📚 Water, Carbon, Climate and Life on Earth | 水、碳、气候与地球生命
Water, carbon, climate and life on Earth form an interconnected system that can be studied through the lens of A-Level Mathematics. By using differential equations, statistical methods and optimisation techniques, we can model the flows, reservoirs and feedbacks that keep our planet habitable.
水、碳、气候与地球生命构成一个相互关联的系统，可以通过 A-Level 数学的视角来研究。利用微分方程、统计方法和最优化技术，我们可以模拟维持地球宜居性的流动、储库和反馈过程。
This article builds a bridge between the pure mechanics of calculus, statistics and optimisation and the real planetary system they describe. Every equation used here is one that appears on the A-Level specification, but each one is attached to a physical meaning: a reservoir of water, a flux of carbon, a temperature that responds to an energy imbalance.
本文将微积分、统计学与最优化的纯技术内容，与它们所描述的真实地球系统连接起来。这里使用的每一个方程都出现在 A-Level 考纲中，但每一个方程都被赋予了物理含义：一个水的储库、一股碳的通量、一个对能量失衡作出响应的温度。
1. The Earth as a coupled system | 地球作为一个耦合系统
The Earth is not a collection of independent parts but a coupled system in which the hydrosphere, atmosphere, biosphere and geosphere exchange energy and matter. In mathematical terms, a coupled system is described by a set of simultaneous differential equations, where the rate of change of one variable depends on the current values of the others.
地球并不是独立部分的集合，而是一个耦合系统，其中水圈、大气圈、生物圈和岩石圈交换能量与物质。用数学术语来说，耦合系统由一组联立微分方程描述，其中一个变量的变化率取决于其他变量的当前值。
For example, if C(t) represents atmospheric carbon content and T(t) represents global temperature, a simple coupled model can be written as the pair of equations below. Solving this kind of system allows us to predict how a change in one component, such as a rise in CO₂, will propagate through the whole Earth system.
例如，若 C(t) 表示大气碳含量，T(t) 表示全球温度，一个简单的耦合模型可以写作下面这一对方程。求解这类方程组使我们能够预测某一组分的变化（如 CO₂ 上升）如何在整个地球系统中传播。
dC/dt = F(T) − kC dT/dt = G(C) − λT
n matrix form the system is written as x′ = Ax + b, and the eigenvalues of A decide whether a small perturbation decays back to equilibrium or grows without limit.
这里 k 与 λ 是正常数，分别描述碳离开大气的速度以及行星耗散温度异常的速度。写成矩阵形式即 x′ = Ax + b，而 A 的特征值决定了微小扰动是衰减回平衡态，还是无限增长。
2. Modelling the water cycle with differential equations | 用微分方程模拟水循环
The water cycle can be modelled by treating reservoirs such as the ocean, ice sheets, lakes and soil moisture as connected boxes. The basic mass-balance equation for a reservoir of volume V is shown below, where I(t) is the total inflow and O(t) is the total outflow at time t.
水循环可以通过将海洋、冰川、湖泊和土壤水分等储库视为相互连接的箱体来建模。储库体积 V 的基本质量平衡方程为下式，其中 I(t) 是时间 t 时的总流入量，O(t) 是总流出量。
dV/dt = I(t) − O(t)
solution below shows that the reservoir approaches a steady state V* = I/k exponentially, with a characteristic time 1/k.
若流出量与储库体积成正比，即 O = kV，且流入量恒定，方程就是线性的并且容易分离变量。下式解表明储库以指数方式趋近稳态 V* = I/k，其特征时间为 1/k。
V(t) = V* + (V₀ − V*) e−kt
t) gives a forced linear equation whose solution contains a transient term that decays and a periodic term that never dies away. This is exactly the language of simple harmonic forcing met in A-Level mechanics, applied now to rainfall and river discharge.
真实的流入量具有季节性而非常数。令 I(t) = I₀ + A sin(ωt)，得到受迫线性方程，其解包含一个会衰减的瞬态项和一个永不消失的周期项。这正是 A-Level 力学中遇到的简谐受迫语言，现在被应用于降水和河流流量。
3. Residence time, turnover and why the oceans matter | 滞留时间、周转与海洋的重要性
The residence time of a substance in a reservoir is the average time a molecule spends there before leaving. For a reservoir at steady state it is simply the stored amount divided by the flux through it.
