Paul Keryn I, Larmour John S, Roxburgh Stephen H, England Jacqueline R, Davies Micah J, Luck Hamish D
CSIRO Agriculture and CSIRO Land and Water, GPO Box 1700, Canberra, ACT, 2601, Australia.
CSIRO Agriculture and CSIRO Land and Water, Private Bag 10, Clayton South, VIC, 3169, Australia.
Environ Monit Assess. 2017 Aug;189(8):416. doi: 10.1007/s10661-017-6109-x. Epub 2017 Jul 26.
Stem diameter is one of the most common measurements made to assess the growth of woody vegetation, and the commercial and environmental benefits that it provides (e.g. wood or biomass products, carbon sequestration, landscape remediation). Yet inconsistency in its measurement is a continuing source of error in estimates of stand-scale measures such as basal area, biomass, and volume. Here we assessed errors in stem diameter measurement through repeated measurements of individual trees and shrubs of varying size and form (i.e. single- and multi-stemmed) across a range of contrasting stands, from complex mixed-species plantings to commercial single-species plantations. We compared a standard diameter tape with a Stepped Diameter Gauge (SDG) for time efficiency and measurement error. Measurement errors in diameter were slightly (but significantly) influenced by size and form of the tree or shrub, and stem height at which the measurement was made. Compared to standard tape measurement, the mean systematic error with SDG measurement was only -0.17 cm, but varied between -0.10 and -0.52 cm. Similarly, random error was relatively large, with standard deviations (and percentage coefficients of variation) averaging only 0.36 cm (and 3.8%), but varying between 0.14 and 0.61 cm (and 1.9 and 7.1%). However, at the stand scale, sampling errors (i.e. how well individual trees or shrubs selected for measurement of diameter represented the true stand population in terms of the average and distribution of diameter) generally had at least a tenfold greater influence on random errors in basal area estimates than errors in diameter measurements. This supports the use of diameter measurement tools that have high efficiency, such as the SDG. Use of the SDG almost halved the time required for measurements compared to the diameter tape. Based on these findings, recommendations include the following: (i) use of a tape to maximise accuracy when developing allometric models, or when monitoring relatively small changes in permanent sample plots (e.g. National Forest Inventories), noting that care is required in irregular-shaped, large-single-stemmed individuals, and (ii) use of a SDG to maximise efficiency when using inventory methods to assess basal area, and hence biomass or wood volume, at the stand scale (i.e. in studies of impacts of management or site quality) where there are budgetary constraints, noting the importance of sufficient sample sizes to ensure that the population sampled represents the true population.
树干直径是评估木本植物生长情况以及其所带来的商业和环境效益（如木材或生物质产品、碳固存、景观修复）时最常用的测量指标之一。然而，其测量的不一致性仍是林分尺度测量（如断面积、生物量和材积）估计中持续存在的误差来源。在此，我们通过对一系列不同林分（从复杂的混交种植到商业单一种植园）中不同大小和形态（即单干和多干）的树木和灌木进行重复测量，评估了树干直径测量中的误差。我们比较了标准直径卷尺和阶梯直径测量仪（SDG）在时间效率和测量误差方面的表现。直径测量误差受树木或灌木的大小、形态以及测量时的树干高度的影响较小（但显著）。与标准卷尺测量相比，SDG测量的平均系统误差仅为 -0.17厘米，但在 -0.10至 -0.52厘米之间变化。同样，随机误差相对较大，标准差（以及变异系数百分比）平均仅为0.36厘米（和3.8%），但在0.14至0.61厘米之间（和1.9至7.1%）变化。然而，在林分尺度上，抽样误差（即就直径的平均值和分布而言，为测量直径而选择的单个树木或灌木对真实林分总体的代表性如何）通常对断面积估计中的随机误差的影响比对直径测量误差的影响至少大十倍。这支持使用高效的直径测量工具，如SDG。与直径卷尺相比，使用SDG几乎将测量所需时间减半。基于这些发现，建议如下：（i）在建立异速生长模型时，或在监测永久样地（如国家森林资源清查）中相对较小的变化时，使用卷尺以最大限度地提高准确性，需注意对不规则形状、大单干个体要格外小心；（ii）在林分尺度（即在管理影响或立地质量研究中）使用清查方法评估断面积，进而评估生物量或木材体积时，若存在预算限制，使用SDG以最大限度地提高效率，需注意足够样本量的重要性，以确保所抽样的总体代表真实总体。
