1
00:00:00,030 --> 00:00:08,309
um hi everyone my name is Wen-Ying Wu and
I'm the PhD student and UT Austin I'm

2
00:00:08,309 --> 00:00:13,380
working on you geological sciences but
today we are going I'm going to tell you

3
00:00:13,380 --> 00:00:17,640
a new project that we are working with a
mechanical engineering group in our

4
00:00:17,640 --> 00:00:24,480
school, so we are trying to put the
flaw prediction into a hurricane

5
00:00:24,480 --> 00:00:29,550
receipt and healthcare infrastructure
modeling basically is about how to

6
00:00:29,550 --> 00:00:33,660
optimize the resources for hospital
evacuation

7
00:00:33,660 --> 00:00:39,620
doing those flood events so here you
can see the NASA satellite image that is

8
00:00:39,620 --> 00:00:46,320
a MODIS image and you can see the the
hurricane Harvey and how dramatic it is

9
00:00:46,320 --> 00:00:49,380
in the Texas Gulf region

10
00:00:51,500 --> 00:00:55,980
so I'm giving
you an overview of the Harvey event so

11
00:00:55,980 --> 00:01:02,699
Harvey is a hurricane and 2017 August
and it's bring tremendous on renzo and

12
00:01:02,700 --> 00:01:06,600
the Texas Gulf Coast on region

13
00:01:09,800 --> 00:01:14,960
and eight
minutes length for around August 26 and

14
00:01:14,960 --> 00:01:20,670
the east part in the north part after
Harvey brings a lot of rainfall in to

15
00:01:20,670 --> 00:01:26,310
Texas Gulf region so actually it and
this caused many hospital and a nursing

16
00:01:26,310 --> 00:01:29,570
home had to evacuate

17
00:01:36,200 --> 00:01:45,100
and unfortunately Harvey kills a hundred
and six lives during several days

18
00:01:53,380 --> 00:01:57,939
yeah so you might be thinking about
there are so many hospital over Texas

19
00:01:57,939 --> 00:02:02,560
and so many nursing home and we have
very limited amount of the ambulance and

20
00:02:02,560 --> 00:02:07,930
the buses so how to optimize all the
resources to get the ambulance moving

21
00:02:07,930 --> 00:02:14,680
around the Texas Gulf region and have the
limited resources optimize and save the

22
00:02:14,680 --> 00:02:23,709
life and I would like to say this
framework is have the global impact

23
00:02:23,709 --> 00:02:29,680
because globally water cycle is moving
every, every day every time everywhere

24
00:02:29,680 --> 00:02:37,480
and here is my visualization of the how
the river moving over there event, so in

25
00:02:37,480 --> 00:02:45,340
a water year you can see there is a big
seasonality of river stream flow because

26
00:02:45,340 --> 00:02:50,709
of like snow melting or the rain flow or
the free throw process over the arctic

27
00:02:50,709 --> 00:02:57,489
regions so about 35% of the
precipitation will go into the runoff

28
00:02:57,489 --> 00:03:06,370
and that accounts for 40 cubic 40,000
cubic kilometres amount of water running

29
00:03:06,370 --> 00:03:09,400
to the ocean every year

30
00:03:10,920 --> 00:03:16,260
so here's a year
water cycle of the streamflow animation

31
00:03:16,269 --> 00:03:23,049
so I run the river multi model with
using the Geo DIS is a NASA learning

32
00:03:23,049 --> 00:03:30,670
surface data assimilation product and to
simulate it how the water transfer from

33
00:03:30,670 --> 00:03:32,880
the upstream to the downstream

34
00:03:35,200 --> 00:03:39,640
so let's
move back to the Texas Gulf region so

35
00:03:39,640 --> 00:03:45,970
this is the time Frame of where and
when Harvey happens later on you will

36
00:03:45,970 --> 00:03:52,209
see a dramatic amount of the water that
is because of the way.. on the rain fall

37
00:03:52,209 --> 00:03:57,700
bring from Harvey quit was appear in
that region and would take all the

38
00:03:57,700 --> 00:04:03,220
plains around and those thought
actually is the location of the hospital

39
00:04:03,220 --> 00:04:07,090
in a nursing home, so it's the really
question that's all

40
00:04:07,090 --> 00:04:12,870
for this amount of the water moving over
the land, how do we optimize the

