Transcript 投影片 1

AgMIP-Pakistan Kickoff Workshop
&
International Seminar on Climate Change
Application of Remote Sensing Technologies to
Mitigate the Impacts of Climate Change on
Crop Production
Yuei-An Liou
Center for Space and Remote Sensing Research
National Central University, Taiwan
President, Taiwan Group on Earth Observations
Email: [email protected]
Yesterday
Center for Space and Remote Sensing Research
2/36
Today
Hydrology Remote Sensing Laboratory (HRSL)
-- Briefing
-- Crop Yield Remote Sensing
3/36
HRSL (1/2)

Land surface processes modeling (freezing, prairie)

Land surface monitoring (soil moisture, evapotranspiration, heat flux,
biomass)

Land
land
use/change
studies:
precision
farming,
agricultural
applications (Taiwan, China, Thailand, 311 Japan); natural disasters
monitoring & mitigation & reduction; regional climate (heat island effect)

Atmosphere (water vapor, typhoon, profiles & waves) by microwave
radiometers, ground- and space-based GNSS approach (radio occultation,
e.g. Formosat3)

Weather forecast (typhoon, extreme weather events)
4
HRSL (2/2)

Cryosphere and Global Warming

Glaciers:
Arctic (2010 Ilulissat Icefjord)
Antarctica
Mainland China
5
Content - Crop Yield Remote Sensing
Remote Sensing on Crop Production
The Great East Japan Earthquake (311)
Impacts of Climate Change on Crop Production
6/36
Remote Sensing on Crop Production –
motivations


7 /36
With the advantage of assessing environmental
change over a large area, remotely sensed imageries
have been extensively used to acquire a wide variety
of information of the earth’s surface.
As the globe is facing more and more unpredictable
natural disasters, the application of remotely sensed
technique on estimation of lost crop yield is vital for
taking further mitigation actions linking to climate
change.
Remote Sensing on Crop Production

A newly developed Rice field Identification and riCe
yield Estimate (RICE) algorithm is utilized to perform
remote sensing of crop production. The RICE
algorithm consists of masking (including forest,
building, cloud, and products of water area & DEM),
identification of rice field, and rice yield estimate.
MOD44W(white: water)
8 /36
SRTM DEM
Remote Sensing on Crop ProductionFlowchart
9 /36
Remote Sensing on Crop Production

Images pre-processing: 1) Mosaic images; 2) Convert
Map Projection as Geographic Lat/Lon-WGS84 (or
other coordinate system), & locate and resample the
study areas; 3) Stack Layers; 4) Calculate MODIS
including NDVI, EVI, LSWI, and NDBI (spatial
resolution: 250 m).
Text
10 /36
MODIS
Images
MATLAB
Remote Sensing on Crop Production
irrigation
LSWI(Land Surface Water
Index)≧NDVI or EVI (Vegetation
index)
11 /36
NDVI=Normalized Difference Vegetation Index & EVI= Enhanced Vegetation Index
Remote Sensing on Crop ProductionMODIS-derived
case study 1
Spatial distribution of paddy over Taiwan

Changhua
2006
2007
2006
2007
Data from National Land
Surveying and Mapping Center(2006)
12 /36
2008
First period of paddy
2008
Second period of paddy
Remote Sensing on Crop ProductionMODIS-derived
case study 1
Spatial distribution of paddy over Taiwan

Yunlin
First period of paddy
Published by Agriculture
and Food Agency(2006)
Second period of paddy
Published by Agriculture
13 /36 and Food Agency(2006)
2006
2006
2007
2007
2008
2008
First period of paddy
Second period of paddy
Remote Sensing on Crop Productioncase study 1
MODIS-derived
Spatial distribution of paddy over Taiwan

Chiayi
First period of paddy
Published by Agriculture
and Food Agency(2006)
Second period of paddy
Published by Agriculture
and Food Agency(2006)
Taiwan
14 /36
2006
2006
2007
2007
2008
2008
First period of paddy
Second period of paddy
Remote Sensing on Crop ProductionMODIS-derived
case study 1
Spatial distribution of paddy over Taiwan

