1
00:00:01,590 --> 00:00:05,460
Hi, and welcome back to
Remote Sensing Foundations.

2
00:00:05,460 --> 00:00:08,160
This week we are gonna
talk about the instruments

3
00:00:08,160 --> 00:00:10,650
that carry out remote sensing.

4
00:00:10,650 --> 00:00:13,560
Those instruments are called sensors.

5
00:00:13,560 --> 00:00:14,700
At the end of the week,

6
00:00:14,700 --> 00:00:18,600
I hope that you'll be able to describe

7
00:00:18,600 --> 00:00:22,290
some common sensors that
we use in remote sensing

8
00:00:22,290 --> 00:00:25,563
and also be able to
define how they differ.

9
00:00:28,230 --> 00:00:31,590
I first wanted to review
what we covered last week.

10
00:00:31,590 --> 00:00:34,080
So if you remember last
week we started off

11
00:00:34,080 --> 00:00:35,850
by introducing particle physics

12
00:00:35,850 --> 00:00:39,510
and explaining how that
relates to remote sensing.

13
00:00:39,510 --> 00:00:42,390
We also touched upon the
electromagnetic spectrum

14
00:00:42,390 --> 00:00:45,190
and talked about how radiation

15
00:00:45,190 --> 00:00:49,680
differs in terms of
wavelength and frequency

16
00:00:49,680 --> 00:00:51,570
and how there's a lot of different sources

17
00:00:51,570 --> 00:00:55,560
of electromagnetic radiation,

18
00:00:55,560 --> 00:00:56,940
including the sun,

19
00:00:56,940 --> 00:00:59,350
which is one of the most important sources

20
00:01:00,600 --> 00:01:03,033
of energy when it comes to remote sensing.

21
00:01:04,230 --> 00:01:08,220
We also talked about how solar
radiation going from the sun

22
00:01:08,220 --> 00:01:10,293
to the earth back to the sensor,

23
00:01:11,220 --> 00:01:16,220
can be sort of lost at
different points of that path.

24
00:01:16,920 --> 00:01:21,000
So energy from the sun can be scattered

25
00:01:21,000 --> 00:01:23,790
or absorbed as it goes
from the sun to the earth,

26
00:01:23,790 --> 00:01:25,500
and it can also be scattered

27
00:01:25,500 --> 00:01:30,500
and absorbed going from the
earth back to the sensor.

28
00:01:30,780 --> 00:01:34,020
And then finally, we talked
about spectral signatures

29
00:01:34,020 --> 00:01:38,550
and how different objects
or land covers on the earth

30
00:01:38,550 --> 00:01:43,510
can emit light that has quite different

31
00:01:45,360 --> 00:01:48,690
percent reflectances at
different wavelengths.

32
00:01:48,690 --> 00:01:52,260
And those spectral signatures
can help us identify

33
00:01:52,260 --> 00:01:57,260
or define those objects when
we are looking at imagery.

34
00:02:01,140 --> 00:02:03,990
Remote sensing instruments or sensors

35
00:02:03,990 --> 00:02:07,080
measure what travel back from the object

36
00:02:07,080 --> 00:02:09,570
or the earth's surface to the sensor.

37
00:02:09,570 --> 00:02:12,900
This figure depicts many but
not close to all of the tools

38
00:02:12,900 --> 00:02:16,890
that we humans utilize to
remotely sense our world.

39
00:02:16,890 --> 00:02:19,560
Though we won't dive into
each one of these tools,

40
00:02:19,560 --> 00:02:21,480
I wanted to put this here

41
00:02:21,480 --> 00:02:24,000
to show you how many different
kinds of tools there are

42
00:02:24,000 --> 00:02:26,520
and how big of an opportunity there is

43
00:02:26,520 --> 00:02:29,250
to use remotely sense data

44
00:02:29,250 --> 00:02:32,820
to understand different
aspects of the earth's surface.

45
00:02:32,820 --> 00:02:35,460
You can also start to
imagine that these tools

46
00:02:35,460 --> 00:02:38,160
can be used in conjunction
with one another

47
00:02:38,160 --> 00:02:40,140
or in conjunction with other kinds

48
00:02:40,140 --> 00:02:42,460
of non-remotely sense data

49
00:02:43,440 --> 00:02:47,040
to gain a deeper understanding
of the earth's surface.

50
00:02:47,040 --> 00:02:49,530
So for example, you could
use data from satellites,

51
00:02:49,530 --> 00:02:52,350
with data from meteorological balloons.

52
00:02:52,350 --> 00:02:55,170
You could use data from
drones with soil sampling.

53
00:02:55,170 --> 00:02:58,740
A lot of agricultural
applications do that.

54
00:02:58,740 --> 00:03:01,980
You could use underwater
acoustic or sonar sampling

55
00:03:01,980 --> 00:03:04,620
with fish DNA sampling.

56
00:03:04,620 --> 00:03:07,830
Actually, one of my first
internships in undergrad

57
00:03:07,830 --> 00:03:12,830
was using sonar videos of fish
swimming at a certain point

58
00:03:13,200 --> 00:03:15,630
in a stream in Maryland.

59
00:03:15,630 --> 00:03:17,700
And I would watch those
and in those videos

60
00:03:17,700 --> 00:03:21,960
and estimate the population
size of this fish species.

