Estimating the Global Distribution of Field Size using Crowdsourcing

There is increasing evidence that smallholder farms contribute substantially to food production globally yet spatially explicit data on agricultural field sizes are currently lacking. Automated field size delineation using remote sensing or the estimation of average farm size at subnational level using census data are two approaches that have been used but both have limitations, e.g. limited geographical coverage by remote sensing or coarse spatial resolution when using census data. This paper demonstrates another approach to quantifying and mapping field size globally using crowdsourcing. A campaign was run in June 2017 where participants were asked to visually interpret very high resolution satellite imagery from Google Maps and Bing using the Geo-Wiki application. During the campaign, participants collected field size data for 130K unique locations around the globe. Using this sample, we have produced an improved global field size map (over the previous version) and estimated the percentage of different field sizes, ranging from very small to very large, in agricultural areas at global, continental and national levels. The results show that smallholder farms occupy no more than 40% of agricultural areas, which means that, potentially, there are much more smallholder farms in comparison with the current global estimate of 12%. The global field size map and the crowdsourced data set are openly available and can be used for integrated assessment modelling, comparative studies of agricultural dynamics across different contexts and contribute to SDG 2, among many others.

The dataset (global field sizes.zip) contains:
- map of dominant field sizes (dominant_field_size_categories.tif) and description of legend items (legend_items.txt)
- table with all submissions by the participant (those who completed more than 10 classifications) and table description
- table with quality score of all the participants and table description
- table with estimated dominant field sizes at each location and table description

越来越多的研究证据表明，小农户农场在全球粮食生产中贡献显著，但目前仍缺乏农业田块面积的空间显式数据。利用遥感技术自动划定田块面积，或是利用普查数据估算次国家层面的平均农场规模，是目前已采用的两种方法，但二者均存在局限——例如遥感数据的地理覆盖范围有限，或是使用普查数据时空间分辨率较粗糙。本研究展示了另一种通过众包方式量化并绘制全球田块面积地图的方法。2017年6月开展了一项众包活动，参与者通过Geo-Wiki应用，目视解译谷歌地图（Google Maps）和必应（Bing）提供的超高分辨率卫星影像。活动期间，参与者为全球13万个独特点位收集了田块面积数据。基于该样本，我们生成了相较于此前版本的优化版全球田块面积地图，并估算了全球、大洲及国家尺度下农业区域内不同规模（从超小型到超大型）田块的占比。研究结果显示，小农户农场占农业区域总面积的比例不超过40%，这意味着相较于当前全球预估的12%，实际小农户农场的数量可能要多得多。全球田块面积地图与众包数据集均公开可获取，可用于综合评估建模、不同情境下农业动态的比较研究，以及助力可持续发展目标2（SDG 2）等诸多研究与应用领域。

本数据集（global field sizes.zip）包含以下内容：
- 优势田块面积分布图（dominant_field_size_categories.tif）及图例说明文件（legend_items.txt）
- 参与者提交的所有有效数据表（仅涵盖完成10次以上分类任务的参与者）及表结构说明
- 所有参与者的质量评分表及表结构说明
- 各点位估算优势田块面积数据表及表结构说明