[day 2] center presentation: iwmi
DESCRIPTION
Presented by Mir Matin at the CGIAR-CSI Annual Meeting 2009: Mapping Our Future. March 31 - April 4, 2009, ILRI Campus, Nairobi, KenyaTRANSCRIPT
1
GIS, Remote sensing and Data
Management at IWMI
Mir Matin
2
Structural Change in IWMI
IDIS GIS/RS
GIS/RS/Data
Management
(part of Information and
Knowledge group)
IWMI
CPWF
Thematic structure where projects are under one of the four themes
GIS/RS application in research projects
3
Outline
Water Availability
Water productivity
Spatial model for site selection
Data harmonization and sharing
Future activities
4
Changes in Water
Availability at Sub basin
level - BFP IGB
Luna Bharati, Priyantha Jayakody
5
Gorai River Catchment -
Bangladesh
6
Methods: SWAT Model
Description Conceptually, SWAT divides a watershed into sub
watersheds. Each sub watershed is connected through a stream channel and further divided in to Hydrologic Response Unit (HRU).
HRU is a unique combination of soil and vegetation type in a sub watershed, and SWAT simulates water balances, vegetation growth, and management practices at the HRU level.
The subdivision of the watershed enables the model to reflect differences in evapo-transpiration for various crops and soils.
Runoff is predicted separately for each HRU and routed to obtain the total runoff for the watershed.
7
Method : input data
Spatial Data• Digital Elevation Model • Land Use Map • Soil Map and Soil Properties• River network
Time Series Data• Meteorological data ( Rainfall ,
Maximum and minimum temperature, Relative humidity, Sunshine hours, Wind speed )
• Flow data
8
-4000
-2000
0
2000
4000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Inp
ut/
Ou
tpu
t (m
m)
Average annual RF (mm) Average annual ET (mm) Average annual RO (mm) Balance closer (mm)
Water Balance Results
-4000
-2000
0
2000
4000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Inp
ut/
ou
tpu
t (m
m)
Average annual RF (mm) Average annual ET (mm) Average annual RO (mm) Balance closer (mm)
Water balance at each sub basin during 1965 to 1975 (1 to 22 are sub
basin numbers)
Water balance at each sub basin during 1990 to 1997
9
Actual ET from the two
time periods:
•As expected, ET changes
can be linked to the
landuse changes in the
catchment.
•ET has decreased in Sub-
basins where landcover
has changed from Forest to
Agriculture and increased
where rice has replaced
some of the traditional
agricultural crops
10
Model Performance Evaluation
Observed and simulated flow during the calibration period for the
4th sub basin outlet. r2 is 0.96
Observed and simulated flow during the validation period
for the 4th sub basin out let. r2is 0.94
11
Water Productivity
Mapping
Cai Xueliang
12
Water productivity – the
conceptWater productivity (WP) is “the physical mass of production or the economic value of production measured against gross inflow, net inflow, depleted water, process depleted water, or available water” (Molden, 1997, SWIM 1). It measures how the systems convert water into goods and services. The generic equation is:
)/m(m inputWater
)$/m or (kg/muse waterfrom derived utputO)$/m or (kg/moductivityPrWater
23
2233
13
Why mapping water productivity
The overarching goal of Water Productivity assessment is
to identify opportunities to improve the net gain from water
by either
• increasing the productivity for the same quantum of
water; or
• reduce the water input without or with little productivity
decrease.
14
Basin WP Analysis – What to Care?
Magnitude – what’s the current status?
Spatial Variation – how does it vary within and among regions?
Causes – why is WP varying (both high and low)?
Irrigated vs. rainfed – what’s the option for sustainable development under water scarcity and food deficit condition?
Crop vs. livestock and fisheries – how is livestock and fisheries contributing to water use outputs?
Scope for improvement – how much potential for, where?