61
00:03:21,960 --> 00:03:23,790
And then we would match that data

62
00:03:23,790 --> 00:03:28,790
with data that we
collected from the field.

63
00:03:28,950 --> 00:03:30,600
So these are just a few examples

64
00:03:30,600 --> 00:03:34,380
of what you can do with
remotely sensed data.

65
00:03:34,380 --> 00:03:37,614
As you can see, the
possibilities are endless.

66
00:03:37,614 --> 00:03:40,920
And start to think, I know
it's early in the semester,

67
00:03:40,920 --> 00:03:43,680
but start to think what you might wanna do

68
00:03:43,680 --> 00:03:45,690
your final project on.

69
00:03:45,690 --> 00:03:49,620
What kind of a remotely sense data

70
00:03:49,620 --> 00:03:52,533
or questions are most interesting to you.

71
00:03:53,370 --> 00:03:54,203
In this class,

72
00:03:54,203 --> 00:03:57,720
we're gonna focus primarily
on satellite missions

73
00:03:57,720 --> 00:04:00,270
and the data that we get from them.

74
00:04:00,270 --> 00:04:02,580
And we're gonna learn
how to process, interpret

75
00:04:02,580 --> 00:04:06,513
and analyze the imagery that
comes from those satellites.

76
00:04:07,881 --> 00:04:10,950
In order to discuss
different satellite sensors,

77
00:04:10,950 --> 00:04:14,400
we must understand the
characteristics that define them.

78
00:04:14,400 --> 00:04:17,190
One important characteristic
is spatial resolution,

79
00:04:17,190 --> 00:04:20,070
which is essentially the size of a pixel

80
00:04:20,070 --> 00:04:22,290
when the sensor's data is mapped.

81
00:04:22,290 --> 00:04:24,030
Another way to understand this is,

82
00:04:24,030 --> 00:04:26,010
it's the size of the smallest area

83
00:04:26,010 --> 00:04:31,010
that can be separately recorded
as an entity on an image.

84
00:04:31,170 --> 00:04:34,200
Spatial resolution is
often expressed in meters,

85
00:04:34,200 --> 00:04:36,420
but it can also be expressed in finer

86
00:04:36,420 --> 00:04:38,700
or coarser measurements as well.

87
00:04:38,700 --> 00:04:39,960
I've seen some satellite data

88
00:04:39,960 --> 00:04:42,120
that has a resolution of 60 centimeters,

89
00:04:42,120 --> 00:04:45,450
which is pretty incredibly small.

90
00:04:45,450 --> 00:04:48,810
That's a really high-resolution image.

91
00:04:48,810 --> 00:04:51,480
In the bottom maps, I'm
showing two different images

92
00:04:51,480 --> 00:04:54,930
that are the same area or location,

93
00:04:54,930 --> 00:04:57,330
but they have different resolutions.

94
00:04:57,330 --> 00:04:58,530
You can see that the left image

95
00:04:58,530 --> 00:05:01,680
has a lower resolution,
spatial resolution,

96
00:05:01,680 --> 00:05:04,170
meaning that its pixel sizes are bigger.

97
00:05:04,170 --> 00:05:07,920
And the right image has a
higher spatial resolution

98
00:05:07,920 --> 00:05:11,160
corresponding to smaller pixel sizes.

99
00:05:11,160 --> 00:05:14,460
High resolution data is great
'cause it's more specific

100
00:05:14,460 --> 00:05:19,440
and it's easier to see objects
or features on the landscape.

101
00:05:19,440 --> 00:05:22,620
So in the right image,
it's pretty clear to see

102
00:05:22,620 --> 00:05:27,060
that the light-colored
pixels represent buildings.

103
00:05:27,060 --> 00:05:29,640
But in the left image it's not as clear

104
00:05:29,640 --> 00:05:32,850
that those light pixels are buildings.

105
00:05:32,850 --> 00:05:37,593
But high-resolution data is
not always the best choice.

106
00:05:39,390 --> 00:05:42,670
Low resolution, spatial resolution imagery

107
00:05:43,920 --> 00:05:45,810
definitely has its pros.

108
00:05:45,810 --> 00:05:48,510
So for example, it's not as data intensive

109
00:05:48,510 --> 00:05:51,510
as high-resolution imagery.

110
00:05:51,510 --> 00:05:53,790
For example, if you're doing an analysis

111
00:05:53,790 --> 00:05:56,370
over a really large area, for example,

112
00:05:56,370 --> 00:05:58,860
the entire state of Vermont,

113
00:05:58,860 --> 00:06:02,703
you don't really need super
high-resolution imagery.

114
00:06:03,780 --> 00:06:07,200
The high-resolution imagery,
it may be way too intensive,

115
00:06:07,200 --> 00:06:10,290
it may take forever to run your analyses.

116
00:06:10,290 --> 00:06:12,630
The files may be really huge.

117
00:06:12,630 --> 00:06:15,358
So in that case, it may
be more advantageous

118
00:06:15,358 --> 00:06:20,358
to have lower resolution
imagery that's easier to manage,

119
00:06:21,210 --> 00:06:25,623
but where you can still see
everything that you need to see.

120
00:06:29,100 --> 00:06:33,750
Another characteristic of
sensors is Spectral Resolution.

