Find the direction in which slope is the steepest

  • Thread starter Thread starter wifi
  • Start date Start date
  • Tags Tags
    Direction Slope
Click For Summary
SUMMARY

The discussion centers on finding the direction of the steepest slope for the function z = (2xy) - (3x²) - (4y²) - (18x) + (28y) + 12 at the point (1,1). Participants confirm that the gradient vector, calculated as ∇z = (-6x + 2y - 18)i + (2x - 8y + 28)j, indicates the direction of steepest ascent. At (1,1), the gradient evaluates to ∇z(1,1) = -22i + 22j, leading to a magnitude of 22√2. The angle with the x-axis is computed using the dot product, resulting in θ = 135°, while the correct answer is noted as θ = -45°.

PREREQUISITES
  • Understanding of multivariable calculus concepts, particularly gradients.
  • Familiarity with vector notation and operations, including dot products.
  • Knowledge of trigonometric functions and angle calculations.
  • Ability to differentiate functions of multiple variables.
NEXT STEPS
  • Study the properties of gradient vectors in multivariable calculus.
  • Learn how to compute and interpret the gradient of a function.
  • Explore the relationship between gradient direction and steepest ascent/descent.
  • Investigate common pitfalls in angle calculations involving vectors.
USEFUL FOR

Students and professionals in mathematics, particularly those studying calculus and optimization, as well as anyone interested in understanding gradient vectors and their applications in determining slope directions.

wifi
Messages
115
Reaction score
1
Problem:

The height of a hill (in meters) is given by [z=(2xy)-(3x^2)-(4y^2)-(18x)+(28y)+12], where x is the distance east, y is the distance north of the origin. In which compass direction is the slope at x=y=1 the steepest?

Solution (so far):

Does it have something to do with the gradient?
 
Physics news on Phys.org
wifi said:
Does it have something to do with the gradient?
Since the gradient vector points in maximal increase of a multivariate function, and z here represents a height then grad(z) will point where the height increases most rapidly, I.e where it is most steep. So yes, the gradient would certainly be helpful here.
 
I thought so, but I don't see how I'd be able to find the direction by calculating grad(z). Any suggestions?
 
wifi said:
I thought so, but I don't see how I'd be able to find the direction by calculating grad(z). Any suggestions?

The ##∇z## will give you a vector ...

So, the gradient of Z at a point is a vector pointing in the direction of the steepest slope.
 
I calculated ## \nabla z(x,y)= (-6x+2y-18)\hat{i}+(2x-8y+28)\hat{j} ##.

So this is a vector pointing in the direction of greatest increase (ie. the steepest).

Thus, ## \nabla z(1,1)= (-6+2-18)\hat{i}+(2-8+28)\hat{j}=(-6+2-18)\hat{i}+(2-8+28)\hat{j}=-22\hat{i}+22\hat{j} ##
 
Crake said:
The ##∇z## will give you a vector ...

So, the gradient of Z at a point is a vector pointing in the direction of the steepest slope.

Thanks. I calculated the gradient. I'm not sure what to do next.
 
I found ## |\nabla z(1,1)|= 22 \sqrt{2} ##. So, by the dot product, the angle this vector makes with the x-axis is given by ## cos \theta = \frac{\nabla z(1,1) \cdot \hat{x}}{|\nabla z(1,1)| \cdot |\hat{x}|} = \frac{-22}{22\sqrt{2} \cdot 1 }= \frac{1}{\sqrt{2}} \Rightarrow \theta = 135°##.
 
I think I messed up somewhere because my text is saying correct answer is ## \theta = -45°##.
 
Last edited:
wifi said:
I think I messed up somewhere because my is saying correct answer is ## \theta = -45°##.

Hum.. Perhaps your initial function lacks a minus somewhere?
 
  • #10
Well, the steepest ascent is also associated with the steepest descent... just the opposite direction (180° turn).
 
  • #11
gneil, I understand what you mean, but in that case shouldn't -45° & 135° both be acceptable answers?
 
  • #12
wifi said:
gneil, I understand what you mean, but in that case shouldn't -45° & 135° both be acceptable answers?

Yes, I would think so.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
Replies
8
Views
1K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 4 ·
Replies
4
Views
5K
Replies
13
Views
3K
Replies
13
Views
2K
Replies
4
Views
2K
Replies
6
Views
4K