15
The Methodology
1. Data collection: production, weather data, MODIS NDVI and Land Surface Temperature (LST) products,
existing LULC maps and GIS layers, GT points;
2. Crop dominance map synthesizing;
3. Land productivity:1. district/state wise agricultural productivity map
from census;
2. Interpolating to pixel wise productivity using MODIS NDVI indices;
4. ET mapping:1. Potential ET map with FAO approach;
2. Actual ET estimation using SSEB model;
5. Water productivity mapping.
16
Data collection
A ground truth mission was conducted in India from 8th -17th
Oct, 2008
Across Indus and Gangetic river basin
>2700km covered
175 samples
• LULC
• Cropping pattern
• Agricultural productivity (cut and farmer survey)
• Water use (rainfed, surface/GW)
• Social-economic survey
17
Crop Dominance MapSynthesizing existing maps to a crop dominance map with
GT data500m, IWMI, 2003
500m, IWMI, 20051km, USGS, 1992-1993
Legend
00 Ocean and other areas
01Irrigated, surfacewater, rice, single crop
02 Irrigated, surfacewater, rice, double crop
03 Irrigated, surfacewater, rice-other crops, single crop
04 Irrigated, surfacewater, rice-other crops, double crop
05 Irrigated, surfacewater, rice-other crops, continuous crop
06 Irrigated, conjunctive use, mixed forest, rice-other crops, continuous crop
07 Irrigated, surfacewater, wheat-other crops, single crop
08 Irrigated, surfacewater, wheat-other crops, double crop
09 Irrigated, surfacewater, wheat-other crops, continuous crop
10 Irrigated, surfacewater, sugarcane-other crops, single crop
11 Irrigated, surfacewater, mixed crop, single crop
12 Irrigated, surfacewater, mixed crops, double crop
13 Irrigated, groundwater, rice-othercrops, single crop
14 Irrigated, groundwater, rice-othercrops, double crop
15 Irrigated, groundwater, cotton-other crops, single crop
16 Irrigated, groundwater, cotton, wheat-other crops, double crop
17 Irrigated, groundwater, cotton, soyabean-other crops, continuous crop
18 Irrigated, groundwater, sugarcane-other crops, single crop
19 Irrigated, groundwater, mixed crops, single crop
20 Irrigated, groundwater, plantations-other crops, continuous crop
21 Irrigated, conjunctive use, rice-other crops, single crop
22 Irrigated, conjunctive use, rice, wheat-other crops, double crop
23 Irrigated, conjunctive use, wheat, rice-other crops, double crop
24 Irrigated, conjunctive use, rice, sugarcane-other crops, continuous crop
25 Irrigated, conjunctive use, wheat-other crops, single crop
26 Irrigated, conjunctive use, cotton-other crops, single crop
27 Irrigated, conjunctive use, cotton, wheat-other crops, double crop
28 Irrigated, conjunctive use, sugarcane-other crops, single crop
29 Irrigated, conjunctive use, soyabean, wheat-other crops, double crop
30 Irrigated, conjunctive use, mixed crops, single crop
state boundary
18
A “crop dominance map” of namely year 2008 shows major crops rice and wheat area, and other mixed croplands. Watering sources are also given for IGB map.
Crop Dominance Map
19
Crop ProductivityStep 1. District wise productivity map using census data
IGB paddy rice yield map of 2005 Crop GVP map of India and Nepal
for 2005 Kharif season
20
Crop ProductivityStep 2. Pixel wise rice productivity map interpolation using MODIS data
paddy rice yield map of 2005NDVI composition
of 29 Aug – 5 Sept 2005 for rice area
MODIS 250m NDVI at rice
heading stage was used to
interpolate yield from
district average to pixel
wise employing rice yield ~
NDVI linear relationship.
21
Actual ET EstimationStep 1. Potential ET calculation (2005-09-21 as example)
Daily data from 58 weather stations
Steps:
1. Hargreaves equation for reference ET.
2. Kc approach for potential ET.
Note: Kc (FAO56) was determined by maximum
Kc values of major crop of the month
potential ET map (2005 Sept 21)
22
Actual ET EstimationStep 2. Actual ET calculation by Simplified Surface Energy Balance (SSEB) approach
Seasonal actual ET map
(2005 Jun 10 – Oct 15)
potential ET map (2005 Sept 21)
ETa – the actual Evapotranspiration, mm.
ETf – the evaporative fraction, 0-1, unitless.
ET0 – Potential ET, mm.
Tx – the Land Surface Temperature (LST)
of pixel x from thermal data.
TH/TC – the LST of hottest/coldest pixels.