121
00:06:33,750 --> 00:06:36,960
Spectral resolution indicates
the sensitivity of a sensor

122
00:06:36,960 --> 00:06:41,100
to distinguish between
a range of frequencies

123
00:06:41,100 --> 00:06:44,100
or define wavelength intervals.

124
00:06:44,100 --> 00:06:46,830
Most applicably, we
use spectral resolution

125
00:06:46,830 --> 00:06:49,080
to describe the number of spectral bands

126
00:06:49,080 --> 00:06:51,240
that a sensor can measure.

127
00:06:51,240 --> 00:06:54,780
So for example, those
bands may be red, green,

128
00:06:54,780 --> 00:06:56,163
blue, near infrared.

129
00:06:57,390 --> 00:06:58,590
On the top right of this slide,

130
00:06:58,590 --> 00:07:02,220
I've shown some examples of
different spectral resolutions

131
00:07:02,220 --> 00:07:05,040
that sensors or satellites may have.

132
00:07:05,040 --> 00:07:09,930
So panchromatic means that that
sensor can measure one band,

133
00:07:09,930 --> 00:07:13,290
and it's usually visualized
as just black and white.

134
00:07:13,290 --> 00:07:16,110
Color means that it can
measure three different bands,

135
00:07:16,110 --> 00:07:18,120
so red, green, blue.

136
00:07:18,120 --> 00:07:21,000
And those bands can be combined

137
00:07:21,000 --> 00:07:23,070
to create a true color image.

138
00:07:23,070 --> 00:07:25,050
So that would basically look like...

139
00:07:25,050 --> 00:07:26,610
If you took a normal camera

140
00:07:26,610 --> 00:07:28,560
and took a photo of the landscape,

141
00:07:28,560 --> 00:07:30,570
it would look like true color.

142
00:07:30,570 --> 00:07:32,620
Multispectral means

143
00:07:33,900 --> 00:07:36,630
that sensor is starting
to measure some bands

144
00:07:36,630 --> 00:07:38,640
that are outside of the visible range,

145
00:07:38,640 --> 00:07:41,820
so near infrared or coastal blue.

146
00:07:41,820 --> 00:07:45,360
And that can help you distinguish

147
00:07:45,360 --> 00:07:49,080
between more objects in the landscape.

148
00:07:49,080 --> 00:07:50,490
And then hyperspectral

149
00:07:50,490 --> 00:07:52,500
means that hundreds of bands are measured,

150
00:07:52,500 --> 00:07:55,533
so those are quite advanced sensors.

151
00:07:56,370 --> 00:07:59,901
This figure at the bottom
just shows some examples

152
00:07:59,901 --> 00:08:04,590
of different spectral band
sets that a sensor might have.

153
00:08:04,590 --> 00:08:09,590
So for example, at the
bottom is a four band subset,

154
00:08:09,720 --> 00:08:14,010
a sensor that has red, green,
blue, and near infrared.

155
00:08:14,010 --> 00:08:18,720
A lot of the earlier sensors
had these four bands.

156
00:08:18,720 --> 00:08:20,860
And then at the top you can see a more

157
00:08:23,820 --> 00:08:25,530
complex band set.

158
00:08:25,530 --> 00:08:28,890
So this satellite, for example,
would have eight bands,

159
00:08:28,890 --> 00:08:32,640
coastal blue, blue, green,
yellow, red, a red edge,

160
00:08:32,640 --> 00:08:35,280
and then two different near infrareds.

161
00:08:35,280 --> 00:08:39,060
So that provides you more information,

162
00:08:39,060 --> 00:08:40,481
more spectral information,

163
00:08:40,481 --> 00:08:45,481
and can help you more
specifically distinguish

164
00:08:45,570 --> 00:08:48,303
between items on the landscape.

165
00:08:49,200 --> 00:08:52,530
All three of these rows in this figure

166
00:08:52,530 --> 00:08:55,050
would be multi-spectral, right?

167
00:08:55,050 --> 00:08:57,620
Because they're not just color...

168
00:08:58,950 --> 00:09:02,160
They all have near infrared,
which is non-visible,

169
00:09:02,160 --> 00:09:03,480
but they're not hyperspectral

170
00:09:03,480 --> 00:09:05,883
because they don't have hundreds of bands.

171
00:09:09,000 --> 00:09:12,720
Another type of resolution
that's important to,

172
00:09:12,720 --> 00:09:15,000
of distinguished between
different satellites

173
00:09:15,000 --> 00:09:16,800
is Temporal Resolution,

174
00:09:16,800 --> 00:09:18,900
which is also known as the revisit period

175
00:09:18,900 --> 00:09:21,420
or the length of time
it takes for a satellite

176
00:09:21,420 --> 00:09:24,570
to complete an entire orbit cycle.

177
00:09:24,570 --> 00:09:27,510
So this is basically the
time it takes for a satellite

178
00:09:27,510 --> 00:09:32,510
to return to approximately
the same location in space.

179
00:09:32,790 --> 00:09:34,170
So for example,

180
00:09:34,170 --> 00:09:36,990
it captures imagery over
Burlington, Vermont.

181
00:09:36,990 --> 00:09:40,410
Captures imagery everywhere else on earth

182
00:09:40,410 --> 00:09:45,410
and then comes back to take
another image of Burlington.

