Section 1
Historical Context & Motivation
Long before graphing calculators or spreadsheets, mathematicians and scientists noticed a common pattern in nature: some quantities don't just increase steadily—they multiply. A colony of bacteria doesn't add a fixed number of cells each hour; instead, each cell divides, causing the population to double again and again. Debts left unpaid don't grow by a constant dollar amount; they compound, with interest accruing on top of interest. The mathematical tools that describe these phenomena—exponential functions and their inverses, logarithms—took centuries to develop, and their story is intertwined with the history of commerce, astronomy, and physics.
1614 • John Napier
1683 • Jacob Bernoulli
1798 • Thomas Malthus
1896 • Henri Becquerel
1935 • Charles Richter
These milestones show that exponential and logarithmic models aren't abstract algebra exercises—they are the language nature uses to describe growth, decay, and scale. The central question this lesson addresses is: How do we translate a real-world situation into an exponential or logarithmic equation, and how do we interpret the results?
Section 2
Core Principles & Definitions
Before we can model a scenario, we need to understand the building blocks. Exponential and logarithmic functions are inverses of one another, much like multiplication and division. Mastering both sides of this relationship is the key to setting up, solving, and interpreting real-world models.
Exponential Function
f(x) = a · bˣ, where a is the initial value (when x = 0), b is the base (b > 0, b ≠ 1), and x is the exponent. When b > 1 the function models growth; when 0 < b < 1 it models decay.Logarithmic Function
log_b(y) = x means bˣ = y. Logarithms answer the question: "What exponent do I need?" Common bases include 10 (common log) and e (natural log, ln).Continuous Growth / Decay
A = A₀ · eᵏᵗ. Here k > 0 gives growth and k < 0 gives decay. The constant e ≈ 2.71828 arises naturally from calculus and continuous compounding.Half-Life & Doubling Time
t½ = ln(2)/|k|.Key Takeaway
Section 3
Visual Explanation — Growth vs. Decay
The graph below shows two exponential functions on the same coordinate plane. The cyan curve represents exponential growth (y = 2ˣ), and the pink curve represents exponential decay (y = (½)ˣ). Notice that both pass through the point (0, 1) because any base raised to the zero power equals one. The growth curve accelerates upward to the right, while the decay curve approaches—but never reaches—zero.
Two critical observations emerge from this graph. First, the horizontal asymptote at y = 0 means an exponentially decaying quantity gets infinitesimally close to zero but never actually reaches it—this is why we say a radioactive sample never fully decays. Second, exponential growth is deceptively slow at the start and then explosively fast: between x = 0 and x = 1 the growth function only rises from 1 to 2, but between x = 9 and x = 10 it jumps from 512 to 1,024. This "hockey-stick" shape catches many real-world planners off guard.
Section 4
Mathematical Framework
There are several standard forms for exponential and logarithmic models. Choosing the right one depends on the context of the problem—whether growth is discrete or continuous, and whether you're solving for the amount or for the time.
Standard Exponential Model
A₀ = initial amount · b = growth/decay factor · t = time
In this form, if a quantity doubles every period, then b = 2. If it loses 15% per period, then b = 1 − 0.15 = 0.85. The key insight is that the base b encodes the rate of change per unit of time. When the problem gives you a percentage rate r, the relationship is b = 1 + r for growth and b = 1 − r for decay, where r is expressed as a decimal.
Continuous Growth / Decay Model
k > 0 → growth · k < 0 → decay · e ≈ 2.71828
When a process compounds continuously—as with bacterial growth, radioactive decay, or continuously compounded interest—we use the natural exponential base e. The continuous rate k is related to the discrete factor b by the equation k = ln(b). This means eᵏ = b, so the two models always agree.
Solving for Time with Logarithms
Equivalently: t = (A / A₀), using any consistent base
This is where logarithms earn their keep. Whenever you know the starting and ending amounts and need to find when something happens (for instance, "When will the population reach 10,000?"), you isolate the exponential, take a logarithm of both sides, and solve for t. The change-of-base formula log_b(x) = ln(x) / ln(b) lets you convert between any logarithmic bases.
Half-Life & Doubling Time
Works for both half-life (decay) and doubling time (growth)
A remarkable property of exponential processes is that the time to halve (or double) is constant, regardless of the current amount. A 10-gram sample of carbon-14 takes 5,730 years to become 5 grams, and a 2-gram sample takes the same 5,730 years to become 1 gram. This constant interval is entirely determined by the rate k.
