5.3 Logarithmic models

Properties of logarithms

A logarithm answers the question: "What exponent do I need to raise this base to in order to get this number?" For example, $\log_2(8) = 3$ because $2^3 = 8$. This concept underpins everything in this section, so make sure it clicks before moving on.

Basic logarithm rules

These rules let you break apart or combine logarithmic expressions, which is essential for solving equations and building models.

• Product rule: $\log_a(xy) = \log_a(x) + \log_a(y)$ Multiplication inside the log becomes addition outside.
• Quotient rule: $\log_a(x/y) = \log_a(x) - \log_a(y)$ Division inside the log becomes subtraction outside.
• Power rule: $\log_a(x^n) = n\log_a(x)$ An exponent inside the log pulls out front as a multiplier.
• Identity property: $\log_a(a) = 1$ for any valid base $a$, because $a^1 = a$.
• Zero property: $\log_a(1) = 0$ for any valid base $a$, because $a^0 = 1$.

Change of base formula

Most calculators only have buttons for $\log_{10}$ and $\ln$. So if you need $\log_3(20)$, you're stuck unless you convert the base.

$\log_a(x) = \frac{\log_b(x)}{\log_b(a)}$

Here $b$ can be any positive base (not equal to 1), but you'll almost always use 10 or $e$. For example:

$\log_3(20) = \frac{\ln(20)}{\ln(3)} \approx \frac{3.00}{1.10} \approx 2.73$

This formula also helps when you need to simplify expressions that mix different log bases.

Natural vs common logarithms

• Natural logarithm $\ln(x)$ uses base $e \approx 2.718$. It shows up constantly in calculus, continuous growth models, and differential equations.
• Common logarithm $\log(x)$ uses base 10. It's the go-to in engineering, scientific notation, and logarithmic scales like pH and decibels.

The two are related through the change of base formula:

$\log_{10}(x) = \frac{\ln(x)}{\ln(10)}$

Since $\ln(10) \approx 2.303$, you can convert between them quickly.

Exponential vs logarithmic functions

Exponential and logarithmic functions are inverses of each other. Understanding how they mirror one another is key to solving equations and interpreting models.

Inverse relationship

If $f(x) = a^x$, then its inverse is $g(x) = \log_a(x)$. "Inverse" means they undo each other:

$a^{\log_a(x)} = x \quad \text{and} \quad \log_a(a^x) = x$

This is the core technique for solving equations. Have an unknown in the exponent? Take a log of both sides. Have an unknown inside a log? Exponentiate both sides.

Graphical representations

• Exponential functions (like $2^x$) grow rapidly and produce J-shaped curves. They always pass through $(0, 1)$ because $a^0 = 1$.
• Logarithmic functions (like $\log_2(x)$) grow slowly and always pass through $(1, 0)$ because $\log_a(1) = 0$.

Since they're inverses, their graphs are mirror images across the line $y = x$.

Asymptotic behavior matters here: the exponential curve approaches the x-axis as $x \to -\infty$ but never touches it. The logarithmic curve approaches the y-axis as $x \to 0^+$ but never touches it.

Domain and range

These flip between the two functions, which makes sense given the inverse relationship:

• Exponential $f(x) = a^x$: Domain is all real numbers; Range is $y > 0$
• Logarithmic $g(x) = \log_a(x)$: Domain is $x > 0$; Range is all real numbers

Always check domain restrictions when solving logarithmic equations. If your answer gives a negative number inside a log, that solution is extraneous and must be thrown out.

Solving logarithmic equations

Logarithm properties application

The basic strategy: use log rules to combine or simplify, then convert to exponential form.

For example, to solve $\log_3(x) + \log_3(x-2) = 1$:

1. Apply the product rule: $\log_3(x(x-2)) = 1$

2. Convert to exponential form: $x(x-2) = 3^1 = 3$

3. Expand and solve: $x^2 - 2x - 3 = 0$, so $(x-3)(x+1) = 0$

4. Get $x = 3$ or $x = -1$

5. Check domain: $x = -1$ makes $\log_3(-1)$ undefined, so reject it. The answer is $x = 3$.

That last step is critical. Always verify that your solutions don't produce negative arguments inside any logarithm.

Exponentiation technique

When you have a single log term isolated on one side, convert directly using the definition of a logarithm:

$\log_a(x) = c \quad \Longleftrightarrow \quad a^c = x$

For example, $\log_2(x) = 3$ becomes $2^3 = x$, so $x = 8$.