CH
xHf
TT
TTET
fpa ETETET
SSEB
ET fraction map (2005 Sept 21)
MODIS LST 2005 Sept 21
23
Water Productivity MapsRice productivity (kg/m3)
Mean AVG SDV Min Max
0.618 0.618 0.306 0.09 2.5
24
Rice water productivity for 4 major IGB countries (unit: kg/m3)
Country ADMIN_NAME WP_MEAN Country ADMIN_NAME WP_MEAN
Bangladesh Chittagong 0.445 Pakistan North-west Frontier 0.451
Bangladesh Dhaka 0.496 Pakistan FAT 0.525
Bangladesh Barisal 0.533 Pakistan Azad Kashmir 0.580
Bangladesh Khulna 0.796 Pakistan Baluchistan 0.657
Bangladesh Rajshahi 0.856 Pakistan Sind 0.732
Pakistan Punjab 0.755
Average 0.625 Average 0.617
Nepal Lumbini 0.542 India Madhya Pradesh 0.393
Nepal Sagarmatha 0.556 India Himachal Pradesh 0.407
Nepal Janakpur 0.578 India Bihar 0.408
Nepal Bagmati 0.583 India Jammu & Kashmir 0.430
Nepal Gandaki 0.607 India Uttar Pradesh 0.560
Nepal Seti 0.699 India West Bengal 0.718
Nepal Bheri 0.713 India Rajasthan 0.720
Nepal Rapti 0.715 India Haryana 0.746
Nepal Narayani 0.754 India Delhi 0.818
Nepal Mahakali 0.792 India Punjab 0.833
Nepal Kosi 0.904
Nepal Mechi 0.964
Average 0.701 Average 0.603
Results: Water Productivity MapsRice productivity (kg/m3)
25
Aral
Sea
Toktogul
Kirgizstan
Galaba study site Kuva study site
Kazakhstan
Tajikistan
Lakes
Basin boundaries
Administrative provinces
Rivers
Canals
# Test sites
Uzbekistan
Irrigated area
Water productivity mapping in Central Asia
Satellite sensor data• MODIS • IRS• Landsat• Quickbird
Ground measurements• NDVI, LAI, Spectra-radiometer
reflectance• Biomass (wet, dry), yield, crop
height, canopy cover• Soil moisture, irrigation, outflow• Weather data
26
Alfalfa
Cotton
Fallow
Home garden
Orchard
Rice
Settlements
Legend
LULC Areas share
ha %
Alfalfa 858.5 8.7
Cotton 4414.9 44.5
Fallow 1853.8 18.7
Home garden 90.3 0.9
Orchard 1.4 0.0
Rice 361.8 3.6
Settlement 573.9 5.8
other 1769.9 17.8
Sum 9924.5 100.0
LULC and the areas
in Galaba site
Crop type mapping
27
Spectro-biophysical and yield modeling
Best bands Best indices
Crop Parameter Sensorsample
sizeBest
model band
R-square
Best model
band combinati
on R-square
Cotton Wet Biomass IRS 140 Exp 2 0.697 Power 2, 3 0.834
QB 41 Multi-linear 1, 4 0.813 Multi-linear 1,4; 3,4 0.506Dry Biomass IRS 136 Power 2 0.620 Power 2, 3 0.821
QB 41 Exp 2 0.521 Exp 1, 2 0.661 LAI IRS 135 Multi-linear 3, 4 0.634 Power 1, 3 0.725
QB 41 Multi-linear 2, 4 0.511 Quadratic 2, 4 0.574 Yield IRSA 14 Linear 2, 3 0.753
QBB 7 Linear 3, 4 0.610 Wheat Wet Biomass IRS 9 Quadratic 2 0.425 Quadratic 1, 3 0.678
Dry Biomass IRS 14 Quadratic 1 0.205 Quadratic 3, 4 0.309LAI IRS 18 Quadratic 4 0.8 Multi-linear 1,3; 2,3 0.465
Yield IRS 12 Linear 2, 3 0.67MaizeD Wet Biomass IRS 19 Power 2 0.815 Power 2, 3 0.871
Dry Biomass IRS 17 Exp 2 0.928 Power 2, 3 0.903LAI IRS 19 Multi-linear 1, 3 0.777 Multi-linear 1,2; 2,3 0.839
RiceE Wet Biomass QB 10 Multi-linear 1, 2 0.535 Multi-linear 1,2; 2,4 0.600Dry Biomass QB 10 Multi-linear 1, 2 0.395 Multi-linear 1,3; 2,3 0.414
LAI QB 10 Multi-linear 2, 4 0.879 Quadratic 2, 3 0.234Alfalfa Wet Biomass IRS 21 Power 2 0.838 Quadratic 1, 2 0.853
QB 8 Multi-linear 2, 4 0.772 Multi-linear1,2; 2,3;
3,40.887
Dry Biomass IRS 21 Power 2 0.817 Exp 1, 2 0.812
QB 8 Multi-linear 2, 4 0.732 Multi-linear1,2; 2,3;
3,40.867
LAI IRS 21 Power 3 0.499 Exp 3, 4 0.639QB 8 Multi-linear 1, 3, 4 0.927 Multi-linear 1,3; 3,4 0.858
The best models for determining biomass, LAI and yield of 5 crops using IRS and QB data
28
2006-04-24 2006-05-10 2006-06-11
2006-07-29 2006-08-14 2006-10-01
0 2 41 km 0 2 41 km 0 2 41 km
0 2 41 km 0 2 41 km 0 2 41 km
Crop water use mapping Evapotranspiration using Landsat ETM+ thermal data
29
Legend
<0.3
0.3-0.4
>0.4
Kg/m3
With an average value
of 0.3 kg/m3, the water
productivity map shows
explicit scope for
improvement: where
and how much.
