Geology Reference
In-Depth Information
An
x-y plot
consists of an array of data points or
a line that joins the points. The line may also be a
"best-fit" line, which shows the approximate trend of
the data points. Each point on an x-y plot has two val-
ues, one value that is plotted for the x-axis, and one
value that is plotted for the y-axis. An x-y plot is a
good way to show the relationship between two
quantities and often a best fit line can be used to make
estimates where data are missing. Usually, the scales
on x-y graphs are arithmetic (also known as linear);
however, these graphs can also be constructed using
logarithmic scales on one or both axes. Logarithmic
scales save space for quantities that have wide varia-
tion and also produce straight lines that can improve
predictions.
that you might make in lab exercises. Some data are
provided in text descriptions in the topic, and more
data are available from textbooks. Data are also
available from the Internet, but it is necessary to be
very careful using Internet data, as much of it is of
poor quality. With computer-based data plotting pro-
grams such as Excel, it is possible to select data from
many sources and plot them in a wide variety of for-
mats. It is important to finally select the graph style
that is most informative in conveying the information
you desire.
Plotting Lines
Joining the points on an x-y graph with a line may be
appropriate for showing the exact number of sales per
month or the calculated variation of mass with vol-
ume. In experimental results, particularly when the
experiment involves measurement of some aspect of
nature, scientific uncertainty often precludes joining
the points with an exactly straight line. Because it is
reasonable to assume there is a relationship between
two quantities that we are plotting, a straight line is
often drawn through the array of points, balancing
those above with those below. We can do this "best-fit"
straight line by eye (that is, estimating what the
straight line that comes closest to being aligned with
all the points looks like) or the line can be calculated
using a hand calculator. If we had joined each point, it
would suggest an irregular relationship or no uncer-
tainty in our measurements.
Selecting Scales
In drawing graphs it is important to select a scale that
includes the range of point values expected or included
in the data. The scale chosen must also be able to be
plotted easily on available graph paper. One technique
for including a wide range of data on a scale is to break
the axis with a zig-zag line and change scales; another
is to use a logarithmic scale. Axes scales must be
labeled to give the values and the units. Scientific nota-
tion may be used to reduce the size of the numbers
needed on axis labels.
Logarithmic scales are usually recognized by the
variable grid pattern made by the changing size of the
units. The scales usually increase by units of 10. Loga-
rithmic scales are often used in plotting quantities
such as stream discharge measurements, grain size
analyses of sands and sediments, and Richter magni-
tudes. These are all units that have broad ranges of
values.
QUESTIONS 3, PART C
1.
On the two types of graph paper provided here, plot the
number of landslides by year, using the data in Table 3.1.
Add captions.
Writing Captions
All graphs must have a caption that explains what the
graph shows, describes and defines symbols used for
points and lines, identifies the source(s) of data shown,
and indicates when the data were collected. If the
graph is based on an earlier graph or graphs, the
source(s) of those earlier graphs must be acknowl-
edged in the title. In scientific papers, graphs are usu-
ally categorized as figures. Figures are numbered for
reference in the text of scientific papers or topics. Cap-
tions are typically placed below figures but above
tables. In some figures the key to the symbols used is
included in the figure. Being able to write a clear cap-
tion for graphs you construct is an important aspect of
understanding the meaning of the graphs.
TABLE 3.1 Total number of landslides, by year, in
Creepville, for the period from 1991-2001
Number of Landslides
Year
20
1991
6
1992
7
1993
10
1994
8
1995
12
1996
45
1997
Data Sources
Gathering data (which is a plural word; the singular
word is datum) to plot in graphs and interpret is an
important aspect of the scientific method. Data in this
topic are available from measurements or information
29
1998
47
1999
30
2000
10
2001