For equations like $5^x = 200$ where the variable is in the exponent, take a log of both sides:

1. $\log(5^x) = \log(200)$
2. $x \cdot \log(5) = \log(200)$
3. $x = \frac{\log(200)}{\log(5)} \approx 3.29$

Graphical solutions

When an equation is hard to solve algebraically, you can graph each side as a separate function and find where they intersect. For instance, to solve $\log_2(x) = 5 - x$, graph $y = \log_2(x)$ and $y = 5 - x$, then read off the intersection point. This approach also helps you see whether there are multiple solutions or none at all.

Logarithmic modeling

Logarithmic models show up whenever data spans several orders of magnitude or when growth slows over time. They turn unwieldy exponential relationships into manageable linear ones.

Growth and decay scenarios

Exponential growth (like population increase or compound interest) and exponential decay (like radioactive decay or drug metabolism) both produce data that's hard to analyze on a standard scale. Taking the logarithm of exponential data linearizes it, meaning it turns a curve into a straight line.

For example, if a bacterial population follows $N(t) = N_0 \cdot e^{kt}$, then taking the natural log gives $\ln(N) = \ln(N_0) + kt$. That's a linear equation in $t$, so you can plot $\ln(N)$ vs. $t$ and find the growth rate $k$ from the slope.

Earthquake intensity scale

The Richter scale measures earthquake magnitude using:

$M = \log_{10}\left(\frac{A}{A_0}\right)$

where $A$ is the measured seismic wave amplitude and $A_0$ is a reference amplitude.

Because it's logarithmic, each whole-number increase means 10 times more ground motion. A magnitude 6 earthquake produces 10 times the ground shaking of a magnitude 5, and 100 times that of a magnitude 4. In terms of energy released, each whole number corresponds to roughly 31.6 times more energy.

Sound intensity measurement

The decibel scale measures sound intensity:

$dB = 10 \log_{10}\left(\frac{I}{I_0}\right)$

where $I$ is the measured intensity and $I_0 = 10^{-12}$ W/m² (the threshold of human hearing).

A 10 dB increase means 10 times the sound intensity. Normal conversation is about 60 dB, while a rock concert hits around 110 dB. That 50 dB difference means the concert is $10^5 = 100{,}000$ times more intense. The logarithmic scale matches how our ears actually perceive loudness: we hear equal ratios as equal steps.

Applications in various fields

Computer science and algorithms

Logarithmic time complexity $O(\log n)$ describes algorithms that cut the problem in half at each step. Binary search is the classic example: searching a sorted list of 1,000,000 items takes at most about $\log_2(1{,}000{,}000) \approx 20$ steps.

Logarithms also appear in data structures like B-trees (used in database indexing), data compression, and cryptographic algorithms.

Finance and compound interest

The compound interest formula is $A = P(1 + r)^t$, where $P$ is principal, $r$ is the rate per period, and $t$ is the number of periods. Logarithms let you solve for $t$.

For example, to find how long it takes to double your money at 7% annual interest:

1. Set $2P = P(1.07)^t$
2. Divide: $2 = 1.07^t$
3. Take the log: $t = \frac{\log(2)}{\log(1.07)} \approx 10.24$ years

This is the math behind the Rule of 72: divide 72 by the interest rate to estimate doubling time ($72/7 \approx 10.3$ years).

Biology and population growth

• Logarithmic transformations linearize exponential growth curves, making it easier to fit models and extract growth rates.
• Allometric scaling relates body size to metabolic rate, lifespan, and other traits through power laws, which become linear on a log-log plot.
• The Shannon diversity index $H = -\sum p_i \ln(p_i)$ uses logarithms to quantify species diversity in an ecosystem, where $p_i$ is the proportion of each species.
• Gene expression data is routinely log-transformed to normalize distributions and make fold-changes comparable.

Logarithmic scales

Logarithmic scales compress huge ranges of values into a manageable format. They're used whenever the quantities involved span many orders of magnitude.

Decibel scale

• Measures sound intensity or power ratios
• Formula: $dB = 10 \log_{10}(P/P_0)$, where $P$ is measured power and $P_0$ is reference power
• Each 10 dB increase represents a 10-fold increase in power
• Range: roughly 0 dB (threshold of hearing) to about 120 dB (threshold of pain)
• Used in acoustics, telecommunications, and audio engineering

pH scale

The pH scale measures how acidic or alkaline a solution is:

$pH = -\log_{10}[H^+]$

where $[H^+]$ is the hydrogen ion concentration in moles per liter.

