available. This dataset consists of a raw, unprocessed classification map of key woody invasive alien plant taxa (hereafter “trees”) within the Cape Floristic Region. This classification was generated using Sentinel-2 satellite imagery at a 10 m resolution using a Random Forest machine learning classifier. There has been no post-processing, and therefore it should be used with caution, especially in combination with the uncertainty layer provided.Dataset specificationsData type: Raster (tiff)Resolution: 10 mGeographic Coordinate System: WGS 1984Projected Coordinate System: Albers Equal Area projection (AEA)There is no EPSG for this, however the Well-Known Text (WKT) for this projected coordinate system is as follows:var wktAEA = 'PROJCS["AEA_RSA_WGS84",' +'GEOGCS["GCS_WGS_1984",' +'DATUM["D_WGS_1984",' +'SPHEROID["WGS_1984",6378137.0,298.257223563]],' +'PRIMEM["Greenwich",0.0],' +'UNIT["Degree",0.0174532925199433]],' +'PROJECTION["Albers"],' +'PARAMETER["False_Easting",0.0],' +'PARAMETER["False_Northing",0.0],' +'PARAMETER["Central_Meridian",25.0],' +'PARAMETER["Standard_Parallel_1",-24.0],' +'PARAMETER["Standard_Parallel_2",-33.0],' +'PARAMETER["Latitude_Of_Origin",0.0],' +'UNIT["Meter",1.0]]';BandsThis raster has two bands, the details of which are described in Table 1.ClassesSix classes of woody invasive alien plants were mapped in this classification (Table 2). This classification goes to genus level only and gives only presence/absence information together with information on uncertainty. Some classes were much harder to map than others and hence have higher uncertainty, such as Poplar and Hakea. A class was added for all the other invasive alien taxa which had large coverage though not enough for their own classes: “Other”.

背景：本项研究得到美国国家航空航天局（NASA）生物科学项目“入侵性外来物种对生物多样性和生态系统功能的影响”的支持。用于生成该分类的训练数据亦可供使用。本数据集包含开普植物区关键木质入侵性外来植物类群（以下简称“树木”）的原始、未处理分类图。该分类利用10米分辨率的Sentinel-2卫星影像，采用随机森林机器学习分类器生成。未进行后处理，因此在使用时需谨慎，尤其在与提供的不确定性层结合使用时。数据集规格：数据类型：栅格（tiff）；分辨率：10米；地理坐标系：WGS 1984；投影坐标系：阿尔伯斯等面积投影（AEA）。该投影坐标系无EPSG代码，但其已知文本（WKT）如下：
var wktAEA = 'PROJCS["AEA_RSA_WGS84",...];
波段：此栅格包含两个波段，其详细信息见表1。
类群：本分类图中绘制了六类木质入侵性外来植物（见表2）。该分类仅达到属的水平，并提供存在/不存在信息及不确定性信息。有些类群比其他类群更难以绘制，因此具有更高的不确定性，例如白杨和桃金娘。对于其他具有广泛覆盖但不足以形成独立类群的外来植物类群，增加了“其他”这一类。