Urban Climate: A Statistical Approach | 城市气候：统计分析方法

📚 Urban Climate: A Statistical Approach | 城市气候：统计分析方法

Urban climate presents a fascinating context for applying A-Level statistical techniques. Using real meteorological data from the Edexcel Large Data Set, students can investigate the urban heat island effect, compare temperature distributions between city and rural locations, and explore relationships among weather variables such as temperature, humidity, and rainfall. This article demonstrates how descriptive statistics, hypothesis testing, correlation, and regression can be used to analyse urban climate data, providing both conceptual understanding and exam-relevant practice.

城市气候为应用 A-Level 统计方法提供了引人入胜的情境。借助 Edexcel 大型数据集中的真实气象数据，学生可以探究城市热岛效应，比较城市与乡村站点的温度分布，并研究温度、湿度和降雨等天气变量之间的关系。本文展示如何利用描述性统计、假设检验、相关分析和回归分析来研究城市气候数据，既有概念理解，也提供与考试相关的练习。

1. Urban Climate and the Large Data Set | 城市气候与大数据集介绍

The Edexcel A-Level Mathematics specification includes a Large Data Set (LDS) comprising daily weather records from five UK Met Office stations for the years 1987 and 2015. Two of these stations — Heathrow and Leeming — are often used to illustrate urban climate effects, with Heathrow representing a heavily urbanised area (London) and stations such as Camborne or Leuchars serving as rural or coastal comparisons. An understanding of the geographical context helps in formulating hypotheses about urban temperature anomalies.

Edexcel A-Level 数学考试大纲包含一个大型数据集（LDS），该数据集由英国气象局五个站点 1987 年和 2015 年的每日天气记录组成。其中两个站点——希思罗（Heathrow）和利明（Leeming）——常被用来阐释城市气候效应，希思罗代表高度城市化的伦敦地区，而像坎伯恩（Camborne）或卢赫斯（Leuchars）这样的站点则作为乡村或沿海对照。了解这一地理背景有助于提出关于城市温度异常的假设。

2. The Variables: Temperature, Rainfall, Sunshine, Humidity | 变量：温度、降雨量、日照、湿度

The LDS contains the following daily variables: mean temperature (°C), total rainfall (mm), total sunshine duration (hours), maximum relative humidity (%), and wind speed (knots). For urban climate studies, the mean temperature and the derived daily maximum and minimum temperatures are particularly useful. High urban temperatures combined with modified humidity and rainfall patterns can be explored statistically. It is important to know the units and measurement precision, as exam questions often require correct handling of rounding and class boundaries.

大型数据集包含以下每日变量：平均温度（°C）、总降雨量（mm）、总日照时长（小时）、最大相对湿度（%）和风速（节）。对于城市气候研究而言，平均温度以及由此衍生的每日最高温度和最低温度尤其有用。城市地区的高温，以及与之相关的湿度和降雨模式变化，都可以用统计方法进行探索。了解这些变量的单位和测量精度非常重要，因为考试题目常常要求正确处理四舍五入和组距边界。

3. Extracting and Cleaning Meteorological Data | 气象数据的提取与清洗

Before any analysis can take place, the raw data must be filtered by station, year, and season. For an urban climate investigation, one may extract Heathrow's 2015 summer months (June, July, August) and compare them with Camborne's data for the same period. Cleaning involves checking for missing values — denoted by 'tr' (trace) for rainfall or 'n/a' — and deciding whether to remove those days or use interpolation. In an exam context, students are often given pre-cleaned summary statistics, but understanding the process aids interpretation.

在进行任何分析之前，必须先按站点、年份和季节对原始数据进行筛选。对于城市气候研究，可以提取希思罗 2015 年夏季月份（6 月、7 月、8 月）的数据，并与坎伯恩同期的数据进行比较。数据清洗包括检查缺失值——降雨量中可能标为“tr”（微量）或“n/a”——并决定是删除这些记录还是采用插值处理。在考试情境中，学生通常会得到已清洗的汇总统计量，但理解这一过程有助于正确解读。

4. Descriptive Statistics: Central Tendency and Spread | 描述性统计：集中趋势与离散程度

For a selected urban station, calculate the mean, median, and mode of daily mean temperature. The sample mean is given by x̄ = Σx / n. Compare the mean with the median to assess skewness: if mean > median, the distribution is positively skewed, possibly due to a few exceptionally hot days. Measures of spread — range, interquartile range (IQR), variance and standard deviation — quantify the variability. A larger standard deviation in an urban station might indicate more extreme temperature fluctuations due to heat-retaining surfaces.