41
00:04:12,870 --> 00:04:16,180
resources to save the life

42
00:04:19,180 --> 00:04:22,520
so the Harvey
when it made its first land fall

43
00:04:22,520 --> 00:04:24,780
August 26 but the

44
00:04:24,790 --> 00:04:29,980
flooding last for several days because
Harvey's surrounding around in these

45
00:04:29,980 --> 00:04:39,400
regions for a long time that's beyond
our expectation so that its first test

46
00:04:39,400 --> 00:04:47,020
test case um of the project I'm doing
this semester to see how we can evacuate

47
00:04:47,020 --> 00:04:53,910
the hospital resources um in this story

48
00:04:55,320 --> 00:05:04,030
so that is our framework, um, I'm working on
the Geo science team which is about this

49
00:05:04,030 --> 00:05:12,240
part so we used national water model
which is a continental scale flood

50
00:05:12,240 --> 00:05:18,520
forecast model and we use the streamflow
and then we use a method to simulate the

51
00:05:18,520 --> 00:05:23,800
flood heads and that is for the river
flooding from them I mean the the fresh

52
00:05:23,800 --> 00:05:29,470
water flooding from the stream flow
but also we use the / model to simulate

53
00:05:29,470 --> 00:05:36,100
in the storm surge which is salient water
that will also cause flooding especially

54
00:05:36,100 --> 00:05:43,380
in the coastal region like this and that
is from the front end to the break

55
00:05:43,380 --> 00:05:49,920
backend the engineering team they would
use the stochastic optimization model so

56
00:05:49,930 --> 00:05:59,120
basically is optimization so it was
minimize the resources between the

57
00:05:59,139 --> 00:06:05,620
hospital and see how many so the final
result of that is so we was simulated

58
00:06:05,620 --> 00:06:12,700
that how the brute Map going to be
like to try to optimize this resources

59
00:06:12,700 --> 00:06:18,789
around different hospital so later on
I'm going to show you the result of two

60
00:06:18,789 --> 00:06:22,729
simulations
one with using the national water model

61
00:06:22,729 --> 00:06:32,090
only to, to account for the flooding
from the river and the other ones use

62
00:06:32,090 --> 00:06:38,120
both national water model and the slash
model to account for the flash

63
00:06:38,120 --> 00:06:43,080
freshwater flooding but also the storm
surge together

64
00:06:44,980 --> 00:06:47,180
so here's the result of

65
00:06:47,180 --> 00:06:57,530
the Ruth meds for account for the
natural water model only for the red

66
00:06:57,530 --> 00:07:03,680
dots mean the hospital that's going to
receive the patient that is come from

67
00:07:03,680 --> 00:07:08,810
the sender so the the greens out mean
that's the hospital that is really in the

68
00:07:08,810 --> 00:07:12,900
dangerous because of the flooding they
they have to send a patient to somewhere

69
00:07:12,920 --> 00:07:18,349
else and the staging area means that the
the location of the ambulance or

70
00:07:18,349 --> 00:07:27,440
original is so here we are those centers
is pick up because of those regions are

71
00:07:27,440 --> 00:07:32,570
going to be flooded based on the weather
forecast and then to the hydrological

72
00:07:32,570 --> 00:07:39,320
forecast and to the optimization model
and later I'm going to show you the

73
00:07:39,320 --> 00:07:44,720
results with another model together so
you might expect this for those coastal

74
00:07:44,720 --> 00:07:53,240
regions those hospitals though are going to
pick out because of the storm surge so

75
00:07:53,240 --> 00:08:01,460
now you can see that those hospitals in
those coastal regions are really I need

76
00:08:01,460 --> 00:08:07,040
to evacuate the patient because of the
storm surge so in this project we

77
00:08:07,040 --> 00:08:14,990
account for the fresh water but also the
storm surge together and we start to do

78
00:08:14,990 --> 00:08:18,740
this project at the beginning of
September so we have very limited

79
00:08:18,740 --> 00:08:26,960
primary results but I'm happy to make
the science from useful to usable to

80
00:08:26,960 --> 00:08:30,770
make our science have the social impact
and try to save some lives

81
00:08:30,770 --> 00:08:38,510
yeah and for this project and like to
thank for our engineering team and also

82
00:08:38,510 --> 00:08:46,149
AGU and NASA to support this conference
yeah thank you