Tainan
2006
First period of paddy
Published by Agriculture
and Food Agency(2006)
15 /36
Second period of paddy
Published by Agriculture
and Food Agency(2006)
2006
2007
2007
2008
2008
First period of paddy
Second period of paddy
Remote Sensing on Crop Productioncase study 1 Comparison of MODIS-imagery-derived and official paddy yields. (Unit: ton)
Year
Area
First
Chang Official
hua
RICE Estimate
Yunlin
2007
Second
First
2008
Second
First
Second
175333
134944
176124
81776
182582
90870
164823
78115
130915
82804
168526
70614
Diff (%)
-6
-42
-26
1
-8
-22
Official
182087
90132
169861
52673
200635
64914
RICE Estimate
224110
65544
193073
74599
163658
67244
23
-27
14
42
-18
4
116633
84556
115354
62002
119001
66593
129334
70584
122636
64753
97215
74482
11
-17
6
4
-18
12
95827
53294
94775
37234
96036
39393
84276
46087
102124
37794
90433
43420
-12
-14
8
2
-6
10
7.6
-26.5
1.6
9.5
-12.7
0.2
Diff (%)
Chiayi Official
RICE Estimate
Diff (%)
Tainan Official
RICE Estimate
16 /36
2006
Diff (%)
Overall error
Remote Sensing on Crop Productioncase study 2



17 /36
Northeastern Thailand is one of the representative
rainfed lowland rice agriculture areas in Asia, where rice
yield is limited due to unstable rainfall and poor soil.
Heavy monsoon rainfall over central and northern
Thailand began in July 2011 and lasted until October,
causing a great impact on national agriculture.
We applied the RICE algorithm by using the MODIS data
to estimate the loss of paddy yield after the severe
flooding events.
Remote Sensing on Crop Productioncase study 2

The flooded map over northeast Thailand in 2011
was drawn by THA_flood map_111013 (from OCHA,
United Nations Office for the Coordination of
Humanitarian Affairs) using Editor tool of GIS.
Severe flooding area on
Northeast Thailand
18 /36
Remote Sensing on Crop Productioncase study 2

Rice paddy map comparison using MODIS data (a)
and (b) IRRI data.
19 /36
IRRI is the abbreviation of International Rice Research Institute
Remote Sensing on Crop Productioncase study 2

20 /36
To predict the toll of rice paddy, we overlay the
flooded map with the estimated rice paddy from
MODIS imagery. The influenced region by the severe
flooded area is approximately 7,890,850.86 ha, which
occupied 43.13% of the northeast Thailand area.
Remote Sensing on Crop Productioncase study 2



21 /36
The damage of rice paddy
The rice paddy planted area influenced by severe
flooded area is about 123,950 ha, which is 2.32% of
total rainfed rice planted area (5,336,369 ha). The
corresponding affected rainfed rice yield is about
227,304 tons.
Even though the rice planted area is not seriously
influenced by the severe flood, the rice planted
condition and harvest in the region would be likely
influenced in the near future.
The Great East Japan Earthquake (311)


22 /36
A 9.0 magnitude earthquake struck Japan on 11 March 2011,
triggered an extremely destructive tsunami that hit the Tohoku
region of Japan severely.
On 12 September 2012, a Japanese National Police
Agency report confirmed 15,883 deaths, 6,144 injured, and
2,676 people missing across twenty prefectures, as well as
129,225 buildings totally collapsed, with a further 254,204
buildings 'half collapsed', and another 691,766 buildings
partially damaged.
Impacts of Climate Change on Crop
Production


23 /36
The Tohoku region is located in the northeastern
portion of Honshu, the largest island of Japan.
Miyagi and Fukushima are the most damaged
prefectures by the Great East Japan Earthquake.
Impacts of Climate Change on Crop
Production

