The unrealised potential of agroforestry (Raw drone images GHA 2022 Part F)

This dataset is part of a larger collection accompanying the analysis presented in “The unrealized potential of agroforestry for an emissions-intensive agricultural commodity” (Becker et al., Nature Sustainability, 2025). The full dataset has been published via UQ eSpace as a series of interlinked records, each representing a different stage of the research workflow—from raw imagery to processed data products and analysis code.

This subset (Part F) includes imagery from farms P401 to P518, collected in 2022 as part of a stratified sampling design in Ghana. It is one of several folders containing raw drone imagery—unprocessed aerial photographs (JPEGs)—captured during field surveys across cocoa farms in Ghana between 2021 and 2022.

These images were collected as part of a broader effort to map shade-tree cover and aboveground biomass using drone-based ground-truth data and machine learning. The raw images in this folder served as the basis for generating orthomosaics, digital surface models, digital terrain models, and vegetation height estimates.

本数据集是配套于《高排放农用商品农林业（agroforestry）未实现潜力》（Becker等人，《自然·可持续性（Nature Sustainability）》，2025年）一文分析的大型数据集合集的组成部分。完整数据集已通过昆士兰大学eSpace（UQ eSpace）平台发布，包含一系列相互关联的数据集条目，分别对应研究工作流的不同阶段——从原始影像数据到处理后的数据产品与分析代码。

本子集（F部分）涵盖P401至P518号农场的影像数据，这些影像于2022年采集，为加纳境内分层抽样设计研究的组成部分。本文件夹是多个存储原始无人机航拍影像的文件夹之一，这批未处理的航空照片（JPEGs）采集于2021至2022年间加纳境内可可种植园的野外调查过程中。

本次影像采集是一项更大规模研究的组成部分，该研究旨在借助基于无人机的地面实测数据与机器学习（machine learning）技术，绘制遮荫树覆盖率与地上生物量分布图。本文件夹内的原始影像为生成正射影像镶嵌图（orthomosaics）、数字表面模型（digital surface models）、数字地形模型（digital terrain models）以及植被高度估算值提供了核心基础。