对选定的城市站点，计算日平均温度的均值、中位数和众数。样本均值的计算公式为 x̄ = Σx / n。比较均值和中位数可评估偏态：若均值大于中位数，则分布呈正偏态，可能是由于少数异常炎热的日子所致。离散程度的度量——极差、四分位距（IQR）、方差和标准差——用于量化变异性。城市站点的标准差较大，可能意味着由于蓄热表面导致温度波动更极端。

s = √[ Σ(x − x̄)² / (n − 1) ]

样本标准差 s = √[ Σ(x − x̄)² / (n − 1) ]

5. Box Plots and Outlier Identification | 箱线图与异常值识别

sing the five-number summary: minimum, Q1, median, Q3, maximum. Outliers are usually defined as observations below Q1 − 1.5×IQR or above Q3 + 1.5×IQR. An urban site may display a higher median and a compressed IQR due to the consistent retention of heat overnight, whereas a rural site might show a lower median and greater diurnal range, leading to more outliers at the lower end.

箱线图是比较城市与乡村温度分布的上佳工具。利用五数概括（最小值、第一四分位数 Q1、中位数、第三四分位数 Q3 和最大值）为希思罗和坎伯恩绘制并列箱线图。异常值通常定义为低于 Q1 − 1.5×IQR 或高于 Q3 + 1.5×IQR 的观测值。城市站点可能显示较高的中位数和压缩的 IQR，这是因为夜间持续保温；而乡村站点可能显示较低的中位数和较大的昼夜温差，从而导致低端出现更多异常值。

6. Comparing Urban and Rural Temperatures: The Two-Sample t-Test | 比较城市与乡村温度：双样本 t 检验

on mean daily temperature for the urban station and μ₂ for the rural station. The null hypothesis is H₀: μ₁ = μ₂, and the alternative is H₁: μ₁ > μ₂ (one-tailed test for urban warming). The two-sample t-statistic is calculated as:

正式的假设检验可以判断观测到的平均温度差异是否具有统计显著性。设 μ₁ 为城市站点日均温度的总体均值，μ₂ 为乡村站点。原假设为 H₀: μ₁ = μ₂，备择假设为 H₁: μ₁ > μ₂（城市增温的单尾检验）。双样本 t 统计量的计算公式为：

t = (x̄₁ − x̄₂) / √s₁²/n₁ + s₂²/n₂

广告 / Sponsored

t = (x̄₁ − x̄₂) / √s₁²/n₁ + s₂²/n₂

The degrees of freedom can be approximated using the Welch-Satterthwaite equation if variances are unequal. At a 5% significance level, comparing the calculated t-value with the critical value from the t-distribution provides evidence for or against the urban heat island effect.

若方差不齐，自由度可用 Welch-Satterthwaite 公式近似计算。在 5% 显著性水平下，将计算出的 t 值与 t 分布的临界值进行比较，即可为城市热岛效应是否存在提供证据。

7. Checking Assumptions: Normality and Equal Variances | 检验假设：正态性与方差齐性

roximately normally distributed and that the two samples have equal variances (for Student's t-test) or does not require the latter (Welch' e to heatwaves, but with large sample sizes (n ≥ 30) the Central Limit Theorem ensures robustness. The F-test for equality of variances compares s₁² and s₂²: F = s₁² / s₂². If the variances differ significantly, Welch's adjustment should be applied.

t 检验要求数据近似服从正态分布，且对于 Student t 检验还需两样本方差相等，而 Welch t 检验则无此要求。正态性可通过直方图、Q-Q 图或 Shapiro-Wilk 检验进行评估。对于城市温度数据，由于热浪等原因常出现轻微偏态，但在大样本（n ≥ 30）下，中心极限定理保证了稳健性。方差齐性的 F 检验用于比较 s₁² 和 s₂²：F = s₁² / s₂²。若方差差异显著，则应采用 Welch 校正。

8. Interpreting the Results: Urban Heat Island Effect | 结果解读：城市热岛效应

If the p-value from the t-test is less than the chosen significance level (e.g., 0.05), we reject H₀ and conclude that there is significant evidence that the urban mean temperature is higher. The magnitude of the difference, along with a confidence interval for μ₁ − μ₂, quantifies the effect. For instance, a 95% confidence interval of (1.2°C, 2.8°C) suggests that the true urban-rural temperature difference likely lies within this range, confirming the urban heat island effect. This result aligns with geographical knowledge: concrete and asphalt store heat and release it slowly, raising nighttime minima.

