quadratAnalysis

描述

🌐 Description

样方法将一组等大小的区域（样方）覆盖在研究区域上，并统计每个样方中的特性数量，进而创建一个频率表。该表列出了不包含任何特性的样方数、包含一个特性的样方数、包含两个特性的样方数，依此类推，直到包含最多特性的样方。然后，该方法基于随机分布（通常是泊松分布）创建频率表。该方法使用该分布计算0个特性出现、1个特性出现、2个特性出现等等的概率，并将这些概率列在频率表中。通过比较这两个频率表，可以判断特性是否形成某种模式。如果观测分布的表中包含许多特性的样方数比随机分布的表多，那么这些特性就形成了聚集模式。

🌐 Quadrat analysis lays a set of equal-size areas(quadrat) over the study area and counts the number of features in each quadrat and creates a frequency table. The table lists the number of quadrats containing no features, the number containing one feature, two features, and so on, all the way up to the quadrat containing the most features. The method then creates the frequency table for the random distribution, usually based on a Poisson distribution. The method uses the distribution to calculate the probability for 0 feature occuring, 1 feature occuring, 2 features, and so on, and lists these probabilities in the frequency table. By comparing the two frequency tables, you can see whether the features create a pattern. If the table for the observed distribution has more quadrats containing many features than the table for the random distribution dose, then the features create a clustered pattern.

仅仅通过观察，很难判断频率表是相似还是不同。因此，我们可以使用几种统计检验方法来找出频率表的差异程度。我们使用科尔莫哥罗夫-斯米尔诺夫检验。此方法计算两个分布的累积概率，然后比较每个类别水平的累积概率，并选取最大的绝对差值D。随后，检验将D与你指定置信水平的临界值进行比较。如果D大于临界值，则观察到的分布与随机分布之间的差异显著。数值越大，差异越大。

🌐 It is hard to judge the frequency tables are similar or different just by looking at them. So, we can use serval statistical tests to find out how much the frequency tables differ. We use Kolmogorov-Smirnov test.This method calculates cumulative probabilities for both distributions, and then compares the cumulative probabilities at each class level and selects the largest absolute difference D. Then, the test compares D to the critical value for a confidence level you specify. If D is greater than the critical value, the difference between the observed distribution and the random distribution is significant. The greater the value the bigger the difference.

传统上，方形用于样方法中的样方形状，排列成规则网格（方形网格）。一些研究人员建议样方的大小等于每个特性平均面积的两倍，这只是研究区域的面积除以特性的数量。

🌐 Traditionally, squares are used for the shape of the quadrats, in a regular grid(square-grid). Some researchers suggest that the quadrat size equal twice the size of mean area per feature, which is simply the area of the study area divided by the number of features.

参数

🌐 Parameters

名称	类型	描述
pointFeatureSet	FeatureCollection<Point>	要研究的点集
options?	Object	可选参数 (默认 {})
options.studyBbox?	[number, number, number, number	表示研究区域的边界框
options.confidenceLevel?	20 | 15 | 10 | 5 | 2 | 1	置信水平。单位为百分比。5 表示 95%，取值必须在 K_TABLE 中 (默认 20)

返回

🌐 Returns

QuadratAnalysisResult result

示例

🌐 Examples

var bbox = [-65, 40, -63, 42];
var dataset = turf.randomPoint(100, { bbox: bbox });
var result = turf.quadratAnalysis(dataset);

安装

🌐 Installation

$ npm install @turf/quadrat-analysis

import { quadratAnalysis } from "@turf/quadrat-analysis";
const result = quadratAnalysis(...);

$ npm install @turf/turf

import * as turf from "@turf/turf";
const result = turf.quadratAnalysis(...);