01 / 21

Logarithmic
Models

Learn how logarithmic functions model real-world phenomena: from earthquake intensity to sound levels and beyond.

College Algebra

What Is a Logarithm?

A logarithm answers the question: "To what exponent must I raise a base to get a given number?" If \( b^y = x \), then \( \log_b(x) = y \).

\( \log_{10}(1000) \)
Equals 3 because \( 10^3 = 1000 \)
\( \ln(e^5) \)
Equals 5 because the natural log uses base \( e \)

Shape of the Logarithmic Curve

\(x\)
\(y\)
1 2 3 4 0 1 2 -1 -2
\(x = 0\)
(1, 0)
\(y = \ln(x)\)

Why Use Logarithmic Models?

Logarithmic models are ideal when data grows rapidly at first, then levels off. They compress enormous ranges of values into manageable scales.

Richter Scale
Earthquake magnitude: each whole number is a 10x increase in amplitude
Decibel Scale
Sound intensity: measures vast loudness differences on a compact scale
pH Scale
Acidity: each unit represents a 10x change in hydrogen ion concentration

The Logarithm-Exponential Bridge

Exponential Form
\( b^y = x \)
equivalent Logarithmic Form
\( \log_b(x) = y \)

Essential Log Properties

1
Product Rule: \( \log_b(MN) = \log_b(M) + \log_b(N) \)
2
Quotient Rule: \( \log_b\!\left(\frac{M}{N}\right) = \log_b(M) - \log_b(N) \)
3
Power Rule: \( \log_b(M^p) = p \cdot \log_b(M) \)
4
Change of Base: \( \log_b(x) = \dfrac{\ln(x)}{\ln(b)} = \dfrac{\log(x)}{\log(b)} \)

Logarithmic vs Exponential

Logarithmic Model Exponential Model
General Form \( y = a + b\ln(x) \) \( y = ab^x \)
Growth Pattern Fast then slow Slow then fast
Asymptote Vertical (\( x = 0 \)) Horizontal (\( y = 0 \))
Best For Diminishing returns Compound growth
Example Learning curves Population growth

The General Log Model

\( y = a + b \ln(x) \)
\( a \)
Vertical shift: the y-intercept adjustment. It shifts the entire curve up or down.
\( b \)
Growth rate: if \( b > 0 \), the model is increasing. If \( b < 0 \), it is decreasing.

How to Build a Log Model

1
Identify the pattern: Does the data increase quickly at first, then level off? If yes, a log model may fit.
2
Choose two data points \( (x_1, y_1) \) and \( (x_2, y_2) \) to set up a system of equations.
3
Substitute into \( y = a + b\ln(x) \) to create two equations in \( a \) and \( b \).
4
Solve the system: subtract equations to eliminate \( a \) and solve for \( b \), then back-substitute for \( a \).
5
Validate: check the model against a third data point or use regression technology.

Example: Richter Scale

The Richter magnitude \( M \) of an earthquake is modeled by:

Problem: An earthquake has intensity \( I = 10{,}000 \cdot I_0 \). What is its Richter magnitude?

1
Substitute: \( M = \log\!\left(\dfrac{10{,}000 \cdot I_0}{I_0}\right) = \log(10{,}000) \)
2
Simplify: \( \log(10{,}000) = \log(10^4) = 4 \)
3
Answer: The earthquake has magnitude \( M = 4 \).

Example: Learning Curve

A student's test score after \( t \) hours of study is modeled by \( S(t) = 25 + 18\ln(t) \). Find the score after 1 hour and after 10 hours.

After 1 Hour
\( S(1) = 25 + 18\ln(1) = 25 + 18(0) = \) 25
After 10 Hours
\( S(10) = 25 + 18\ln(10) \approx 25 + 41.4 \approx \) 66.4

Example: Finding \( a \) and \( b \)

A logarithmic model \( y = a + b\ln(x) \) passes through the points \( (1, 3) \) and \( (e^2, 11) \). Find \( a \) and \( b \).

1
Using \( (1, 3) \): \( 3 = a + b\ln(1) = a + 0 \), so \( a = 3 \).
2
Using \( (e^2, 11) \): \( 11 = 3 + b\ln(e^2) = 3 + 2b \).
3
Solve: \( 2b = 8 \), so \( b = 4 \).
4
Model: \( y = 3 + 4\ln(x) \).

Graphing \( y = 3 + 4\ln(x) \)

\(x\)
\(y\)
1 2 3 4 0 2 4 6 8 10 -2 (1, 3)
\(y = 3 + 4 \ln(x)\)

Example: Decibel Scale

Sound intensity in decibels is modeled by:

Problem: A rock concert has intensity \( I = 10^{-2} \text{ watts/m}^{2}\). Find \( D \).

1
Substitute: \( D = 10\log\!\left(\dfrac{10^{-2}}{10^{-12}}\right) = 10\log(10^{10}) \)
2
Simplify: \( D = 10 \cdot 10 = \) 100 dB

Common Mistakes

✕ Logging a Negative
Writing \( \ln(-5) \). The domain of \( \ln(x) \) is \( x > 0 \) only. Check that the argument is positive.
✕ Product Rule Error
Writing \( \log(a+b) = \log(a) + \log(b) \). The product rule applies to \( \log(a \cdot b) \), not to sums.
✕ Forgetting the Base
Mixing up \( \log \) (base 10) and \( \ln \) (base \( e \)) in a calculation, leading to wrong answers.

Summary of Log Models

Model Formula Domain Use Case
Richter Scale \( M = \log(I/I_0) \) \( I > 0 \) Earthquakes
Decibel Scale \( D = 10\log(I/I_0) \) \( I > 0 \) Sound levels
pH Scale \( \mathrm{pH} = -\log[\mathrm{H}^+] \) \( [\mathrm{H}^+] > 0 \) Acidity
General Model \( y = a + b\ln(x) \) \( x > 0 \) Curve fitting

Question 1 of 5

If \( \log_b(x) = 3 \), what is \( x \) in terms of \( b \)?

Question 2 of 5

An earthquake measures magnitude 6 on the Richter scale using \( M = \log(I/I_0) \). How many times more intense is it than a magnitude 4 earthquake?

Question 3 of 5

Using the model \( S(t) = 25 + 18\ln(t) \), approximately how many hours of study are needed to reach a score of 70?

Question 4 of 5

Which expression is equivalent to \( \log_5(75) - \log_5(3) \)?

Question 5 of 5

A solution has a hydrogen ion concentration of \( [\mathrm{H}^+] = 10^{-9} \). Using \( \mathrm{pH} = -\log[\mathrm{H}^+] \), what is the pH?