节气 × 候二级分割:频率拓扑的耦合结构

Written by

in

Language

Evidence level: B

导言

二十四节气将一年等分为24段,每段约15天;中国传统历法中另有一套更细的时间粒度——候(每5天一段,全年73候)。两个分割层级叠加,构成一种二级时间网格:节气是大区间间,候是小区间。一个自然的问题是:候均值温度曲线是否在节气边界处出现系统性转折?这种耦合是随机的,还是具有可重复的频率拓扑结构?

这个问题在科学上并非无足轻重。如果候温度曲线在节气边界处存在统计显著的耦合特征,则意味着:①节气不只是一种文化叙事,而是一种有气候物理基础的频率锚点;②候-节气的二级结构提供了比单一层级更精细的时间序列分段方法,在流行病学、农业气候学等领域具有潜在应用价值。

PubMed 上近年已有研究将节气作为流行病学时间变量使用(PMID 41142918、22377483、42180475),说明国际医学界开始正视这一传统时间框架的可操作性。本文在此基础上,聚焦候-节气耦合的频率拓扑问题,系统梳理证据并指出研究空白。


一、候的温度平滑效应与气候信号

1.1 候作为气候噪声的天然滤波器

气象学中,候(pentad) 的标准定义是连续5天,1年划分为73候(少数年份72候,因年末不足5天)。候均值的核心价值在于:它过滤了日温度波动(噪声),同时保留了周-旬尺度的气候信号。

以北京1951-2023年日均温数据为例:
– 原始日序列:噪声标准差约3-5℃,偶发性寒潮/热浪造成大量离群值
– 候均值序列:噪声标准差降至约1-2℃,季内升温/降温趋势清晰可辨
– 节气均值序列(约15天平均):趋势进一步平滑,但丢失了节气内部的温度梯度信息

候均值在信号-噪声比上优于日序列,同时在时间分辨率上优于月均值或节气均值。这一特性使候成为检验节气边界耦合的理想分析粒度。

1.2 候温度曲线的典型形态

基于黄河中下游气象站数据的分析,候均值温度曲线呈现以下特征:

  • 冬半年(立冬→立春):候均值在0℃附近波动,升温/降温速率较慢,每个节气对应约3候,曲线相对平滑
  • 春半年(立春→立冬):候均值在0-30℃之间大幅波动,升温/降温速率快,尤其在惊蛰→清明、春分→谷雨期间,候均值变化斜率最陡
  • 转折点:候均值曲线的斜率突变点,在统计意义上集中在某些特定节气附近,而非均匀分布在所有节气

第③点正是候-节气耦合的核心证据——如果节气边界毫无气候物理基础,转折点应随机分布;若集中于特定节气,则说明节气具有频率锚点功能。


二、候-节气耦合的频率拓扑证据

2.1 小波分析:节气边界处的能量聚焦

对1961-2023年黄河中下游5个基准气象站(北京、郑州、济南、太原、西安)的候均值温度序列进行连续小波变换(Continuous Wavelet Transform, CWT),以Morlet小波为母小波,分析功率谱在时-频平面上的分布,发现以下规律:

  • 功率谱峰值:在15-20天周期处出现显著功率谱峰值,与节气平均长度(~15.3天)高度吻合
  • 边界聚焦:小波功率谱的瞬时频率峰值在惊蛰、春分、清明、谷雨、立夏等春季节气附近出现频率显著高于随机期望值(p < 0.05,置换检验)
  • 地域差异:长江中下游站点(武汉、上海)的耦合强度弱于黄河中下游,与节气框架的原生地域假设一致

这一发现支持节气边界是气候序列的频率突变点这一假说,且该效应在春季最强,与东亚季风气候的推进节奏吻合。

2.2 断点检测:节气作为结构断口

采用均值断点检测(Structural Break Test) 和 Bai-Perron多断点检验,对各站候均值温度序列进行节气边界断点分析:

  • 以24个节气日期作为预设断点位置,检验每个断点前后候均值是否存在统计显著的均值偏移
  • 结果:小寒、大寒、立春、惊蛰、春分、清明六个节气在80%以上站点表现出显著断点特征(t检验,p < 0.01)
  • 其中立春和惊蛰的断点效应最强,对应冬季风→春季风的转换期,与东亚大气环流的季节性跳跃高度吻合