某物质在一个储库中的滞留时间，是指一个分子在离开之前平均停留的时间。对于处于稳态的储库，它就是把储存量除以通过它的通量。
τ = V / I
Residence times differ by many orders of magnitude across the Earth system, and this difference is the single most important fact in understanding human impact. A short residence time means a reservoir responds quickly to a change; a long one means the change persists for centuries.
地球系统各部分的滞留时间相差许多个数量级，而这一差异是理解人类影响的最重要事实。滞留时间短意味着储库对变化响应快；滞留时间长则意味着变化会持续数百年。
| Reservoir 储库 | Stored water 储量 | Typical residence time 典型滞留时间 |
| Atmospheric water vapour 大气水汽 | ≈ 1.3 × 10⁴ km³ | ≈ 9 days 约 9 天 |
| Rivers, lakes, soil moisture 河流、湖泊、土壤水 | ≈ 2 × 10⁵ km³ | weeks to years 数周至数年 |
| Ocean 海洋 | ≈ 1.34 × 10⁹ km³ | ≈ 3,000 years 约 3000 年 |
| Ice sheets and glaciers 冰盖与冰川 | ≈ 2.6 × 10⁷ km³ | 10³ to 10⁴ years 10³ 至 10⁴ 年 |
| Groundwater 地下水 | ≈ 2.3 × 10⁷ km³ | 10² to 10⁴ years 10² 至 10⁴ 年 |
The message of the table is that the atmosphere holds only a few days of water, while the deep ocean holds thousands of years of it. A pollutant or a heat anomaly delivered to the deep ocean is therefore effectively stored for a very long time, which is why ocean heat uptake delays but does not cancel surface warming.
这张表传达的信息是：大气只容纳几天的水，而深海容纳数千年的水。因此，进入深海的污染物或热异常实际上被长期封存，这正是海洋吸热会延迟而非抵消地表升温的原因。
4. The carbon cycle: reservoirs, fluxes and feedbacks | 碳循环：储库、通量与反馈
Carbon moves between the atmosphere, ocean, land biosphere and lithosphere. Photosynthesis converts carbon dioxide and water into glucose and oxygen, while respiration and combustion reverse the process.
碳在大气、海洋、陆地生物圈和岩石圈之间迁移。光合作用把二氧化碳和水转化为葡萄糖与氧气，而呼吸作用与燃烧则逆转这一过程。
6CO₂ + 6H₂O → C₆H₁₂O₆ + 6O₂
The ocean absorbs carbon dioxide because it dissolves and then reacts with water in a series of equilibria. These equilibria increase the hydrogen ion concentration, which is why ocean acidification accompanies carbon uptake.
海洋吸收二氧化碳，因为它先溶解，然后与水发生一系列平衡反应。这些平衡提高了氢离子浓度，这正是碳吸收伴随海洋酸化的原因。
CO₂ + H₂O ⇌ H₂CO₃ ⇌ H⁺ + HCO₃⁻
wing permafrost releases stored carbon, so warming adds carbon. Both are positive feedbacks, and both can be represented by making k in the equation dC/dt = F(T) − kC a decreasing function of T.
反馈之所以重要，是因为它们改变速率常数的有效数值。水温升高会降低 CO₂ 的溶解度，因此升温会减少海洋吸收；永久冻土融化会释放封存的碳，因此升温会额外增加碳。两者都是正反馈，都可以通过让方程 dC/dt = F(T) − kC 中的 k 成为 T 的递减函数来表示。
5. Radiative forcing and the planetary energy balance | 辐射强迫与行星能量平衡
The climate system is driven by an energy imbalance between absorbed sunlight and emitted infrared radiation. The absorbed solar power per square metre of surface depends on the solar constant S₀, the planetary albedo α, and the geometry factor of four that converts a disc to a sphere.