183
00:09:46,380 --> 00:09:48,390
For example, Landsat satellites,

184
00:09:48,390 --> 00:09:50,160
which we'll talk about in a few slides,

185
00:09:50,160 --> 00:09:53,460
they have a revisit period of 16 days.

186
00:09:53,460 --> 00:09:57,840
Revisit periods are
dependent on orbital speeds

187
00:09:57,840 --> 00:10:00,750
of the satellite as well
as the type of orbit

188
00:10:00,750 --> 00:10:04,105
and what the global coverage
of the instrument is.

189
00:10:04,105 --> 00:10:08,430
Coverage refers to not just how
much of the earth it images,

190
00:10:08,430 --> 00:10:13,430
but also it refers to the
swath width and the overlap.

191
00:10:14,370 --> 00:10:17,160
So swath width is the width of the imagery

192
00:10:17,160 --> 00:10:18,780
that the satellites collect.

193
00:10:18,780 --> 00:10:21,000
And then overlap refers to the overlap

194
00:10:21,000 --> 00:10:24,780
between adjacent images.

195
00:10:24,780 --> 00:10:25,920
So as you can imagine,

196
00:10:25,920 --> 00:10:29,250
if a sensor has a low swath width

197
00:10:29,250 --> 00:10:33,240
or a high overlap between two adjacent,

198
00:10:33,240 --> 00:10:37,170
you know, images, then
it may take it more time

199
00:10:37,170 --> 00:10:38,973
to complete that orbit cycle.

200
00:10:40,020 --> 00:10:41,940
One fun fact is that some sensors,

201
00:10:41,940 --> 00:10:45,270
like the sensors that are aboard

202
00:10:45,270 --> 00:10:47,430
the International Space Station,

203
00:10:47,430 --> 00:10:49,620
they don't actually
measure the entire planet

204
00:10:49,620 --> 00:10:52,623
because of the orbit path that they take.

205
00:10:53,460 --> 00:10:55,200
There's a lot of reasons why it's valuable

206
00:10:55,200 --> 00:10:58,650
to have low temporal
resolution or in other words,

207
00:10:58,650 --> 00:11:02,190
to get imagery that is
close together in time.

208
00:11:02,190 --> 00:11:04,350
For example, a lot of
regions in the world,

209
00:11:04,350 --> 00:11:06,990
like in the tropics,
they're really cloudy.

210
00:11:06,990 --> 00:11:10,200
And so the more times that
a satellite flies over

211
00:11:10,200 --> 00:11:11,880
and captures imagery,

212
00:11:11,880 --> 00:11:16,110
the higher a chance there
is that you'll get an image

213
00:11:16,110 --> 00:11:17,523
when it's not cloudy.

214
00:11:18,690 --> 00:11:21,720
You may also want imagery
close together in time

215
00:11:21,720 --> 00:11:24,540
if you're studying short-lived phenomena.

216
00:11:24,540 --> 00:11:28,230
So for example, I've studied
the effects of hurricanes

217
00:11:28,230 --> 00:11:29,133
in the tropics,

218
00:11:30,318 --> 00:11:33,210
and for those events we really wanna get

219
00:11:33,210 --> 00:11:35,100
an image right before the hurricane

220
00:11:35,100 --> 00:11:37,500
and right after the hurricane.

221
00:11:37,500 --> 00:11:40,080
There's also a lot of
reasons why we might want

222
00:11:40,080 --> 00:11:42,210
data from specific seasons.

223
00:11:42,210 --> 00:11:46,740
So for example, if we are
studying corn production

224
00:11:46,740 --> 00:11:50,970
and we wanna get images
each year in the Midwest

225
00:11:50,970 --> 00:11:53,250
when corn production is at its peak,

226
00:11:53,250 --> 00:11:56,130
it's important that the
satellite flies over

227
00:11:56,130 --> 00:12:00,030
frequently enough that we can
capture that moment every year

228
00:12:00,030 --> 00:12:01,863
at that approximate time.

229
00:12:04,890 --> 00:12:07,609
Finally, the last type of resolution

230
00:12:07,609 --> 00:12:11,733
to describe satellites is
Radiometric Resolution.

231
00:12:12,600 --> 00:12:17,130
Radiometric resolution is
basically the sensor's sensitivity

232
00:12:17,130 --> 00:12:20,790
to the magnitude of electromagnetic energy

233
00:12:20,790 --> 00:12:22,413
that it's measuring.

234
00:12:23,610 --> 00:12:27,090
Another way to describe
this is its bit depth,

235
00:12:27,090 --> 00:12:30,483
or how many levels of
data can be collected.

236
00:12:31,470 --> 00:12:35,250
This kind of resolution
can be kind of complicated

237
00:12:35,250 --> 00:12:36,450
to wrap your head around

238
00:12:36,450 --> 00:12:38,550
if you haven't worked in data science

239
00:12:38,550 --> 00:12:40,770
or computer science before,

240
00:12:40,770 --> 00:12:43,413
but I think this figure describes it well.

241
00:12:44,330 --> 00:12:47,040
So a low radiometric resolution

242
00:12:47,040 --> 00:12:49,560
might only have two levels of information

243
00:12:49,560 --> 00:12:50,970
that it can collect.

244
00:12:50,970 --> 00:12:54,780
So that would be like a
binary data collection.