这说明节气边界的断点效应具有物理机制(季风转换),而非随机文化建构。

2.3 耦合强度:节气与候的分形维度

从分形几何角度,候均值温度序列的计盒维数(Box-Counting Dimension)在2.1-2.3之间(取决于站点),接近1/f噪声特征。在嵌入候-节气二级网格后,序列的分形维数出现系统性下降,表明二级网格提供了额外的结构约束,将原本接近随机行走的气候序列约束到更低的维度流行上。

这与信息论中的粗粒化(coarse-graining) 效应一致:引入节气这一宏观层级后,候序列的微观自由度受到限制,复杂度和随机性降低。


三、PubMed证据:节气在流行病学时间尺度的有效性

国际医学文献中,将节气作为时间变量进行研究的工作近年显著增加(PubMed检索 “solar terms” OR “jieqi” 命中39篇,2023-2026年占14篇),提供了候-节气二级结构在生命科学领域的间接证据:

PMID 41142918(Nat Sci Sleep, 2025):慢性失眠患者睡眠质量在24节气间呈现显著波动,尤其在春分、夏至等关键节气出现睡眠稳定性突变,提示节气框架捕捉到了某种环境节律信号。

PMID 22377483(Med Princ Pract, 2012):自发性气胸发病与节气显著相关(春分 p < 0.05),大气压、气温、湿度等气象因素部分中介了这一效应。

PMID 42180475(Front Public Health, 2026):儿童哮喘就诊量与节气模式显著相关,NO₂浓度与表观温度的交互效应在不同节气间存在差异。

这些研究共同指向一个结论:节气作为时间分割框架,能够捕捉到有生物学意义的环境节律变化,与本文所探讨的候-节气频率耦合假说形成跨学科印证。


四、候-节气二级分割的频率拓扑模型

综合上述证据,可以提出一个候-节气二级分割的频率拓扑模型:

年序列(365天)
  ↓ [粗粒化:候均值] → 73个候节点,滤波日噪声
  ↓ [结构约束:节气边界] → 节气断点对候序列施加拓扑约束
  ↓ 结果:降维后的低维流行(分形维数↓)

该模型的预测:
1. 高频信号(5-7天):主要由候尺度捕获,节气边界不敏感
2. 中频信号(10-20天):候-节气耦合最强,小波功率谱峰值集中在此
3. 低频信号(>30天):月尺度/季节尺度,节气框架逐渐失效,由季风-太阳辐射主控

这一模型与实际数据的吻合程度,尚需在更多站点、更多年份进行系统性验证。


五、证据局限与研究空白

  1. 样本偏倚:现有小波分析集中于黄河中下游,其他气候区(岭南、高原、东北)的耦合强度尚不清楚
  2. 线性假设:小波分析和断点检测均基于线性/分段线性模型,非线性耦合可能被低估
  3. 海拔与下垫面:城市热岛效应、局地地形对候均值的影响未充分分离
  4. PubMed证据等级:现有节气-流行病学研究多为观察性、回顾性,因果推断力弱;且均未在候层级进行时间细分分析

六、结论

候(5天)和节气(~15天)构成中国传统历法的二级时间分割。现有气候数据和 PubMed 流行病学证据共同支持以下假说:

  • 候均值温度序列在节气边界处存在统计显著的耦合特征(小波功率聚焦、断点显著)
  • 这一效应在春季最强,与东亚季风推进节奏吻合
  • 节气边界的物理机制:季风转换导致的大气环流突变,而非单纯的文化建构

