-
Notifications
You must be signed in to change notification settings - Fork 1
/
Chapter_11_Drawing_Maps.qmd
232 lines (175 loc) · 12.5 KB
/
Chapter_11_Drawing_Maps.qmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
# Drawing Maps
## Introduction
Climatic data is often presented in map form. Common examples include
details of the stations in each country, plus further information about
each station.
Preparing a map is usually a 2-stage process. The first stage, described
in Section 10.2, uses a shape file to provide an outline of the country,
and other general details. Then climatic details are added, as is
described in Section 10.3.
## Using a shape file
R-Instat does not currently have any shape files in the library, so they
have to be downloaded. There are various sites that offer shape files
and GADM (GADM, 2019) is comprehensive, Fig. 10.2a.
-----------------------------------------------------------------------
***Fig. 10.2a The GADM data page***
-----------------------------------------------------------------------
{width="6.11749343832021in"
height="3.305452755905512in"}
-----------------------------------------------------------------------
C licking on country in Fig. 10.1a allows the input of a country name,
as shown in Fig. 10.2b. For Kenya, as for most other formats, there are
different ways to download the data.
-------------------------------------------------------------------------------------------------------------
***Fig. 10.2b*** ***Fig. 10.2c***
------------------------------------------------------ ------------------------------------------------------
{width="2.7050929571303586in" {width="3.3892760279965004in"
height="2.183115704286964in"} height="1.7678455818022747in"}
-------------------------------------------------------------------------------------------------------------
Data at level 0 is simply the country boundary, while level 1 gives the
outline for each of Kenya's 47 counties. Levels 2 and 3 provide even
more detail.
In Fig. 10.2b the R(sf) option provides a file, with an ***rds***
extension, that can be opened into R-Instat through the ***File \>
Open*** dialogue. However, we suggest clicking on Shapefile in Fig.
10.2. This downloads a zip file that is about 20 Mbytes for Kenya and
which contains data at all 4 levels, 0, 1, 2, 3.
Unzip this file. In R-Instat use ***Climatic \> File \> Open and Tidy
Shapefile***, Fig. 10.2c. Go to the unzipped files and open the Level 1
shapefile. There are 47 rows of data, shown in Fig. 10.2d, (after
reordering the variables for clarity.)
------------------------------------------------------------------------------------------------------------
***Fig. 10.2d*** ***Fig. 10.2e***
------------------------------------------------------ -----------------------------------------------------
{width="3.5070898950131233in" {width="2.380729440069991in"
height="3.419274934383202in"} height="2.9443088363954506in"}
------------------------------------------------------------------------------------------------------------
Go to ***Climatic \> Mapping***. Part of this dialogue is shown in Fig.
10.2e. Choose the data frame with the shape file if it isn't chosen
automatically. Then just press Ok for the map shown in Fig. 10.2f.
What could be easier?
Return to the ***Climatic \> Mapping*** dialogue and add the
***NAME_1*** variable from Fig. 10.2d into the ***Fill field*** of Fig.
10.2e. The result is in Fig. 10.2g. The map is now nicely colourful. If
there were, say, 10 regions, this would be fine, but with 47 counties
the legend is taking too much space and can't easily be associated with
each county.
----------------------------------------------------------------------------------------------------------
***Fig. 10.2f Kenya county map*** ***Fig. 10.2g***
---------------------------------------------------- -----------------------------------------------------
{width="2.97536198600175in" {width="3.021402012248469in"
height="3.437853237095363in"} height="2.575886920384952in"}
----------------------------------------------------------------------------------------------------------
So, return to the ***Climatic \> Mapping*** dialogue. Press ***Plot
Options,*** then use the ***Theme*** tab and set the ***Legend to
None***, Fig. 10.2h.
-------------------------------------------------------------------------------------------------------------
***Fig. 10.2h*** ***Fig. 10.2i***
------------------------------------------------------ ------------------------------------------------------
{width="2.9793011811023624in" {width="3.0055807086614172in"
height="2.9853805774278217in"} height="3.5880883639545056in"}
-------------------------------------------------------------------------------------------------------------
The resulting map is shown in Fig. 10.2i.
Sometimes it is useful for an initial map to include the names of the
districts as labels.
Return to the ***Climatic \> Mapping*** dialogue and include the same
data frame also on the right-hand side of the dialogue, Fig 10.2j.
-----------------------------------------------------------------------
***Fig. 10.2j***
-----------------------------------------------------------------------
{width="5.439518810148732in"
height="2.9879385389326334in"}
-----------------------------------------------------------------------
The initial results are in Fig. 10.2k. It shows that, with 47 labels we
have given the software a considerable challenge to make all the labels
visible. The geom is called geom_label_repel and has done its best, but
perhaps it needs some help. Two possibilities are to make the labels
smaller, Fig. 10.2l, or to omit the box round each label. This then uses
geom_text_repel instead, Fig. 10.2m.