若 t 检验的 p 值小于选定的显著性水平（例如 0.05），则拒绝 H₀，并得出结论：有显著证据表明城市平均温度更高。差异的大小以及 μ₁ − μ₂ 的置信区间可以量化这一效应。例如，95% 置信区间为（1.2°C, 2.8°C），表明真实的城乡温差很可能落在此范围内，从而证实了城市热岛效应。这一结果与地理知识相符：混凝土和沥青储存热量并缓慢释放，抬高了夜间最低温度。

9. Correlation Between Temperature and Humidity | 温度与湿度的相关性

mmer month, the product moment correlation coefficient (PMCC) can be calculated: r = Sxy / √Sxx Syy, where Sxy = Σ(x − x̄)(y − ȳ s constant. Testing H₀: ρ = 0 against H₁: ρ < 0 using a t-test for correlation helps validate this inverse relationship.

城市气候研究常会考察温度与相对湿度之间的关系。利用夏季月份的配对数据，可以计算积矩相关系数（PMCC）：r = Sxy / √Sxx Syy，其中 Sxy = Σ(x − x̄)(y − ȳ)。由于较暖的空气能容纳更多水分，在绝对含水量不变的情况下相对湿度会降低，因此预期存在负相关。用相关性的 t 检验来检验 H₀: ρ = 0 与 H₁: ρ < 0，可以验证这一反向关系。

10. Rainfall Patterns: Urban vs Rural | 降雨模式：城市与乡村的对比

Urban areas may experience enhanced convectional rainfall due to the heat island lifting air masses, potentially increasing total rainfall compared to surrounding rural areas. Using the LDS, one can compare the mean daily rainfall at Heathrow and Camborne with a two-sample t-test or a Mann-Whitney U test if the data are heavily skewed (rainfall data often contain many zero values). Percentage difference and box plots can illustrate the contrast. Additionally, the frequency of rainfall days above a threshold can be analysed using a chi-squared test for independence.

城市地区可能因热岛效应促使气团抬升而增强对流性降雨，从而导致总降雨量高于周边乡村地区。利用大型数据集，可以对希思罗和坎伯恩的日均降雨量进行双样本 t 检验，或在数据严重偏态时（降雨量数据常包含许多零值）使用 Mann-Whitney U 检验。百分比差异和箱线图可以展示这种对比。此外，超过某一阈值的降雨日频率也可用卡方独立性检验进行分析。

11. Using Regression to Predict Urban Temperatures | 使用回归分析预测城市温度

l temperature. The least squares regression line is y = a + bx, where b = Sxy / Sxx and a = ȳ − bx̄ ication. The coefficient of determination, R², assesses the proportion of variance in urban temperature explained by the rural temperature. Residual plots check for patterns that would violate linear model assumptions.

可以构建一个简单的线性回归模型，用乡村温度预测城市平均温度。设 y 为城市温度，x 为乡村温度。最小二乘回归直线为 y = a + bx，其中 b = Sxy / Sxx，a = ȳ − bx̄。系数 b 表示乡村温度每升高一个单位，城市温度相应的升幅。b > 1 表明存在城市放大效应。决定系数 R² 用于评估乡村温度所解释的城市温度方差的比例。残差图可用于检查违背线性模型假设的模式。

12. Conclusion and Exam Tips | 结论与考试技巧

Investigating urban climate through the Edexcel Large Data Set provides a rich opportunity to apply a wide range of statistical methods. When answering exam questions, always define your hypotheses clearly, state the significance level, and interpret your results in the context of urban climate. Remember to refer to the geographical setting, use correct notation, and check assumptions. The ability to move seamlessly between statistical analysis and real-world interpretation is a key skill assessed in the A-Level Mathematics specification.

通过 Edexcel 大型数据集研究城市气候，为应用多种统计方法提供了丰富的机会。在回答考试题目时，务必清晰地定义假设，说明显著性水平，并结合城市气候的背景来解读结果。要记得提及地理环境，使用正确的符号，并检查假设条件。能够在统计分析与现实解读之间自如转换，是 A-Level 数学考纲评估的一项关键技能。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）