Section 5
Detailed Breakdown — Common Model Types
The exponential and logarithmic framework applies to a wide variety of real-world scenarios. The diagram below maps out the most common types of models and the contexts where each appears, followed by a reference table of model parameters.
| Scenario | Model Type | Equation Form | Key Parameter |
|---|---|---|---|
| Compound interest (annual) | Discrete growth | A = P(1 + r/n)^(nt) | r = annual rate, n = compounds/year |
| Continuously compounded interest | Continuous growth | A = Pe^(rt) | r = continuous rate |
| Population growth | Growth | P(t) = P₀ · e^(kt) | k = growth rate constant |
| Radioactive decay | Decay | N(t) = N₀ · (½)^(t/t½) | t½ = half-life |
| Car depreciation | Discrete decay | V(t) = V₀(1 − r)^t | r = depreciation rate per year |
| Earthquake magnitude | Logarithmic | M = log₁₀(I / I₀) | I₀ = reference intensity |
| Sound intensity (decibels) | Logarithmic | dB = 10 · log₁₀(I / I₀) | I₀ = 10⁻¹² W/m² |
The table above serves as a quick-reference guide. Notice that every growth or decay model has the same structure: an initial value multiplied by a base raised to a power involving time. The logarithmic models "flip" the relationship, mapping enormous ranges of physical intensity onto a compact, human-friendly scale.
Section 6
Worked Example
A biologist records that a bacterial colony in a petri dish contains 500 bacteria at time t = 0 hours and 4,000 bacteria at t = 3 hours. Assuming exponential growth, find the continuous growth rate k, write the model, and determine when the colony will reach 100,000 bacteria.
Bacterial Growth Model
- 1
Step 1 — Identify the model form
Since the problem says "exponential growth" and asks for a continuous rate, we use the continuous model:A(t) = A₀ · e^(kt). We know A₀ = 500 and A(3) = 4,000. - 2
Step 2 — Substitute the known data point to find k
4,000 = 500 · e^(3k)→ Divide both sides by 500:8 = e^(3k) - 3
Step 3 — Take the natural logarithm of both sides
ln(8) = 3k→k = ln(8) / 3 = 2.0794 / 3 ≈ 0.6931The continuous growth rate is approximately k ≈ 0.693 per hour. This makes intuitive sense: ln(2) ≈ 0.693, so the doubling time is about 1 hour (the colony multiplied by 8 = 2³ in 3 hours—three doublings). - 4
Step 4 — Write the complete model
A(t) = 500 · e^(0.6931t) - 5
Step 5 — Solve for when A(t) = 100,000
100,000 = 500 · e^(0.6931t)→200 = e^(0.6931t)→ln(200) = 0.6931t→t = ln(200) / 0.6931 = 5.2983 / 0.6931 ≈ 7.64 hours - 6
Step 6 — Interpret the result
The bacterial colony will reach 100,000 bacteria approximately 7 hours and 38 minutes after the initial observation. At this point roughly 7.64 doublings have occurred (since .64 ≈ 200, confirming our answer). The biologist should plan to take the next observation no later than 7.5 hours to catch the milestone.
Section 7
Strengths, Limitations & Comparisons
Exponential and logarithmic models are powerful but not universal. Understanding when they apply—and when they don't—is just as important as knowing how to set them up.
| Aspect | Strengths | Limitations |
|---|---|---|
| Accuracy | Excellent fit for short-to-medium time spans when growth rate is genuinely proportional to current size. | Over long periods, real populations hit resource limits; exponential models overpredict. |
| Simplicity | Only two parameters needed (A₀ and k or b), making them easy to calibrate from minimal data. | Oversimplifies processes with multiple interacting factors (predator-prey dynamics, economic cycles). |
| Predictive power | Logarithmic models compress wide-range data beautifully (earthquake magnitudes span 10⁸ in intensity). | Extrapolation beyond the data range is risky; a small error in k compounds over time. |
| Mathematical elegance | Exponential functions are their own derivatives—they integrate seamlessly into calculus and differential equations. | Not appropriate for processes that change at a constant absolute rate (use linear models instead). |
| Real-world fit | Ideal for radioactive decay, early-stage epidemics, compound interest, and cooling processes. | Logistic models (S-curves) are more realistic for populations approaching carrying capacity. |
Key Takeaway
Section 8
Connection to Advanced Theory
The exponential and logarithmic models you learn in Algebra 2 are the foundation for much more sophisticated mathematics. Understanding how these simple models evolve will help you see where you're headed—and appreciate the power of what you already know.