-----------------------------------------------------------------------------------------------------------
***Fig. 10.2k*** ***Fig. 10.2l***
------------------------------------------------------ ----------------------------------------------------
{width="3.0282250656167977in" {width="2.979179790026247in"
height="3.608007436570429in"} height="3.6041830708661418in"}
-----------------------------------------------------------------------------------------------------------
Finally, in this section, we go "down a level". This uses the level 2
data, which has 310 rows. Then we filter to give just one district,
choosing Kisumu. (A few neighbouring districts could be chosen.) The
resulting map is shown in Fig. 10.2n.
The GADM site is limited to administrative data only. There are other
sites, such as
[[http://www.diva-gis.org/]{.underline}](http://www.diva-gis.org/) with
additional information, such as roads and elevation.
---------------------------------------------------------------------------------------------------------
***Fig. 10.2m Text, rather than label*** ***Fig. 10.2n***
---------------------------------------------------- ----------------------------------------------------
{width="2.865278871391076in" {width="3.192403762029746in"
height="3.3938265529308835in"} height="1.2317147856517936in"}
---------------------------------------------------------------------------------------------------------
## Adding station data
Adding station data to a map is just as easy as adding the county
information in Section 10.2. If your data are in CLIMSOFT, and you have
the necessary permissions, then transferring the data may be done with
the ***Climatic \> File \> Import from Climsoft*** dialogue. Here we
illustrate with the Kenya data from the Instat library.
Use ***File \> Open from Library \> Instat \> Browse \> Climatic \>
Kenya*** and open the file called western_kenya.rds. This has 3 data
frames, including one called wkenya_stationinfo, Fig. 10.3a.
----------------------------------------------------------------------------------------------------------
***Fig. 10.3a*** ***Fig. 10.3b***
---------------------------------------------------- -----------------------------------------------------
{width="4.219044181977253in" {width="1.6362029746281714in"
height="3.1557469378827645in"} height="2.9429680664916886in"}
----------------------------------------------------------------------------------------------------------
There are 52 stations, Fig. 10.3a, and the information includes the
County as well as the latitude and longitude. An initial step is to
examine how many counties have data. This uses the ***Right-Click***
from the top of the County column, and the ***Levels/Labels*** option,
also shown in Fig. 10.3a.
The Levels/Labels dialogue shows there are data from 12 counties. The
spelling is consistent with the County=level file, used in Section 10.2
except for Homabay, which we change, in this dialogue, to Homa Bay.
This initial check is important, as any spelling mistakes in the county
name, when entering the data, will invent a new county!
----------------------------------------------------------------------------------------------------------
***Fig. 10.3d Filter the map file*** ***Fig. 10.3e Map file for Western Kenya***
---------------------------------------------------- -----------------------------------------------------
{width="2.791648075240595in" {width="3.346146106736658in"
height="2.561923665791776in"} height="2.414307742782152in"}
----------------------------------------------------------------------------------------------------------
Return to the level 1 map file, Right-Click and Filter to choose the
same 12 counties, Fig. 10.3d. The data file should now have just 12
rows, to match the number of counties that will be mapped.
Return to the Climatic \> Mapping dialogue. The map information is on
the left-hand side of the dialogue and the new station information is
now on the right, Fig. 10.3f.
-----------------------------------------------------------------------
***Fig. 10.3f***
-----------------------------------------------------------------------
{width="5.3781485126859145in"
height="3.0781649168853895in"}
-----------------------------------------------------------------------
The result is shown in Fig. 10.3g with labels, and with the county given
as a legend. Fig. 10.3h uses text and omits the legend. This is quite a
challenging map, because so may stations are close together.
-----------------------------------------------------------------------------------------------------------
***Fig. 10.3g*** ***Fig. 10.3h***
---------------------------------------------------- ------------------------------------------------------
{width="3.244919072615923in" {width="2.7297692475940507in"
height="2.782950568678915in"} height="2.7890474628171478in"}
-----------------------------------------------------------------------------------------------------------
The results show there are at least 2 obvious errors in the data. The
most obvious is Chemilil, which should be in Kisumu county, but is shown
in the middle of Kajaido. And Akira, that should be in Kajaido is shown
slightly outside any of the counties. It is tempting to correct these
obvious errors. However, better is the message that there may be other
errors and a strategy for checking the geographical information would be
useful.
We wonder whether google maps could be used for checking? This aspect
has become important, partly because of the potential of combining
station and satellite information. However, the combining is only
sensible if the locations of the station data are correct.
## Information about the stations
[To be added.]{.mark}
## Further types of mapping
[Contours, gridded data]{.mark}