• The scale runs from 0 (strongly acidic) to 14 (strongly alkaline), with 7 being neutral.
• Each unit change represents a 10-fold change in $[H^+]$. A solution with pH 3 has 10 times the hydrogen ion concentration of pH 4, and 100 times that of pH 5.
• Vinegar has a pH around 2.4; pure water is 7; bleach is about 12.5.

Richter scale

• Quantifies earthquake magnitude based on seismic wave amplitude
• Formula: $M = \log_{10}(A/A_0)$, where $A$ is maximum wave amplitude and $A_0$ is a reference amplitude
• Each whole number increase means 10 times more ground motion
• Range: roughly 1 (barely detectable) to 9+ (catastrophic)
• The 2011 Tōhoku earthquake (magnitude 9.1) produced about 1,000 times the ground motion of a magnitude 6 event

Error analysis and estimation

When working with logarithmic models, understanding the precision and reliability of your results matters. These tools help you assess how trustworthy your calculations are.

Significant figures

Significant figures reflect the precision of a measurement. The rules shift slightly when logarithms are involved:

• Addition/subtraction: the result keeps the same number of decimal places as the least precise input.
• Multiplication/division: the result keeps the same number of significant figures as the least precise factor.
• Logarithms: the number of decimal places in the result (the mantissa) should match the number of significant figures in the argument. So $\log(3.45)$ has 3 significant figures in the argument, meaning you report 3 decimal places: $0.538$.

Relative error calculation

Relative error tells you how big the error is compared to the actual value:

$\text{Relative Error} = \frac{|\text{Measured Value} - \text{True Value}|}{|\text{True Value}|} \times 100\%$

This is more informative than absolute error when comparing measurements at different scales. Being off by 1 cm matters a lot more when measuring a coin than when measuring a building.

Logarithmic calculations can amplify or compress errors depending on where you are on the curve, so tracking relative error through your calculations is good practice.

Order of magnitude estimation

An order of magnitude estimate rounds a quantity to the nearest power of 10. Since $\log_{10}(x)$ tells you the power of 10 that $x$ is closest to, logarithms are the natural tool here.

For example, the distance from Earth to the Sun is about 150,000,000 km. That's $\log_{10}(1.5 \times 10^8) \approx 8.2$, so it's on the order of $10^8$ km. These "back-of-the-envelope" estimates help you quickly check whether an answer is reasonable before doing precise calculations.

Computational methods

Logarithm table usage

Before electronic calculators, mathematicians and engineers relied on printed tables of pre-calculated logarithm values. To find a value between entries, they used linear interpolation (estimating proportionally between two known values). This method dominated scientific computation for over 300 years and is worth understanding as context for why logarithm properties were so heavily emphasized in earlier eras.

Calculator functions

Modern scientific calculators have built-in log (base 10) and ln (base $e$) buttons. For other bases, use the change of base formula. Most calculators also provide the inverse functions 10^x and e^x.

One thing to watch: calculators have finite precision (typically 10-12 digits). For very large or very small arguments, rounding errors can accumulate, so be mindful of significant figures in your final answer.

Numerical approximation techniques

Computers calculate logarithms using series expansions and iterative methods. The most common is the Taylor series for the natural log:

$\ln(1+x) = x - \frac{x^2}{2} + \frac{x^3}{3} - \frac{x^4}{4} + \cdots \quad \text{for } -1 < x \leq 1$

More terms give more precision. Newton's method can also be used to solve exponential equations numerically by iterating toward a root. These techniques underpin how programming languages and software libraries compute logarithms behind the scenes.

Historical development

Discovery of logarithms

John Napier published his work on logarithms in 1614, motivated by the need to simplify the enormous multiplication and division problems in astronomy and navigation. His key insight was linking arithmetic progressions (adding) to geometric progressions (multiplying). Henry Briggs then collaborated with Napier to develop base-10 (common) logarithms, which became the standard for practical computation.

Slide rule invention

In 1622, William Oughtred built on Napier's logarithms to create the slide rule. By placing two logarithmic scales side by side and sliding them, you could multiply and divide numbers mechanically. Slide rules were the primary computing tool for engineers and scientists for over three centuries, until electronic calculators replaced them in the 1970s.

Modern computational advancements

Electronic calculators made logarithmic computation instant and accessible starting in the 1970s. Today, logarithms are embedded in programming languages, spreadsheet software, and scientific computing libraries. They remain central to modern fields like machine learning (log-loss functions), information theory (entropy), and data science (log-transformations for skewed data).