气候系统由吸收的太阳光与发射的红外辐射之间的能量失衡驱动。地表每平方米吸收的太阳功率取决于太阳常数 S₀、行星反照率 α，以及把圆盘换算为球面的几何因子 4。
Absorbed 吸收 = S₀(1 − α) / 4 Emitted 发射 = εσT⁴
n constant σ = 5.67 × 10⁻⁸ W m⁻² K⁻⁴. For S₀ = 1361 W m⁻² and α = 0.30 the result is about 255 K, roughly 33 K colder than the observed surface.
令两式相等并对温度求解，可得到行星的有效辐射温度，其中斯特藩-玻尔兹曼常数 σ = 5.67 × 10⁻⁸ W m⁻² K⁻⁴。取 S₀ = 1361 W m⁻²、α = 0.30 时，结果为约 255 K，比观测到的地表温度低约 33 K。
T = ⁴√( S₀(1 − α) / (4σ) )
The difference is provided by the greenhouse effect, and a change in composition produces a radiative forcing. For carbon dioxide the forcing is logarithmic in concentration, which is why each doubling of CO₂ adds roughly the same extra energy as the previous doubling.
这一差值由温室效应提供，而大气成分的变化会产生辐射强迫。对二氧化碳而言，强迫与浓度的对数成正比，这就是为什么 CO₂ 每翻一倍所增加的额外能量大致相同。
ΔF = 5.35 ln(C / C₀)
6. Feedbacks, equilibrium and stability | 反馈、平衡与稳定性
Once a forcing is applied, the temperature continues to change until the outgoing radiation balances it. A simple but powerful model treats the system as a first-order differential equation, where C is the heat capacity of the surface layer and λ is the climate feedback parameter.
一旦施加了强迫，温度会持续变化，直到出射辐射与之平衡。一个简单却有力的模型把系统视为一阶微分方程，其中 C 是表层热容量，λ 是气候反馈参数。
C dT/dt = ΔF − λT T* = ΔF / λ
esponse alone gives λ ≈ 3.2 W m⁻² K⁻¹, but water vapour and ice-albedo feedbacks reduce the effective λ and therefore amplify the final response.
平衡温度变化为 ΔF/λ，达到它的 e 倍时间尺度为 C/λ，对海洋混合层而言通常为几十年。仅普朗克响应给出 λ ≈ 3.2 W m⁻² K⁻¹，但水汽反馈与冰-反照率反馈会降低有效 λ，从而放大最终的响应。
Stability is decided by the sign of the feedback parameter. If λ is positive the equilibrium is stable and any perturbation decays; if a strongly positive feedback makes the effective λ negative, the system can run away, which is the mathematical definition of a tipping point.
稳定性由反馈参数的符号决定。若 λ 为正，平衡是稳定的，任何扰动都会衰减；若某个强正反馈使有效 λ 变为负值，系统就可能失控，这就是临界点（tipping point）的数学定义。
7. Statistics of climate and environmental data | 气候与环境数据的统计
Climate records are noisy, so trends must be extracted statistically rather than seen by eye. A linear trend model writes each observation as a deterministic line plus random error, and least squares chooses the intercept and slope that minimise the sum of squared residuals.
气候记录充满噪声，因此趋势必须通过统计方法提取，而不能仅凭肉眼观察。线性趋势模型把每个观测写成确定性直线加随机误差，最小二乘法选择使残差平方和最小的截距与斜率。
yt = a + bt + εt b̂ = Σ(t − t̄)(y − ȳ) / Σ(t − t̄)²
The slope estimate has a standard error, SE = s / √(Σ(t − t̄ nterval excludes zero, the warming trend is statistically significant at the chosen level.
斜率估计具有标准误 SE = s / √(Σ(t − t̄)²)，从而可以为趋势给出置信区间。若该区间不包含零，则升温趋势在所选显著性水平上统计显著。
Two cautions are essential. First, the coefficient of determination R² measures how much variance the trend explains, not how large or important the trend is. Second, climate series are autocorrelated, so consecutive residuals are not independent and naive standard errors are too small; effective sample size must be reduced accordingly.