245
00:12:54,780 --> 00:12:59,780
So it can, for example,
distinguish zeros and ones,

246
00:13:00,570 --> 00:13:03,120
but if you have four levels
of data you can collect,

247
00:13:03,120 --> 00:13:07,380
you can collect zeros,
one, twos, and threes,

248
00:13:07,380 --> 00:13:09,480
so then you have four different levels.

249
00:13:09,480 --> 00:13:11,490
Eight levels is more information,

250
00:13:11,490 --> 00:13:14,403
16 levels is even more information.

251
00:13:17,010 --> 00:13:19,230
I wanted to also describe bit depth

252
00:13:19,230 --> 00:13:23,880
for those of us that may not
be familiar with this concept.

253
00:13:23,880 --> 00:13:28,770
So bit depth is basically the
number to the power of two.

254
00:13:28,770 --> 00:13:32,610
So two bits means there's
four levels of data

255
00:13:32,610 --> 00:13:34,290
that that sensor can collect.

256
00:13:34,290 --> 00:13:38,770
Eight bits means there's
256 levels of data

257
00:13:39,930 --> 00:13:42,330
that that sensor can collect.

258
00:13:42,330 --> 00:13:46,440
So obviously, if a sensor
has a higher bit depth,

259
00:13:46,440 --> 00:13:51,440
then that data that it can
collect is more specific.

260
00:13:54,270 --> 00:13:57,720
So just to summarize the four
different kinds of resolution

261
00:13:57,720 --> 00:13:59,970
that I just discussed,

262
00:13:59,970 --> 00:14:02,280
I found this figure really helpful

263
00:14:02,280 --> 00:14:04,080
in distinguishing between these.

264
00:14:04,080 --> 00:14:06,210
So spatial resolution, again,

265
00:14:06,210 --> 00:14:09,060
is referring to this pixel size.

266
00:14:09,060 --> 00:14:11,220
Temporal resolution is referring

267
00:14:11,220 --> 00:14:14,190
to how frequently imagery is collected.

268
00:14:14,190 --> 00:14:17,490
Spectral resolution is how many bands

269
00:14:17,490 --> 00:14:22,490
or wavelength ranges
that a sensor can detect.

270
00:14:24,000 --> 00:14:26,370
And then radiometric resolution

271
00:14:26,370 --> 00:14:30,750
is how specific is the
data that it's collecting,

272
00:14:30,750 --> 00:14:33,573
how many levels of data
is being collected.

273
00:14:36,600 --> 00:14:39,570
Another way to define satellite sensors

274
00:14:39,570 --> 00:14:43,380
is whether they're Passive or Active.

275
00:14:43,380 --> 00:14:47,232
So passive sensors rely on natural energy.

276
00:14:47,232 --> 00:14:50,130
They record energy that
is naturally reflected

277
00:14:50,130 --> 00:14:52,080
or emitted from the earth's surface.

278
00:14:52,080 --> 00:14:54,720
So this is energy from the sun

279
00:14:54,720 --> 00:14:57,480
that travels down to the earth

280
00:14:57,480 --> 00:15:00,180
and is emitted from the earth's surface.

281
00:15:00,180 --> 00:15:02,790
The downside of passive sensors

282
00:15:02,790 --> 00:15:05,250
is that they can only detect energy

283
00:15:05,250 --> 00:15:07,530
when it is naturally occurring.

284
00:15:07,530 --> 00:15:11,730
So they can't really
capture imagery at night

285
00:15:11,730 --> 00:15:16,410
because there's not
energy being emitted back.

286
00:15:16,410 --> 00:15:19,710
Landsat sensors, which again
we'll discuss in a second,

287
00:15:19,710 --> 00:15:22,440
these are the most well-known examples

288
00:15:22,440 --> 00:15:25,240
of a passive sensor

289
00:15:26,100 --> 00:15:31,100
and they've mapped the Earth's
surface for over 40 years.

290
00:15:33,240 --> 00:15:37,620
Active sensors are the other
main category of a sensor type,

291
00:15:37,620 --> 00:15:40,560
and they're known for using
artificial forms of energy

292
00:15:40,560 --> 00:15:43,320
that they emit and they send this energy

293
00:15:43,320 --> 00:15:44,970
to the object that they're detecting

294
00:15:44,970 --> 00:15:47,550
and then they measure the response

295
00:15:47,550 --> 00:15:50,670
or how much energy they received back.

296
00:15:50,670 --> 00:15:53,970
And then perhaps also
the timing of the energy

297
00:15:53,970 --> 00:15:55,830
that they received back.

298
00:15:55,830 --> 00:15:57,160
Active remote sensing

299
00:15:58,440 --> 00:16:00,990
or sensors can transmit light or waves

300
00:16:00,990 --> 00:16:03,000
and they can control the wavelength,

301
00:16:03,000 --> 00:16:06,730
frequency, power, and
duration of the signal.

302
00:16:06,730 --> 00:16:08,970
So there's a lot of different ways

303
00:16:08,970 --> 00:16:11,340
that they can control the energy.

304
00:16:11,340 --> 00:16:15,330
Using this method, they can
measure distance, elevation

305
00:16:15,330 --> 00:16:17,790
or atmospheric conditions.