2010 Official rice production (MAFF)
Region
Miyagi
Rice area(ha)
(%)
Rice yield(ha)
(%)
Fukushima
Tohoku
Japan
73,400
80,600
419,300
1,625,000
4.52
4.96
25.80
100
400,000
445,700
2,339,000
8,478,000
4.72
5.26
27.59
100
24 /36
MAFF: Ministry of Agriculture, Forestry, and Fishes of Japan.
Impacts of Climate Change on Crop
Production

25 /36
The pre- and postearthquake MODIS multispectral images (250 m)
are collected in Tohoku
after tsunami from the
MODIS Website
(http://modis.gsfc.nasa.go
v/).
Impacts of Climate Change on Crop
Production


26 /36
The standard MODIS products are organized in a tile
system with the sinusoidal projection.
We obtained 23 tiles including Jan., Apr., May, June,
July, Nov., and Dec. 2010 of MODIS Surface
Reflectance 8-Day L3 Global 250 m (MOD09Q1) and
500 m (MOD09A1) imageries from NASA LP DAAC
(http://lpdaac.usgs.gov/) to calculate vegetation
indices.
Impacts of Climate Change on Crop
Production

27 /36
Images mosaic using ENVI
Impacts of Climate Change on Crop
Production

Indices calculation
ρ means reflectance, NIR is near infrared
(841-845 nm), the wavelength of red band is
620-670 nm, blue band is 459-479 nm, and
SWIR is shortwave infrared (1628-1652 nm).
28 /36
Impacts of Climate Change on Crop
Production


29 /36
Parameters used for each mask to distinguish paddy
from other land cover
Exclude (image) areas with cloud cover.
Parameters
Paddy
Cloud
Building
Forest
Snow
Study area
LSWI +
0.05≧ EVI or
LSWI +
0.05≧ NDVI
Blue ref.
≧0.08
NDBI>0
NDVI≧0.6
or EVI>0.45
& LSWI>0.1
NDSI>0.4
Impacts of Climate Change on Crop
Production


According to the historical data from Japan MAFF, the
flooding period of Fukushima and Miyagi is May per year.
Determination formula of Paddy:
LSWI + T≧ EVI or LSWI + T≧ NDVI
where T (threshold) can be varying and indeed depends
on the local rice planting system, such as
flooding/transplanting practices, and single, early, or late
rice growth period. In this study, a global threshold value
of 0.05 recommended by Xiao et al. is adopted.
30 /36
Impacts of Climate Change on Crop
Production

Comparison of total rice field and yield in Miyagi and
Fukushima derived from MODIS imagery with the
statistic data from MAFF.
Statistics
31 /36
MAFF
MODIS
Diff.
Diff. (%)
Miyagi (ha)
73,400
76,676
3,275
4.46
(ton)
400,000
412,066
12,066
3.01
Fukushima (ha)
80,600
89,050
8,450
10.48
(ton)
445,700
478,752
33,052
7.41
Impacts of Climate Change on Crop
Production


32 /36
Disaster loss in rice field.
The disaster losses in rice
field are subsequently
calculated, 1,932.52 ha for
Miyagi and 718.43 ha for
Fukushima, accounting for
2.63% and 0.89% of the total
rice planting areas of the two
prefectures, respectively.
Impacts of Climate Change on Crop
Production

Disaster loss in rice yield
Statistics
Disaster loss of
rice yield (ton)
Disaster loss of
rice yield (%)
Miyagi
9,472.60
2.37
Fukushima
2,939.10
0.66
Image: Dave Tappin
33 /36
Image: Dr. Toshiaki Mizuno
http://biofreshblog.com/2011/04/04/how-the-japanese-earthquake-may-drastically-impact-freshwater-ecosystems/
http://www.bgs.ac.uk/research/highlights/2011/japanTsunamiFieldWork.html
Impacts of Climate Change on Crop
Production-Conclusions

The disaster losses in rice field are found to be 1,932.52 ha for
Miyagi and 718.43 ha for Fukushima. They will result in
corresponding expected losses of rice yield by 9,472.60 tons
and by 2,939.10 tons, respectively, equivalent to a direct total
loss of $US 31 Mio in a year (based on an exchange rate of 1
USD vs. 80 JPY).
34 /36
Mio= million
Impacts of Climate Change on Crop
Production-Conclusions