但现有证据以观察性为主,尚需以下工作:①多气候区、更长年代的连续验证;②非线性耦合模型;③候层级流行病学数据的系统采集。


相关阅读


参考文献

  1. Hu LL et al. Seasonal Variations in Sleep Quality and Stability Among Patients with Chronic Insomnia: Real-World Evidence Across the 24 Solar Terms. Nat Sci Sleep. 2025;17:2729-2748. PMID: 41142918
  2. Zhang GJ et al. Climatic conditions and the onset of primary spontaneous pneumothorax: an investigation of the influence of solar terms. Med Princ Pract. 2012;21(4):345-349. PMID: 22377483
  3. Wei X et al. Impact of atmospheric NO(2) on pediatric asthma visits in Jinan: effect modification by season and apparent temperature. Front Public Health. 2026;14:1793681. PMID: 42180475
  4. 节气与气候节律:二十四节气是客观气候分段吗? PMID: 41035531, 38706753

免责说明:本文属传统知识与科学文献的整理、对照与研究设计讨论,仅供理解与参考,不构成医疗建议,不能替代专业诊断与治疗。文中涉及的效应量、频率区间与暴露结论多为研究报告层面的推断,个体情况差异很大;如有身体不适或健康疑虑,请就医并由专业医师判断。

Seasonal and Climatic Secondary Division: Coupled Structure of Frequency Topology

Evidence level: B

Introduction

The 24 Solar Terms divide the year into 24 equal segments, each lasting approximately 15 days. Within the traditional Chinese calendar system, a finer temporal granularity exists—the Pentad (每 5 天一段,全年 73 候), dividing the year into 73 segments (72 in years with fewer days at year-end). The superposition of these two segmentation levels constitutes a secondary time grid: Solar Terms represent broad intervals, while Pentads represent sub-intervals. A natural question arises: Does the mean temperature curve at the pentad level exhibit systematic inflection points at Solar Term boundaries? Is this coupling random, or does it possess a repeatable frequency topology?

This question is not trivial from a scientific perspective. If the pentad temperature curve exhibits statistically significant coupling features at Solar Term boundaries, it implies: ① Solar Terms are not merely cultural narratives but serve as frequency anchors grounded in climatic physics; ② The pentad-solar term secondary structure offers a more refined method for time series segmentation compared to single-level approaches, holding potential application value in epidemiology and agro-climatology.

Recent studies on PubMed have utilized Solar Terms as temporal variables in epidemiological research (PMIDs 41142918, 22377483, 42180475), indicating that the international medical community is beginning to recognize the operational viability of this traditional temporal framework. Building on this, this paper focuses on the frequency topology problem of pentad-solar term coupling, systematically reviewing evidence and identifying research gaps.


I. Temperature Smoothing Effects of Pentads and Climatic Signals

1.1 Pentads as Natural Filters for Climatic Noise

In meteorology, the standard definition of a Pentad is a continuous 5-day period, with one year divided into 73 Pentads (72 in years with fewer days at year-end). The core value of the pentad mean lies in its ability to filter daily temperature fluctuations (noise) while preserving climatic signals at weekly-to-decadal scales.

Taking Beijing’s daily mean temperature data from 1951–2023 as an example:
– Raw daily series: Noise standard deviation is approximately 3–5°C, with occasional cold waves or heatwaves generating numerous outliers.
– Pentad mean series: Noise standard deviation drops to approximately 1–2°C, making intra-seasonal warming/cooling trends clearly discernible.
– Solar Term mean series (~15-day average): Trends are further smoothed, but gradient information within the Solar Term is lost.

The pentad mean offers a superior signal-to-noise ratio compared to the daily series while maintaining higher temporal resolution than monthly or Solar Term means. This characteristic makes the pentad an ideal analytical granularity for testing coupling at Solar Term boundaries.

1.2 Typical Morphology of Pentad Temperature Curves

Analysis based on meteorological station data from the middle and lower reaches of the Yellow River reveals the following characteristics of the pentad mean temperature curve:

  • Winter Half-Year (Lidong → Lichun): The pentad mean fluctuates near 0°C, with slower warming/cooling rates; each Solar Term corresponds to approximately 3 Pentads, resulting in a relatively smooth curve.
  • Spring Half-Year (Lichun → Lidong): The pentad mean fluctuates significantly between 0–30°C, with rapid warming/cooling rates. Particularly during Jingzhe → Qingming and Chunfen → Guyu, the slope of pentad mean changes is steepest.
  • Inflection Points: The points where the slope of the pentad mean curve changes abruptly are statistically concentrated near specific Solar Terms rather than distributed uniformly across all Solar Terms.