306
00:16:17,790 --> 00:16:19,800
And they're unique to passive sensors

307
00:16:19,800 --> 00:16:21,420
and that they can be used at night

308
00:16:21,420 --> 00:16:23,640
or in different weather conditions.

309
00:16:23,640 --> 00:16:28,260
For example, many types of
signals that active sensors use,

310
00:16:28,260 --> 00:16:33,260
for example, microwaves
can penetrate cloud covers.

311
00:16:33,420 --> 00:16:35,370
One common type of active sensor,

312
00:16:35,370 --> 00:16:37,980
which I'll talk about at
the end of this presentation

313
00:16:37,980 --> 00:16:41,130
is called LiDAR, which stands for

314
00:16:41,130 --> 00:16:43,173
Light Detection And Ranging.

315
00:16:44,280 --> 00:16:49,280
And this is a really useful
type of active sensor

316
00:16:49,650 --> 00:16:52,623
in Earth system modeling.

317
00:16:54,810 --> 00:16:57,360
As I mentioned, Landsat satellites

318
00:16:57,360 --> 00:17:00,510
are a really common type of passive sensor

319
00:17:00,510 --> 00:17:04,980
that we use a lot in
remote sensing analysis.

320
00:17:04,980 --> 00:17:08,250
Beginning in the 1960s,

321
00:17:08,250 --> 00:17:10,620
the US began an effort
to develop and launch

322
00:17:10,620 --> 00:17:13,590
the first civilian Earth
observation satellite.

323
00:17:13,590 --> 00:17:16,050
And this is where Landsat was born.

324
00:17:16,050 --> 00:17:18,150
There have been nine
Landsat missions to date

325
00:17:18,150 --> 00:17:21,660
with the first beginning in 1972

326
00:17:21,660 --> 00:17:24,330
that was the launch of Landsat One.

327
00:17:24,330 --> 00:17:27,810
And then Landsat Nine was the last Landsat

328
00:17:27,810 --> 00:17:31,050
that was launched in 2021.

329
00:17:31,050 --> 00:17:33,510
And it replaces Landsat Seven,

330
00:17:33,510 --> 00:17:37,170
which was in orbit but
had some complications,

331
00:17:37,170 --> 00:17:39,303
which we'll talk about at a later date.

332
00:17:41,340 --> 00:17:44,940
Here's a summary of the
different Landsat missions

333
00:17:44,940 --> 00:17:47,190
and how they differ.

334
00:17:47,190 --> 00:17:52,190
So Landsat's One, Two,
and Three were the first

335
00:17:52,410 --> 00:17:56,850
and they had a spatial
resolution of 80 meters

336
00:17:56,850 --> 00:17:59,790
and a spectral resolution of four bands.

337
00:17:59,790 --> 00:18:02,400
So they had red, green and blue bands

338
00:18:02,400 --> 00:18:04,350
and near infrared as well.

339
00:18:04,350 --> 00:18:07,503
And they had a revisit time of 18 days.

340
00:18:08,370 --> 00:18:12,810
Landsat Four and Five were
slightly more advanced,

341
00:18:12,810 --> 00:18:15,428
so their spatial
resolution improved a lot.

342
00:18:15,428 --> 00:18:19,920
It was 30 meters, their spectral
resolution was seven bands,

343
00:18:19,920 --> 00:18:24,540
and the revisit time decreased to 16 days.

344
00:18:24,540 --> 00:18:27,930
Landsat Six unfortunately
failed at launch.

345
00:18:27,930 --> 00:18:30,150
It never made it to orbit.

346
00:18:30,150 --> 00:18:33,870
So RIP, that was in 1993.

347
00:18:33,870 --> 00:18:36,180
And then Landsat Seven

348
00:18:36,180 --> 00:18:39,180
was basically what Landsat
Six was supposed to be

349
00:18:39,180 --> 00:18:41,400
with a few improvements.

350
00:18:41,400 --> 00:18:43,770
It had a spatial resolution
of 30 meters, again,

351
00:18:43,770 --> 00:18:45,640
a spectral resolution of eight bands,

352
00:18:45,640 --> 00:18:48,450
and a revisit time of 16 days.

353
00:18:48,450 --> 00:18:52,830
Landsat Eight then improved even more.

354
00:18:52,830 --> 00:18:55,680
Spatial resolution, again, 30 meters.

355
00:18:55,680 --> 00:18:59,130
It added some new bands
including the coastal blue bands,

356
00:18:59,130 --> 00:19:00,690
so it had 11 bands.

357
00:19:00,690 --> 00:19:03,030
Revisit time of 16 days.

358
00:19:03,030 --> 00:19:06,603
And then finally Landsat Nine,
which is pretty brand new.

359
00:19:08,250 --> 00:19:10,590
It's pretty similar to the Landsat Eight,

360
00:19:10,590 --> 00:19:12,668
but it has a higher bit depth,

361
00:19:12,668 --> 00:19:15,690
so a higher radiometric resolution.

362
00:19:15,690 --> 00:19:18,450
Again, spatial resolution of 30 meters,

363
00:19:18,450 --> 00:19:23,433
spectral resolution 11 bands
and revisit time of 16 days.

364
00:19:27,270 --> 00:19:31,470
I really love this video
showing the Landsat legacy.