Water productivity map
30
Spatial models for best site selections
of inland valley wetlands for rice
cultivation in Ghana
Muralikrishna Gumma, Prasad S. Thenkabail, and Fujii Hideto
31
Approach
Identify critical spatial data layers needed for the land suitability model for inland valley (IV) wetland rice cultivation;
Provide weightages to spatial data layers and for classes within each spatial data layer based on expert knowledge;
Develop spatial model that will provide answers to relevant questions and identify best sites (e.g., IV wetland rice cultivation) based on the spatial data layers and their weightages.
32
Methodology
33
Model Development
Factor
Factor
weight
Score
range
Maximu
m score
Scores
given Weighted score Factor
Factor
weight
Score
range
Maximum
score
Scores
given Weighted score
01-Annual-rainfall 1.89 1 - 5 3 3 (1.89*3)=5.67 01-Annual-rainfall 1.89 1 - 5 3 3, ( 1.89 * 3 ) = 5.67
02-PET 1.47 1 - 5 3 3,2 (1.47*3)=4.41 02-PET 1.47 1 - 5 3 3,2 ( 1.47 * 3 ) = 4.41
03-LPG 2.05 1 - 5 5 5 (2.05*5)=10.25 03-LPG 2.05 1 - 5 3 3,2 ( 2.05 * 3 ) = 6.15
04-specificdischarge 1.89 1 - 5 5 5,4,3,2,1 (1.89*5)=9.45 05-Stream order 2.05 1 - 5 3 3,2,1 ( 2.05 * 3 ) = 6.15
05-Stream order 2.05 1 - 5 5 5,4,3,2,1 (2.05*5)=10.25 07-Slope-percent 2.95 1 - 5 5 5,4,3,2,1 ( 2.95 * 5 ) = 14.75
07-Slope-percent 2.95 1 - 5 5 5,4,3,2,1 (2.95*5)=14.75 08-Lulc 1.37 1 - 5 5 5,4,3,2,1 ( 1.37 * 5 ) = 6.85
08-Lulc 1.37 1 - 5 5 5,3,2,1 (1.37*5)=6.85 09-Soils 1.53 1 - 5 3 3,2,1 ( 1.53 * 3 ) = 4.59
12-Experience in rice cultivation 1.42 1 - 5 5 5,4 (1.42*5)=7.1 10-Soil depth 1.68 1 - 5 5 5,4,3,2,1 ( 1.68 * 5 ) = 8.4
13-Agro., technology (yield) 1.11 1 - 5 4 4,3 (1.11*4)=4.44 11- Soil fertility 2.32 1 - 5 5 5,4,3,2,1 ( 2.32 * 5 ) = 11.6
14-Watermangement tech,. 1.68 1 - 5 2 2,1 (1.68*2)=3.36 16a-Major settlement 1.5 1 - 5 5 5,4,3,2,1 ( 1.5 * 5 ) = 7.5
15-Postharvest tech., 1.05 1 - 5 5 5,4,3,2,1 (1.05*5)=5.25 16b-Minor settlement 1.5 1 - 5 5 5,4,3,2,1 ( 1.5 * 5 ) = 7.5
16a-Major settlement 1.5 1 - 5 5 5,4,3,2,1 (1.5*5)=7.5 17a-Major roads 1.7 1 - 5 5 5,4,3,2,1 ( 1.7 * 5 ) = 8.5
16b-Minor settlement 1.5 1 - 5 5 5,4,3,2,1 (1.5*5)=7.5 17b-Minor roads 1.7 1 - 5 5 5,4,3,2,1 ( 1.7 * 5 ) = 8.5
17a-Major roads 1.7 1 - 5 5 5,4,3,2,1 (1.7*5)=8.5 18a-Major markets 1.4 1 - 5 5 5,4,3,2,1 ( 1.4 * 5 ) = 7
17b-Minor roads 1.7 1 - 5 5 5,4,3 (1.7*5)=8.5 18b-Minor markets 1.4 1 - 5 5 5,4,3,2,1 ( 1.4 * 5 ) = 7
18-Markets 1.4 1 - 5 5 5,4,3,2,1 (1.4*5)=7 25-Malaria 0.41 1 - 5 2 2,1 ( 0.41 * 2 ) = 0.82
19-Land tenure 1.74 1 - 5 5 5,4,3,2,1 (1.74*5)=8.7 Total Score 115.39