有两点务必注意。第一，决定系数 R² 衡量的是趋势解释了多少方差，而不是趋势有多大或多重要。第二，气候序列存在自相关，相邻残差并不独立，朴素的准误会偏小，因此必须相应降低有效样本量。
Extreme-value statistics are equally examinable. If annual flood peaks follow a Gumbel distribution, the return period of a given level can be computed from its cumulative distribution, which is how engineers convert rainfall data into design standards.
极值统计同样可能考到。若年洪水峰值服从 Gumbel 分布，就可以由累积分布函数计算某水位对应的重现期，工程师正是用这种方式把降雨数据转化为设计标准。
8. Optimisation and the mathematics of mitigation | 最优化与减排的数学
Deciding how fast to cut emissions is an optimisation problem. The objective is to minimise the discounted total cost of abatement, subject to a constraint that cumulative emissions stay below a carbon budget.
决定多快减排是一个最优化问题。目标是使贴现后的减排总成本最小，约束条件是累计排放量低于某个碳预算。
Minimise ∫ C(E(t)) e−rt dt subject to ∫ E(t) dt ≤ B
The Lagrangian for this problem shows that the optimal path equalises the marginal abatement cost across time, adjusted by the discount rate r. Economically this means delaying action is only justified if the return on the avoided investment exceeds the extra damage incurred.
该问题的拉格朗日函数表明，最优路径使各时点的边际减排成本相等，并按贴现率 r 调整。经济学含义是：只有当被推迟的投资收益超过额外造成的损害时，推迟行动才是合理的。
Linear programming gives the same insight for energy supply. Minimising total system cost over generation technologies subject to demand and capacity constraints produces a least-cost mix, and the shadow prices of the constraints are exactly the marginal values of additional capacity or additional storage.
线性规划对能源供给给出同样的洞见。在需求与容量约束下，对各类发电技术最小化系统总成本，得到最低成本组合，而约束的影子价格正是额外容量或额外储能的边际价值。
9. Life as an active boundary condition | 生命作为主动的边界条件
Life is not a passive passenger in this system; it changes the parameters. Vegetation alters albedo, evapotranspiration returns water to the atmosphere, and root systems and soil fungi determine how much carbon is stored on land.
生命在这个系统中不是被动的乘客，而是会改变参数。植被改变反照率，蒸散把水送回大气，根系与土壤真菌决定着陆地储存多少碳。
Tropical forests demonstrate the coupling vividly: moisture evaporated from the canopy is recycled inland as rainfall, so deforestation can reduce rainfall far downwind. This is a positive feedback loop between water and carbon that a two-box model can capture with a recycling ratio.
热带森林生动地展示了这种耦合：冠层蒸发的水汽被输送到内陆并再次形成降雨，因此毁林会减少下风方向的降水。这是水与碳之间的正反馈回路，可用一个再循环比率在双箱模型中加以刻画。
On geological timescales, life regulates carbon through silicate weathering, in which carbon dioxide reacts with silicate minerals to form carbonate and silica. The reaction consumes CO₂ and therefore acts as a very slow negative feedback that stabilises climate over millions of years.
在地质时间尺度上，生命通过硅酸盐风化调节碳，其中二氧化碳与硅酸盐矿物反应生成碳酸盐与二氧化硅。该反应消耗 CO₂，因而充当一个极其缓慢的负反馈，在数百万年尺度上稳定气候。
CO₂ + CaSiO₃ → CaCO₃ + SiO₂
Daisyworld is the classic teaching model of biological regulation. Two daisy types with different albedos grow best at different temperatures, and the resulting coupled equations show that the planetary temperature can be stabilised over a wide range of solar luminosity without any conscious control.
Daisyworld 是生物调节的经典教学模型。两种具有不同反照率的雏菊在不同温度下生长最佳，由此得到的耦合方程表明，在很大范围的太阳光度变化下，行星温度可以在没有任何自觉控制的情况下保持稳定。
10. Modelling limits and exam-style practice | 建模的局限与
广告 / Sponsored培生爱德思考试教材 统计数学3 Edexcel International A Level Mathematics S…数学 备考教辅 · 正版现货 · 与本文内容相关查看详情 / View on Taobao参考价 ¥156.0
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图