365
00:19:31,470 --> 00:19:33,960
So all of the different Landsat missions

366
00:19:33,960 --> 00:19:36,480
and how long they were in orbit.

367
00:19:36,480 --> 00:19:40,680
You can see that some of
these Landsat satellites

368
00:19:40,680 --> 00:19:44,790
have lasted literal decades in space,

369
00:19:44,790 --> 00:19:46,920
whereas some of the earlier ones lasted

370
00:19:46,920 --> 00:19:49,170
from five to 10 years.

371
00:19:49,170 --> 00:19:50,910
It's also interesting to see

372
00:19:50,910 --> 00:19:53,010
which of these satellites
are still in orbit.

373
00:19:53,010 --> 00:19:56,460
So Landsat Seven, Eight,
and Nine are in orbit,

374
00:19:56,460 --> 00:20:00,223
though again, Landsat Seven's
data is kind of questionable.

375
00:20:00,223 --> 00:20:02,670
Again, as we'll discuss later.

376
00:20:02,670 --> 00:20:06,540
You can also see the projected lifetimes

377
00:20:06,540 --> 00:20:08,640
of Landsat Eight and Nine.

378
00:20:08,640 --> 00:20:11,100
So they project that Landsat Eight

379
00:20:11,100 --> 00:20:13,890
should be in orbit until around 2030

380
00:20:13,890 --> 00:20:18,890
and Landsat Nine hopefully
up until 2040 and beyond.

381
00:20:22,920 --> 00:20:24,480
I really love this figure showing

382
00:20:24,480 --> 00:20:26,370
the different spectral resolutions

383
00:20:26,370 --> 00:20:28,860
of the different Landsat satellites.

384
00:20:28,860 --> 00:20:32,190
So again, as you can see,
Landsat One, Two and Three

385
00:20:32,190 --> 00:20:33,330
had four bands,

386
00:20:33,330 --> 00:20:36,600
so they had a relatively
low spectral resolution.

387
00:20:36,600 --> 00:20:39,900
Landsat Four and Five had seven bands.

388
00:20:39,900 --> 00:20:43,680
Landsat Seven advanced to eight bands.

389
00:20:43,680 --> 00:20:46,210
And then Landsat Eight and Nine

390
00:20:47,510 --> 00:20:51,660
have the highest spectral
resolutions, they have 11 bands.

391
00:20:51,660 --> 00:20:53,440
So again, you can just see that

392
00:20:54,390 --> 00:20:56,850
as we've advanced these satellites

393
00:20:56,850 --> 00:20:59,490
with you know each and every one,

394
00:20:59,490 --> 00:21:02,043
we've increased the spectral resolutions.

395
00:21:05,460 --> 00:21:06,930
Finally, I wanted to talk about

396
00:21:06,930 --> 00:21:10,170
two other important passive sensors

397
00:21:10,170 --> 00:21:13,440
that we will use a lot in remote sensing.

398
00:21:13,440 --> 00:21:15,720
So the first is MODIS which stands

399
00:21:15,720 --> 00:21:19,977
for Moderate Resolution
Imaging Spectralradiometer.

400
00:21:19,977 --> 00:21:22,790
This is basically two satellites,

401
00:21:22,790 --> 00:21:25,920
the Aqua and the Terra satellites.

402
00:21:25,920 --> 00:21:28,560
They have differing spatial resolution,

403
00:21:28,560 --> 00:21:32,850
so 250 meter, 500 meter, 1,000 meter.

404
00:21:32,850 --> 00:21:35,700
Again, it's termed the
moderate resolution,

405
00:21:35,700 --> 00:21:37,800
so it doesn't claim to be high resolution.

406
00:21:38,670 --> 00:21:43,140
But where these satellites
are really advanced

407
00:21:43,140 --> 00:21:46,530
is that they have a spectral
resolution of 36 bands

408
00:21:46,530 --> 00:21:49,380
and they have a revisit
time of one to two days.

409
00:21:49,380 --> 00:21:52,050
So that is pretty darn high.

410
00:21:52,050 --> 00:21:56,640
They were launched in 1999 and 2002

411
00:21:56,640 --> 00:22:00,240
and also replaced in 2011.

412
00:22:00,240 --> 00:22:03,720
And this satellite is also really useful

413
00:22:03,720 --> 00:22:06,780
because they make a bunch
of summary data sets

414
00:22:06,780 --> 00:22:10,110
like cloud free imagery every eight days

415
00:22:10,110 --> 00:22:14,340
or a vegetation index every eight days.

416
00:22:14,340 --> 00:22:16,500
So they're really useful

417
00:22:16,500 --> 00:22:21,500
to use in larger area analysis.

418
00:22:21,810 --> 00:22:23,910
And then another passive sensor

419
00:22:23,910 --> 00:22:28,350
that I really love and
use a lot is Sentinel-Two.

420
00:22:28,350 --> 00:22:30,753
Again, this is two active satellites.

421
00:22:31,782 --> 00:22:35,250
They use two satellites to
increase the revisit time.

422
00:22:35,250 --> 00:22:39,690
It has a really relatively
high spatial resolution

423
00:22:39,690 --> 00:22:43,863
compared to other satellites
in orbit, so 10 meters.