35 /36
It is thus estimated that the direct economic loss in
total agricultural products will be around $US 1411
Mio in a year since rice yield of Miyagi and Fukushima
accounts for about 2.2 % of the value of all kinds of
agricultural products.
Impacts of Climate Change on Crop
Production-Conclusions


Nevertheless, the situation is even worse with the
contamination of nuclear radiation.
It is inevitably that economic impact will persist for
decades.
Remote
Sensing
Natural
disaster
GIS
36 /36
Resource
loss
Economic Statistics
impact Data
Conclusion



37 /36
Satellite imagery can be used to monitor the
environmental change after severe natural disaster
timely.
A Rice field Identification and riCe yield Estimate
(RICE) algorithm is developed to identify the rice
paddy/field and estimate its yield, which is useful to
assess the loss in rice paddy production associated
with disasters immediately.
Impacts of climate change on crop production may be
conducted in future with the application of the RICE
algorithm.
Reference (remote sensing)






38/36
Liou,Y.-A.*, H.-C. Sha, T.-M. Chen, T.-S. Wang, Y.-T. Li, Y.-C. Lai, M.-H. Chiang, and L.-T. Lu,
2012/12: Assessment of disaster losses in rice field and yield after tsunami induced
by the 2011 Great East Japan earthquake. Journal of Marine Science and Technology, 20(6),
618-623, doi: 10.6119/JMST-012-0328-2.
Chang, T.-Y., Y.C. Wang, C.-C. Feng, A.D. Ziegler, T. W. Giambelluca, and Y.-A. Liou, 2012/6:
Estimation of Root Zone Soil Moisture using Apparent Thermal Inertia with
MODIS Imagery over the Tropical Catchment of Northern Thailand. IEEE Journal of
Selected Topics in Applied Earth Observations and Remote Sensing, 5 (3), pp. 752-761, doi:
10.1109/JSTARS.2012.2190588. (June 2012)
Lin, C.Y., H.-M. Hsu, Y.-F. Sheng, C.-H. Kuo, and Y.-A. Liou, 2011, Mesoscale Processes for
Super Heavy Rainfall of Typhoon Morakot (2009) over Southern Taiwan,
Atmospheric Chemistry and Physics, 11, 345–361, 2011, doi:10.5194/acp-11-345-2011.
Wang, Y.-C., T.-Y. Chang, Y.-A. Liou, and A. Ziegler, 2010: Terrain correction for increased
estimation accuracy of evapotranspiration in a mountainous watershed. IEEE Geosci.
Remote Sensing Letters, 7(2), pp. 352-356, April 2010, doi: 10.1109/LGRS.2009.2035138.
Chang, T.-Y., Y.-A. Liou*, C.-Y. Lin, C.-S. Liu, and Y.-C. Wang, 2010/7: Evaluation of surface heat
fluxes in Chiayi plain of Taiwan by remotely sensed data. Int. J. Remote Sensing, 31(14), pp.
3885-3898, DOI: 10.1080/01431161.2010.483481.
Lin, C.-Y., F. Chen, J.C. Huang, W.-C. Chen, Y.A. Liou, and W.-N. Chen, 2008b: Urban heat island
effect and its impact on boundary layer development and land-sea circulation
over Northern Taiwan, Atmospheric Environment, 42, 5639-5649, doi:10.1016/j.atmosenv.2008.03.01.
Reference (GNSS Meteorology/Climate -1)