20-Labour force 1.53 1 - 5 5 5,4,3,2,1 (1.53*5)=7.65
21-Crdit system 1.58 1 - 5 3 3,2,1 (1.58*3)=4.74
22-Extension system 1.05 1 - 5 5 5,4 (1.05*5)=5.25
24-Incentives_net benfit 1.37 1 - 5 3 3,4,5 (1.37*3)=4.11
25-Malaria 0.41 1 - 5 3 3,4 (0.41*3)=1.23
Total score 152.46
Kumasi Tamale
Summary of Variables considered in Model runs: variable weights for layers and variable weights for classes
34
Model Output
35
Irrigated Area Mapping
for China
Cai Xueliang
36
Irrigated area mapping for China
MODIS 500m monthly NDVI as main dataset
37
Ground truth collection
±
0 680 Km
140°E
140°E
130°E
130°E
120°E
120°E
110°E
110°E
100°E
100°E
90°E
90°E
80°E
80°E
50°N 50°N
40°N 40°N
30°N 30°N
20°N 20°N
Gu F. X., Sun D.B.(84 GT)
who? CAAS(370 GT)
Hao, W.P., Gu F.
X., Xu J.(156 GT)Gu F. X., Liu S., Sun
D.B. (138 GT)
Xu J., Liu Q.(197 GT)
Liu S.,Liu Q.,
Huang(200 GT)
38
Ideal Spectral Data Bank on Irrigated crops of China
Irrigated-SW-Rice-SC
Irrigated-GW-Rice-DC Irrigated-SW-Rice-DC
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Jan-05 Feb-05 Mar-05 Apr-05 May-05 Jun-05 Jul-05 Aug-05 Sep-05 Oct-05 Nov-05 0.00
0.20
0.40
0.60
0.80
1.00
1.20
Jan-05 Feb-05 Mar-05 Apr-05 May-05 Jun-05 Jul-05 Aug-05 Sep-05 Oct-05 Nov-05
0.00
0.20
0.40
0.60
0.80
Jan-05 Feb-05 Mar-05 Apr-05 May-05 Jun-05 Jul-05 Aug-05 Sep-05 Oct-05 Nov-05 Dec-05
0.00
0.20
0.40
0.60
0.80
1.00
Jan-05 M ar-05 M ay-05 Jul-05 Sep-05 Nov-05
Irrigated-SW-Sugarcane-SC
39
Data Harmonization and
Sharing
40
Progress
Research data management policy and procedure for projects
IDIS user interface improved
Implementing open source version of IDIS to replicate at regional offices for localized access
Integrating IDIS and IWMIDSP for one stop data portal for all data
Develop frame work for water resources audit –case study for Sri Lanka
IWMI Geo Network updated to version 2.2 with PostGreSQL
41
IDIS – Improved Interface
42
Browse Available Data
43
Time series Data Download
44
Download Survey Data
45
IWMIDSP Data Access
http://www.iwmidsp.org/
River Basins
Nations
Regions
Entire Globe
Web access ftp access
46
IDIS – IWMIDSP Integration
Provide one stop access to all IWMI data and metadata
Data access by category, location, country, basin, project, sources
Decentralized across regions
47
Water Resources Audit Framework
Organize data and information to support water resources assessment for a country
Provide access to data and maps on water availability, productivity, poverty, quality, disaster and governance
First case study for Sri Lanka
48
Discussion
49
Thanks!