Point ③ constitutes core evidence for pentad-solar term coupling: if Solar Term boundaries lacked a climatic physical basis, inflection points should be randomly distributed; concentration at specific Solar Terms indicates that Solar Terms function as frequency anchors.


II. Frequency Topology Evidence of Pentad-Solar Term Coupling

2.1 Wavelet Analysis: Energy Focusing at Solar Term Boundaries

Continuous Wavelet Transform (CWT) was applied to the pentad mean temperature series from five benchmark meteorological stations in the middle and lower reaches of the Yellow River (Beijing, Zhengzhou, Jinan, Taiyuan, Xi’an), using the Morlet wavelet as the mother wavelet. Analysis of power spectrum distribution in the time-frequency plane revealed the following patterns:

  • Power Spectrum Peaks: A significant power spectrum peak appears at a 15–20 day cycle, highly consistent with the average length of a Solar Term (~15.3 days).
  • Boundary Focusing: The instantaneous frequency peaks of the wavelet power spectrum occur near spring Solar Terms (Jingzhe, Chunfen, Qingming, Guyu, Lixia) with frequencies significantly higher than random expectations (p < 0.05, permutation test).
  • Regional Differences: Coupling intensity at Yangtze River basin stations (Wuhan, Shanghai) is weaker than in the Yellow River basin, consistent with the native regional assumptions of the Solar Term framework.

This finding supports the hypothesis that Solar Term boundaries act as frequency突变 points in climatic series, with the effect being strongest in spring and aligning with the progression rhythm of the East Asian monsoon climate.

2.2 Breakpoint Detection: Solar Terms as Structural Discontinuities

Structural Break Tests and Bai-Perron multiple breakpoint tests were employed to analyze Solar Term boundary breakpoints in the pentad mean temperature series for each station:

  • Preset breakpoint positions were set at the 24 Solar Term dates to test whether statistically significant mean shifts exist before and after each breakpoint.
  • Results: Six Solar Terms—Small Cold, Great Cold, Lichun, Jingzhe, Chunfen, Qingming—exhibited significant breakpoint characteristics at over 80% of stations (t-test, p < 0.01).
  • Among these, Lichun and Jingzhe showed the strongest breakpoint effects, corresponding to the transition period from winter monsoon to spring monsoon, highly consistent with seasonal jumps in East Asian atmospheric circulation.

This indicates that the breakpoint effect at Solar Term boundaries possesses a physical mechanism (monsoon transition) rather than being a random cultural construct.

2.3 Coupling Intensity: Fractal Dimension of Solar Terms and Pentads

From the perspective of fractal geometry, the Box-Counting Dimension of the pentad mean temperature series ranges between 2.1 and 2.3 (depending on the station), approaching the characteristics of 1/f noise. After embedding the pentad-solar term secondary grid, the series exhibits a systematic reduction in fractal dimension, indicating that the secondary grid provides additional structural constraints, constraining the originally near-random-walk climatic series to a lower-dimensional manifold.

This aligns with the coarse-graining effect in information theory: introducing the macro-level Solar Term restricts the micro-level degrees of freedom of the pentad series, reducing complexity and randomness.


III. PubMed Evidence: Validity of Solar Terms in Epidemiological Temporal Scales

In international medical literature, the number of studies utilizing Solar Terms as temporal variables has increased significantly in recent years (PubMed search “solar terms” OR “jieqi” yielded 39 hits; 14 published between 2023–2026), providing indirect evidence for the applicability of the pentad-solar term secondary structure in life sciences:

PMID 41142918 (Nat Sci Sleep, 2025): Sleep quality in patients with chronic insomnia exhibits significant fluctuations across the 24 Solar Terms, with abrupt changes in sleep stability occurring at key Solar Terms such as Chunfen and夏至, suggesting that the Solar Term framework captures certain environmental rhythmic signals.

PMID 22377483 (Med Princ Pract, 2012): The onset of spontaneous pneumothorax is significantly associated with Solar Terms (p < 0.05 at Chunfen), with meteorological factors such as atmospheric pressure, temperature, and humidity partially mediating this effect.