Chane Ming, F., C. Ibrahim, S. Jolivet, P. Keckhut, Y.-A. Liou, and Y. Kuleshov, 2013: Observation and a
numerical study of gravity waves during tropical cyclone Ivan~(2008), Atmos. Chem. Phys. Discuss.,
13, 10757-10807, doi:10.5194/acpd-13-10757-2013, 2013.
Pavelyev, A.G., Y.-A. Liou, et al., 2012/1: Identification and localization of layers in the ionosphere
using the eikonal and amplitude of radio occultation signals. Atmos. Meas. Tech., 5, 1–16,
doi:10.5194/amt-5-1-2012.
Aragon-Angel, Angela, Yuei-An Liou, et al., 2011/09: Improvement of retrieved FORMOSAT3/COSMIC electron densities validated by using Jicamarca DPS measurements. Radio Science,Vol
46, RS5001, DOI:10.1029/2010RS004578, 1 SEP 2011.
Pavelyev, A.G., K. Zhang, S.S. Matyugov, Y.-A. Liou, et al. 2011/02: Analytical model of bistatic reflections
and radio occultation signals. Radio Science,Vol. 46, RS1009, doi:10.1029/2010RS004434.
Chen, Q.-M., S.-L. Song, S. Heise, Y.-A. Liou*, et al., 2011/1: Assessment of ZTD derived from
ECMWF/NCEP data with GPS ZTD over China, GPS Solutions, 15 (4), pg. 415-425, DOI
10.1007/s10291-010-020.
Pavelyev, A.G., Y.-A. Liou*, et al., 2010: Analytical model of electromagnetic waves propagation and
location of inclined plasma layers using occultation data. Progress in Electromagnetics Research (PIER),
pp. 177-202, July 2010, doi: 10.2528/PIER10042707.
 Pavelyev, A.G., Y.-A. Liou*, et al., 2009: Eikonal acceleration technique for studying of the
earth and planetary atmospheres by radio occultation method, Geophys. Res. Lett., 36,

39/36
L21807, doi:10.1029/2009GL040979.
Lee, C. C., Y.-A. Liou, et. al, 2008: Nighttime medium-scale traveling ionospheric
disturbances detected by network GPS receivers in Taiwan. J. Geophys. Res., Vol.
113,A12316, doi:10.1029/2008JA013250, 2008.
Reference (GNSS Meteorology/Climate -2)

Wang, C., Y.-A. Liou*, and T.Yeh (2008), Impact of surface meteorological measurements on GPS
height determination, Geophys. Res. Lett., 35, L23809, doi:10.1029/2008GL035929.

Chiu, T.-C., Y.A. Liou*, W.-H.Yeh, and C.-Y. Huang, 2008: NCURO data retrieval algorithm in
FORMOSAT-3 GPS radio constellation mission, IEEE Trans. Geosci. Remote Sensing, Vol. 46, No. 11,
doi:10.1109/TGRS.2008.2005038.

*Liou,Y.-A., A.G. Pavelyev, et. al, 2007: FORMOSAT-3 GPS radio occultation mission: preliminary
results, IEEE Trans. Geosci. Remote Sensing, Vol. 45, No. 10, pp. 3813-3826, doi:10.1109/TGRS.2007.903365.

Pavelyev, A.G., Y.-A. Liou*, et al. , 2010: Analytical model of electromagnetic waves propagation
and location of inclined plasma layers using occultation data. Progress in Electromagnetics Research
(PIER), pp. 177-202, July 2010, doi: 10.2528/PIER10042707.

*Liou,Y.-A., and A. G. Pavelyev (2006), Simultaneous observations of radio wave phase and
intensity variations for locating the plasma layers in the ionosphere, Geophys. Res. Lett., 33, L23102,
doi:10.1029/2006GL027112.

*Liou,Y.-A., A.G. Pavelyev, et al., 2006: Application of GPS radio occultation method for
observation of the internal waves in the atmosphere, J. Geophys. Res., 111, D06104, doi:
10.1029/2005JD005823.

*Liou,Y.A., A.G. Pavelyev, and J. Wickert, 2005: Observation of the gravity waves from GPS/MET
radio occultation data. J. Atmos. Solar-Terr. Phys., 67(3), 219-228, February 2005,
doi:10.1016/j.jastp.2004.08.001.
*Liou,Y.-A.,Y.-T. Teng,T.Van Hove, and J. Liljegren, 2001b: Comparison of precipitable water
observations in the near tropics by GPS, microwave radiometer, and radiosondes. J. Appl.
Meteor, 40(1), 5-15.

40/36
Thanks for your
attention