424
00:22:44,910 --> 00:22:47,220
Three times, as you know,

425
00:22:47,220 --> 00:22:49,080
three times higher spatial resolution

426
00:22:49,080 --> 00:22:50,460
as the Landsat satellites

427
00:22:50,460 --> 00:22:53,460
which have a 30 meter spatial resolution.

428
00:22:53,460 --> 00:22:57,000
Their spectral resolution is 13 bands

429
00:22:57,000 --> 00:23:00,150
and their revisit time is five days.

430
00:23:00,150 --> 00:23:01,799
And Sentinel-Two is pretty new.

431
00:23:01,799 --> 00:23:06,033
It was launched in 2015 and 2017.

432
00:23:09,060 --> 00:23:10,260
As I mentioned earlier,

433
00:23:10,260 --> 00:23:12,960
LiDAR is a common type of Active Sensor,

434
00:23:12,960 --> 00:23:16,950
meaning that it uses artificial
or non-natural energy forms

435
00:23:16,950 --> 00:23:19,260
to measure the Earth's surface.

436
00:23:19,260 --> 00:23:22,323
LiDAR stands for Light
Detection And Ranging.

437
00:23:23,160 --> 00:23:24,060
I also wanna point out

438
00:23:24,060 --> 00:23:27,540
that LiDAR is a type of
method in remote sensing.

439
00:23:27,540 --> 00:23:30,240
It doesn't actually refer
to a specific sensor.

440
00:23:30,240 --> 00:23:33,033
Lots of sensors have LiDAR capabilities.

441
00:23:34,020 --> 00:23:35,490
In a LiDAR system,

442
00:23:35,490 --> 00:23:39,663
light is rapidly admitted from a sensor,

443
00:23:41,032 --> 00:23:42,570
from a laser in a sensor,

444
00:23:42,570 --> 00:23:44,550
kind of like a laser light strobe.

445
00:23:44,550 --> 00:23:47,220
And this light travels from
the sensor to the ground

446
00:23:47,220 --> 00:23:50,820
and then it reflects off of
objects like trees or buildings

447
00:23:50,820 --> 00:23:54,570
and the reflected energy
returns to the LiDAR sensor

448
00:23:54,570 --> 00:23:56,220
where it's recorded.

449
00:23:56,220 --> 00:23:58,200
And that sensor can
measure the time it took

450
00:23:58,200 --> 00:24:00,840
for the light to travel
to the ground and back.

451
00:24:00,840 --> 00:24:03,150
And use that to calculate the distance

452
00:24:03,150 --> 00:24:04,300
that the light traveled

453
00:24:05,310 --> 00:24:09,030
and therefore, can measure elevation.

454
00:24:09,030 --> 00:24:11,940
So LiDAR is really
useful in understanding,

455
00:24:11,940 --> 00:24:14,395
for example, the heightened
density of vegetation,

456
00:24:14,395 --> 00:24:16,470
building elevations in city,

457
00:24:16,470 --> 00:24:19,350
the elevation of the Earth's terrain.

458
00:24:19,350 --> 00:24:24,330
And therefore, data acquired
from LiDAR instruments

459
00:24:24,330 --> 00:24:27,150
is super helpful in agricultural sciences.

460
00:24:27,150 --> 00:24:30,630
For example, if you wanna
measure the height of your crops,

461
00:24:30,630 --> 00:24:34,830
LiDAR can also help
measure forest composition

462
00:24:34,830 --> 00:24:37,230
and it help estimate how much carbon

463
00:24:37,230 --> 00:24:39,540
there may be in a forest.

464
00:24:39,540 --> 00:24:42,780
LiDAR is also really valuable
in hard to reach places

465
00:24:42,780 --> 00:24:46,803
such as in the marine sciences
or in rescue missions.

466
00:24:49,140 --> 00:24:51,870
One really fun application of LiDAR

467
00:24:51,870 --> 00:24:54,330
is that it helped us
uncover more information

468
00:24:54,330 --> 00:24:58,020
about a civilization in
Guatemala called Tikal.

469
00:24:58,020 --> 00:25:00,600
So Tikal is located in an area

470
00:25:00,600 --> 00:25:03,180
that is really densely forested.

471
00:25:03,180 --> 00:25:04,800
And so while archeologists

472
00:25:04,800 --> 00:25:07,230
have been trying to
study this civilization

473
00:25:07,230 --> 00:25:09,393
in this region for decades,

474
00:25:10,470 --> 00:25:14,250
the research was hindered
by just the sheer size of it

475
00:25:14,250 --> 00:25:18,390
and also the dense canopy.

476
00:25:18,390 --> 00:25:22,380
But using LiDAR, researchers
were able to understand

477
00:25:22,380 --> 00:25:24,540
what was going on under the canopy.

478
00:25:24,540 --> 00:25:28,590
And you can see in this map here,

479
00:25:28,590 --> 00:25:30,300
a large network of this city.

480
00:25:30,300 --> 00:25:32,460
So through using LiDAR,

481
00:25:32,460 --> 00:25:35,160
we realized that Tikal was much bigger

482
00:25:35,160 --> 00:25:38,168
than we originally understood.

483
00:25:38,168 --> 00:25:40,590
If you're interested in this topic,

484
00:25:40,590 --> 00:25:43,410
I recommend that you pause the video

485
00:25:43,410 --> 00:25:47,097
and you can read this infographic
by "National Geographic."