| Algebra 2 Concept | Advanced Extension | Where You'll See It |
|---|---|---|
A = A₀ · e^(kt) | Differential equation: dA/dt = kA | Calculus (AP Calculus BC, college math) |
| Exponential growth (unlimited) | Logistic model: P = K / (1 + Ce⁻ʳᵗ) | AP Biology, ecology, epidemiology |
| Compound interest formula | Present value, annuities, Black-Scholes model | Finance, economics, actuarial science |
| Logarithmic scales (Richter, pH) | Information theory (Shannon entropy), signal processing | Computer science, electrical engineering |
| Half-life calculations | Nuclear decay chains, pharmacokinetics | Physics, medical science |
In calculus, you'll discover that the exponential function eˣ is special because it is its own derivative: the rate of change of eˣ is eˣ. This single fact explains why so many natural processes are exponential—whenever a quantity's rate of change is proportional to its current size, the solution is automatically an exponential function. The logarithm, as its inverse, then becomes the go-to tool for "undoing" exponential relationships, which is why logarithmic transformations appear throughout statistics, physics, and data science.
In statistics, you'll encounter logarithmic transformations that convert exponential data into linear data, making it easier to analyze with regression techniques. If you plot ln(y) versus x and the result is a straight line, you've confirmed that the original data follows an exponential model—a technique called linearization.
Section 9
Practice Problems
Question 1: Problem 1 — Conceptual
Show answer
Question 2: Problem 2 — Basic Calculation
Show answer
A(t) = 2000(1 + 0.045/12)^(12t) = 2000(1.00375)^(12t). For t = 6 years: A(6) = 2000(1.00375)⁷² ≈ 2000 × 1.30937 ≈ $2,618.74. The investment earned about $618.74 in interest over six years.Question 3: Problem 3 — Intermediate
Show answer
N(t) = 50 · (½)^(t/8.02). To find when N(t) = 5 mg: 5 = 50 · (½)^(t/8.02) → 0.1 = (½)^(t/8.02). Take ln of both sides: ln(0.1) = (t/8.02) · ln(0.5) → t = 8.02 · ln(0.1) / ln(0.5) = 8.02 × (−2.3026) / (−0.6931) ≈ 8.02 × 3.3219 ≈ 26.64 days. After approximately 26.6 days (about 3.3 half-lives), only 5 mg of iodine-131 will remain. This information helps hospitals plan when the sample will be too weak for diagnostic imaging.Question 4: Problem 4 — Applied Multi-Step
Show answer
I_B / I_A = 10^6.7 / 10^4.2 = 10^(6.7 − 4.2) = 10^2.5 ≈ 316.2. The City B earthquake is approximately 316 times more intense than City A's earthquake. This dramatically illustrates why the Richter scale is logarithmic: a difference of just 2.5 points corresponds to a factor of over 300 in actual seismic wave amplitude. Each whole number increase on the Richter scale represents a tenfold increase in measured amplitude.Question 5: Problem 5 — Critical Thinking / Synthesis
Show answer
t ≈ 12.51 months. App 1 overtakes App 2 roughly 12.5 months after its own launch. At that point, both apps have approximately 9,230 users. Discussion: No, these growth rates will not hold indefinitely. As user bases grow, market saturation slows acquisition; server costs, competition, and user churn all dampen growth. A more realistic long-term model would be logistic (S-shaped), where growth tapers off as the app approaches a maximum "carrying capacity" of potential users. The exponential model is excellent for the early-stage forecast but should not be extrapolated to, say, t = 60 months without serious adjustments.Summary
Lesson Summary
Summary
In this lesson, we explored how exponential functions of the form A = A₀ · bᵗ (or A = A₀ · eᵏᵗ) model real-world processes where a quantity changes by a constant percentage per unit of time—whether that's population growth, radioactive decay, or compound interest. We learned that the logarithm is the inverse operation that lets us solve for time: if you know the starting amount and the target, taking a log "unwraps" the exponent. We also saw that logarithmic scales—like the Richter scale, the decibel scale, and pH—use logarithms to compress huge ranges of physical data into manageable numbers.
The key modeling steps are: (1) identify whether the situation involves growth or decay, (2) determine the initial value A₀ and the rate parameter (b, r, or k), (3) write the appropriate equation, and (4) use logarithms to solve for unknowns in the exponent. Remember that half-life and doubling time are constant for any exponential process (t½ = ln 2 / |k|), and that these models work best over short-to-medium time horizons—for long-term forecasting, more advanced models like the logistic function are needed.