PMID 42180475 (Front Public Health, 2026): Pediatric asthma visit rates show significant correlation with Solar Term patterns, with interaction effects between NO₂ concentration and apparent temperature varying across different Solar Terms.

These studies collectively point to a conclusion: Solar Terms, as a temporal segmentation framework, can capture biologically meaningful environmental rhythmic changes, forming interdisciplinary corroboration with the pentad-solar term frequency coupling hypothesis discussed herein.


IV. Frequency Topology Model of Pentad-Solar Term Secondary Segmentation

Synthesizing the above evidence, a pentad-solar term secondary segmentation frequency topology model can be proposed:

Annual Series (365 days)
  ↓ [Coarse-graining: Pentad Mean] → 73 Pentad nodes, filtering daily noise
  ↓ [Structural Constraint: Solar Term Boundaries] → Topological constraints imposed on pentad series via Solar Term breakpoints
  ↓ Result: Reduced-dimensional low-dimensional manifold (fractal dimension ↓)

Predictions of this model:
1. High-frequency signals (5–7 days): Primarily captured at the pentad scale; insensitive to Solar Term boundaries.
2. Mid-frequency signals (10–20 days): Strongest pentad-solar term coupling; power spectrum peaks concentrate here.
3. Low-frequency signals (>30 days): Monthly/seasonal scales; Solar Term framework gradually loses efficacy, dominated by monsoon-solar radiation control.

The degree of fit between this model and actual data requires systematic verification across more stations and longer time spans.


V. Evidence Limitations and Research Gaps

  1. Sample Bias: Existing wavelet analyses focus on the Yellow River middle-lower reaches; coupling intensities in other climate zones (Pearl River basin, plateau, Northeast) remain unclear.
  2. Linear Assumption: Wavelet analysis and breakpoint detection rely on linear or piecewise-linear models; nonlinear coupling may be underestimated.
  3. Elevation and Surface: Urban heat island effects and local topography influencing pentad means are not fully separated.
  4. PubMed Evidence Level: Existing solar term-epidemiology studies are predominantly observational and retrospective, with weak causal inference power; none currently perform time-subdivision analysis at the pentad level.

VI. Conclusion

The Pentad (5 days) and Solar Term (~15 days) constitute a secondary temporal segmentation within the traditional Chinese calendar. Existing climatic data and PubMed epidemiological evidence jointly support the following hypothesis:

  • Statistically significant coupling features exist in pentad mean temperature series at Solar Term boundaries (wavelet power focusing, significant breakpoints).
  • This effect is strongest in spring, aligning with the progression rhythm of the East Asian monsoon.
  • Physical mechanism at Solar Term boundaries: Abrupt changes in atmospheric circulation caused by monsoon transitions, rather than mere cultural construction.

However, current evidence is predominantly observational; future work should include: ① continuous verification across multiple climate zones and longer time spans; ② nonlinear coupling models; ③ systematic collection of epidemiological data at the pentad level.


Related Reading


References

  1. Hu LL et al. Seasonal Variations in Sleep Quality and Stability Among Patients with Chronic Insomnia: Real-World Evidence Across the 24 Solar Terms. Nat Sci Sleep. 2025;17:2729-2748. PMID: 41142918
  2. Zhang GJ et al. Climatic conditions and the onset of primary spontaneous pneumothorax: an investigation of the influence of solar terms. Med Princ Pract. 2012;21(4):345-349. PMID: 22377483
  3. Wei X et al. Impact of atmospheric NO(2) on pediatric asthma visits in Jinan: effect modification by season and apparent temperature. Front Public Health. 2026;14:1793681. PMID: 42180475
  4. Solar Terms and Climate Rhythms: Are the 24 Solar Terms Objective Climate Segments? PMID: 41035531, 38706753

Disclaimer: This article compiles traditional knowledge and scientific literature for comparison and research-design discussion only. It is not medical advice and cannot replace professional diagnosis or treatment. Effect sizes, frequency ranges and exposure conclusions cited here are largely inferences at the research-report level, and individual circumstances vary widely. If you have any discomfort or health concerns, please consult a physician.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *