Book on
Global Irrigated Areas
An Assessment from Spaceborne sensors, other Geospatial techniques, and from National Statistics

Overview
On the sidelines of the Global Irrigated Area Mapping (GIAM) International Workshop, an interested group met to discuss the possibilities of bringing out a peer-reviewed publication on Global irrigated area mapping. After much discussions, majority agreed to:

“Bring out a book on irrigated areas that will be, preferably, preceded by an e-book”

To move forward in this, the following points have been identified.

I. Title
Global Irrigated Areas
An Assessment from Spaceborne sensors, other Geospatial techniques, and from National Statistics
Editors: Prasad S. Thenkabail, Hugh Turral, and Chandrashekhar M. Biradar

III. Peer review panel
Prasad S. Thenkabail (coordinator)
Hugh Turral
Mutulu Ozdogan
Eddy De-Pauw
Jerry Knox

 IV. Manuscript submission
The book chapters will be peer-reviewed before publication. Manuscripts will be coordinated by Prasad S. Thenkabail (p.thenkabail@cgiar.org). Please submit the manuscripts directly to Prasad (either as e-mail attachment or ftp upload\download). Please contact Prasad / Chandru if you need ftp upload location. You can also set up a ftp location and send an e-mail to Prasad to have it downloaded.

Prasad will coordinate with the peer-review panel members (see section III) to have the papers reviewed. Every peer review member will handle few papers and have them peer-reviewed by at least 2 independent reviewers.

V. Page limits and formats
The full length articles must follow the following minimum standard:

A. Length: less than 50 pages including figures and tables
B. Font: times new roman 12”
C. spacing: double, single sided
D. Front page: with title, key words, and contact details of the author\s including e-mail
E. Figures: high resolution tiff. Up to 5 color images allowed (this will depend on agreement with publisher and may change).
F. Text, figures, and tables: for review insert tables and figures at appropriate portion of the text.

VI. Time line for manuscript submission, review, and publication

VII. Accepted article instructions
Detailed instructions will be sent to authors of accepted articles for preparation of the final manuscript.

Appendix I – Email List
loveland@usgs.gov
h.turral@cgiar.org
p.thenkabail@cgiar.org
s.siebert@em.uni-frankfurt.de
ozdogan@hsb.gsfc.nasa.gov
wataru@iis.u-tokyo.ac.jp
roland.geerken@yale.edu
xiangming.xiao@unh.edu
a.mobin@cgiar.org
yousc@igsnrr.ac.cn  
obireddy@nbsslup.ernet.in
na_narayana@hotmail.com   
drroohi@hotmail.com  
aansari@isa.ir
pkgupta@sac.isro.gov.in
seelan@aero.und.edu
bkurz@undeerc.org
vvrao@nrsa.gov.in
j.knox@cranfield.ac.uk
senay@usgs.gov
c.biradar@cgiar.org
russ.congalton@unh.edu
jaishree_kumkar@rediffmail.com
franciscanicius@yahoo.com
w_schweers@caas.ac.cn
qinzh@caas.net.cn


Appendix II - Submitted abstracts
A Global Irrigated Area Map (1999) using remote sensing through advanced Methods*

Thenkabail P.S.¹, Biradar C.M.¹, Turral H.¹, Noojipady, P.¹,
Li, Y.J.¹, Vithanage, J.¹, Dheeravath, V.¹, Velpuri, M.¹, Schull M.², Cai, X. L., ¹, Dutta, R¹

¹International Water Management Institute ²Boston University

A number of new methods and techniques were developed. The study first segmented the world into climate and elevation zones and analyzed satellite images separately for these zones. The class identification and labeling process begins with spectral matching technique (SMTs). Since time-series data are analogous to hyperspectral data, we adopted hyperspectral analysis techniques such as SMTs to identify, group, and label classes with similar time series characteristics. The time-series spectra of classes were also compared with the target ones obtained from ground truthed locations. The spectral correlation similarity was found to be the most useful spectral matching technique (SMTs). The SMTs are followed by selection of 30-50 random spots  for each class that are spatially well distributed and linking them to “zoom in” view of Google earth images for which the resolution varies between sub-meter to 30-meter. In addition ESRI 150-m Landsat Geocover mosaic of the World, a host of image interpretation techniques such as bispectral plots, space time spiral curves (ST-SCs), time-series plots of normalized difference vegetation index (NDVI), and a host of secondary data (e.g., National and global land\use and land cover data) were used. A wide array of ground truth data were extensively used in identifying and labeling classes. First, IWMI’s ground truth data of the World with nearly 2000 points that included three missions in 2004 and 2005 covering entire India, extensive data from the  river basins with vast irrigation such as in the Ganges in India, Krishna in India, Ruhuna in Sri Lanka, Syr Darya in Central Asia, and Limpopo in Southern Africa, and a past data catalogue from Middle East, and 14 Countries in West Africa. Second, data sourced from the Degree Confluence Project with about 4000 points that collates land use data for 1 by 1 degree tile over the globe. In addition nearly 11,000 “zoom in views” of high or very high resolution Google earth points. Decision tree algorithms, NDVI time series plots, NDVI thresholds, principal component analysis, unsupervised clustering algorithms, and GIS spatial modeling using data such an agroecological zones, temperature, precipitation, evapotranspiration, and elevation were widely used to define and refine classes especially to resolve mixed classes.

A 28-class dis-aggregated global irrigated area map at 10-kilometer scale (GIAM10km-28classes) and aggregated 8-class and 3-class (GIAM10km-8 classes and GIAM10km-3 classes) maps of the world were produced. The GIAM10km-28 classes (Figure 39) provide information on watering method (irrigated or rainfed agriculture), irrigation type (surfacewater, groundwater, and conjunctive use), irrigation intensity (single, double, or continuous crop), and crop type or dominance. The GIAM10km-8 class provides watering method, irrigation type, and intensity. The 3 GIAM10km-3 classes provide information on: surfacewater irrigation, ground water irrigation, and conjunctive use (surface and ground water) irrigation. Informal (e.g., small reservoirs, tanks, and groundwater) irrigation was identified and mapped in addition to more conventional large scale surfacewater irrigation found in most irrigated area maps. Annualized irrigated areas (intensity of irrigation) were calculated using time-series satellite imagery from which one can detect how many crops are grown in a same area during a given year. Particular strengths of this work are in: (a) establishing seasonal and annualized irrigated areas (or intensity of irrigation), (b) mapping informal (e.g., small reservoirs, tanks, groundwater) irrigation in addition to conventional surfacewater irrigation, (c) determining irrigated crop calendar, (d) studying historical (e.g., last 20 years, every month) biomass dynamics for every irrigated area class and every pixel within that class. In addition, it must be noted that this is the first satellite sensor based irrigated area map of the world. The ability of remote sensing to map at various scale or pixel resolutions has been demonstrated and highlighted.

The irrigated areas in these maps were calculated based on sub-pixel areas (SPAs). The SPAs, which are areas actually irrigated, were established by multiplying the full pixel areas (FPAs) of the classes with the irrigated area fractions (IAFs) established based on: (a) Google earth estimate (GEE), (b) high resolution imagery (HRI), and (c) sub-pixel decomposition technique (SPDT). The combined coefficients from the IAF-HRI and IAF-SPDT for each of the 28 GIAM classes were used to compute robust and reliable estimates of the seasonal and annualized irrigated areas of the world. The annualized areas are summation of areas from different seasons. Cropping calendar (i.e., single, double, or continuous cropping) for each of the 28 classes were established and their SPAs for each of the season were determined by multiplying the FPA with the combined IAFs from HRI and SPDT. The annualized area is then the sum of the areas from different seasons. For GIAM this was sum of areas from seasons consisting of single, double, or continuous cropping. The IAF-GEE method provides total area available for irrigation (TAAI).

The annualized irrigated areas of the world at the end of the last millennium were 480 Mha. Of which there was: (a) 263 million hectares (Mha) for season 1, (b) 176 Mha for season 2, and (c) 41 Mha for continuous crops. The total area available for irrigation (TAAI) at any given time at the end of last millennium was 412 million hectares of which different proportion of areas are irrigated during different seasons as reported above, leading to an annualized area. The distribution of irrigated areas is highly skewed amongst continents and Countries. Asia accounts for 78 percent of all annualized irrigated areas, followed by Europe (8 percent) and North America (7 percent). South America (4 percent), Africa (2 percent), and Australia (1 percent) have very low proportion of the global irrigation. China has 111 Mha and India has 101 Mha of total area available for irrigation. In this 98 Mha; China has 75 Mha of crops during season 1, 68 Mha during season 2, and 7 Mha during season 3 for a annualized sum of 151 Mha. India has annualized sum of 132 Mha, with a break up of 72 Mha during season 1, 53 Mha during season 2, and 5.9 Mha during season 3.China and India have a staggering 59 percent of the Global annualized irrigation. This is followed by USA (5.06 percent), Russia (3.49 percent), and Pakistan (3.32 percent). Eight other Countries have areas between 1 to 2 percent. All other Countries in the World have less than 1 percent area irrigated relative to global annualized total.

The accuracies of the irrigated areas were determined using three independent  datasets: (a) first, a 1861 ground truth data points of the world sourced from degree confluence project, (b) second, a 890 point ground truth data collected by GIAM team, and (b) third, a randomly picked 670 point “zoom in views” of very high resolution imagery from Google earth. From these 3 methods, the accuracies varied between 84 to 91 percent with errors of omission of not exceeding 16 percent and errors of commission of less than 21 percent.

The IWMI Global irrigated area map (GIAM) 28-class (GIAM10km-28 classes), 8-class (GIAM10km-8 classes), and 3-class (GIAM10km-3 classes) maps, data and products are made available through a dedicated web portal: http://www.iwmigiam.org. These products are supported by class characteristics (e.g., cropping calendar), snap-shots of higher resolution imagery, time-series animations of classes showing their biomass dynamics for last 20 years, area estimation methods, accuracy assessment results, ground truth data links for classes including digital photos, source images, and background documentation on methods and materials. In addition to GIAM, data and products are also available for: (a) a Global Map of Rainfed Cropland Areas (GMRCA), (b) a Global Map of land use/land cover areas (GMLULCA), and (c) a generic IWMI 951-class Global land use/land cover (LULC) Map of the World.

Corresponding author: p.thenkabalil@cgiar.org


Global map of irrigation areas 4.0 - mapping irrigated areas by combining sub-national statistics and geo-spatial data.
Stefan SIEBERT¹, Jippe HOOGEVEEN², Petra DÖLL¹, Jean-Marc FAURES², Sebastian FEICK¹ and Karen FRENKEN^2
1 University of Frankfurt (Main), Institute of Physical Geography, Germany
² Food and Agriculture Organization of the United Nations, Land and Water Development Division, Rome, Italy

Abstract
A new version of a digital global map of irrigation areas was developed by combining irrigation statistics for 26 909 sub-national statistical units and geo-spatial information on the location and extent of irrigation schemes. The difference to map version 3 is the incorporation of a map update for Africa, Europe and parts of Latin America. The map shows the percentage of each 5 arc minute by 5 arc minute grid cell (about 86 km² along the equator) that was equipped for irrigation around the year 2000. It is thus an important data set for global studies related to land and water, but also for assessments on food security or to quantify possible impacts of climate change on agriculture. The article describes the data set and the mapping methodology and gives an estimate of map quality at the scale of countries, world regions and the globe. Two indicators of map quality that take into account the geospatial information density were developed for this purpose. The main results of the study are, that 278.8 Mio ha were equipped for irrigation at the global scale. About 68% of the total irrigated area is located in Asia, 17% in America, 9% in Europe, 5% in Africa and 1% in Oceania. The largest contiguous areas of high irrigation density are found in North India and Pakistan along the rivers Ganges and Indus, in the Hai He, Huang He and Yangtze basins in China, along the Nile river in Egypt and Sudan, in the Mississippi-Missouri river basin, in parts of California, in the Po river plain in Northern Italy and along the lower Danube. Smaller irrigation areas are spread across almost all populated parts of the world. At the global scale, the overall map quality is good, but there are large regional differences of map quality. At the level of world regions, map quality in North America (overall mark 1.03), Southern Europe (1.35), Oceania (1.44), Northern Africa (1.46), Southern Africa (1.50) and Central Asia (1.63) is best. Western Africa (2.90) and the Russian Federation (4.00) have the worst map quality. About 65 Mio ha of areas equipped for irrigation are located in countries where map quality is estimated to be very good, 187 Mio ha in countries with good map quality, 21 Mio ha in countries with fair map quality and 5 Mio ha in countries with poor map quality. Map regions of very poor map quality do not exist anymore at the country scale. Consequently about 90% of the total irrigated area of the world is located in countries where the map quality is assessed to be very good or good. It was found that remote sensing based inventories report higher values for the global extent of irrigated land and that there is a need for a systematic comparison of the different data sets.

Keywords: irrigation, irrigation map, global map, water use, agriculture, crop production, land use, land cover, crop management

Corresponding author: s.siebert@em.uni-frankfurt.de

Mapping irrigated agriculture over the Continental US using multi-temporal MODIs data
Mutlu Ozdogan^1* and Garik Gutman²
1 NASA/GSFC, Code 614.3, Greenbelt, MD 20771, USA
2 Garik Gutman, NASA/HQ, Code SMD, 300 E Street SW, Washington, D.C. 20546, USA

Abstract
Accurate information on irrigation extent is fundamental to many aspects of the Earth Systems Science and global change research.  These include modeling of water exchange between the land surface and atmosphere, analysis of the impact of climate change and variability on irrigation water requirements/supply, and management of water resources that affect global food security.  However, the current extent of irrigated areas over continental to global scales is still uncertain and available maps are derived primarily from country level statistics and spatialized using maps that are often outdated.  Even in locations, such as the US, where general extent of irrigated areas are known, irrigation-related information exists only in disparate datasets and cannot be easily synthesized into a single continental scale database.

To overcome the shortcoming of existing datasets, we developed an irrigation mapping methodology that rely on remotely sensed inputs, better classification methodology, and ever increasing continental and globally extensive ancillary sources of gridded climate and agricultural data.  More specifically, we use multi-temporal data from the MODerate Resolution Imaging Spectroradiometer (MODIS) instrument to characterize the distribution of irrigated lands over the Continental US at 1 km spatial resolution circa 2002.  Development of this mapping methodology is part of a broader effort to map irrigation over the entire globe.

We take a binary (i.e. irrigated vs. non-irrigated) supervised classification approach to the irrigation mapping problem using a non-parametric decision tree algorithm.  To help better characterize irrigation, we introduce climate-based irrigation indices that provide precursory information on irrigation occurrence along with existing datasets on agriculture presence.  We also use remotely-sensed temporal and spectral indices that maximize the irrigation-related information.  Two characteristics of irrigation that are observable with remotely-sensed measurements play a particular role in determining irrigation presence: a) temporal evolution of vegetation greenness as detected by NDVI; and b) water stress detected by spectral indices that are most sensitive chlorophyll presence. 

Implementation of our irrigation mapping procedure over the Continental US reveals a high spatial resolution map of irrigated areas with expected patterns.  For example, there is a strong east-west divide in irrigation presence, most irrigated areas concentrated on the arid western portion of the country along dry lowland valleys such as the Central Valley of California or flat plains such as the Texas Pan Handle.  The final binary irrigation map product has a better than 80 percent classification accuracy with western portion of the country having slightly better estimates than the eastern portion.  Preliminary investigation of estimating subpixel proportion of irrigation using remotely-sensed surface temperature measurements over those pixels only classified as irrigated reveals a moderate but consistent signal with the same east-west divide.

Corresponding author: ozdogan@hsb.gsfc.nasa.gov

Irrigated Area Mapping for China - Challenge and Feasibility
Songcai YOU
Inst. of Geog. Sci. & Natural Resources Research

Challenges
The challenges to map irrigated area in China come from data shortage, poor data quality, complex cropping system and various irrigation intensity in different regions.

(1)        Data Availability
-           Remote sensed data: TM/ETM images covering whole China as of late 1980s is provided, free of charge, by GLCF, according to the MOU signed between Institute of Geographic Sciences and Natural Resources Research (IGSNRR) and GLCF on Sept 3, 2005. (http://glcf.geodata.cn)
-           Cropping system data/map and irrigation intensity information, provided by experimental stations affiliated to IGSNRR in the case study area
-           Land use map: at a scale of 1:100,000~1,000,000, updated every other 5 years by IGSNRR (http://www.resdc.cn).
-           Statistical data: Data Center for Resources and Environmental Sciences in IGSNRR has started data collection and compilation for over two decades, large amount of agricultural statistical data is available now (http://www.data.ac.cn/index.asp)

(2)        Technical Facilities
IGSNRR is strong in GIS technique application, especially ARC/GIS (http://159.226.115.50/echannel/aboutus/eintroduction.asp) and SuperMap, the latter is completely developed by IGSNRR (http://www.supermap.com/)

(3)        Knowledge Capability
IGSNRR is a leading research organization in geographical sciences and resources management as well as an education body in China (http://english.igsnrr.ac.cn/). Currently, there are over 500 employees and nearly 600 students (Master and doctoral course and post-doc)

Irrigated Area Mapping Activities

To overcome above mentioned difficulties, remote sensed data will be fully applied in irrigated area mapping in China under the support of GIS and indigenous knowledge.

(1)  Study Area
North China Plain with an area of 300,000 km2 in China, including Beijing, Tianjin, Hebei, Henan, Shandong, Anhui and Jiangsu, is chose as case study area, because agricultural production in this region highly depends on irrigation.

(2)  Technical Approaches
RS data is used to extract irrigated area under the support of GIS techniques. Indigenous knowledge will be used to help the interpretation during RS data processing and to determine the irrigation intensity in different areas. Land use map is used to distinguish rainfed land especially rainfed paddy land from irrigated land.

Corresponding author: yousc@igsnrr.ac.cn


Anticipated Outputs
The final output is an irrigated area distribution map at a scale of 100,000 with properties such as percentage of irrigated area at each county, contribution of irrigated area to crops production in percentage, irrigation intensity, the potential increase of irrigated area taking into consideration of water resources availability, development and application of water-saving irrigation techniques and changes of land use.

Corresponding author: yousc@igsnrr.ac.cn

 Estimating spatio-temporal patterns of continuous Paddy field cover  using MODIS time series

Wataru Takeuchi
University of Tokyo

Abstract

Two thirds of the rice-growing areas in the World are in Asian countries and hundreds of millions of people depend on rice as their staple food source. At the same time, paddy fields have been considered to be one of the likely and most important sources of atmospheric methane since the rapid increase in atmospheric methane was recognized in the early 1980’s. The improved understanding of paddy field distribution  at large spatial scales has increased the interest in deriving crop yield and methane emission estimations. Nevertheless, the collection of such data through field surveys is time-consuming and expensive in Asian regions. Remotely sensing data from satellite images provide an alternative means of obtaining paddy field distribution. In this study, the patterns observed in metrics calculated for six month of MODIS over Japan is examined. Four analytical approached are used; calculation of temporal mean, maximum and minimum layers for selected metrics showing significant spatial variability of channel 1, 2, NDVI; linear discriminant for input into the spectral mixture analysis is derived from the same multi-temporal metrics used for the classification product using ASTER; the continuous percentage of paddy field is generated using spectral mixture analysis with the training data derived from the above mentioned ASTER data. The derived metrics are not sensitive to time of year or the seasonal cycle and can limit the inclusion of atmospheric contamination. The comparison of 250 m MODIS product with the past efforts on 1 km AVHRR derived IGBP-DIS product shows that the finer resolution and its un-mixing played a crucial role in depicting the paddy field cover over Japan

Corresponding author: wataru@iis.u-tokyo.ac.jp

Information on spatial distribution of irrigated areas and their dynamics are desirable for the planning, management and monitoring programmes at national, regional and local levels for agricultural development. With increase in spatial, spectral and temporal resolutions of the satellite sensors, medium to high resolution satellite data provides valuable information on location, spatial distribution and extent of irrigated areas in the country for accurate mapping of irrigated areas and to analyze their spatial-temporal dynamics in GIS environment with suitable spatial modeling. In heterogeneous irrigated area mapping, the time series NDVI indices derived from NOAA AVHRR 10km, MODIS (Moderate Resolution Imaging Spectroradiometer) 500m resolutions, national statistics and ground truth data play a crucial role. Initially generic classes of land use/land cover are needed to be generated using unsupervised classification for further aggregation on various parameters. In area estimation, the techniques like irrigated fractions and total irrigated area for each class need to be estimated using ground truth data, national statistics and google earth imageries. In this direction, irrigated area maps have been produced by IWMI at 10km and 500m spatial resolution using NOAA AVHRR and MODIS data respectively. The challenge is to study these maps in association with Indian national statistics for further refinement. Discussions are underway between the Indian Council of Agricultural Research (ICAR) and IWMI to study and strengthen the irrigated area maps of India generated in GIAM project by IWMI through further refining in conjunction with national statistics. The overarching goal will be to produce accurate and agreeable maps and products that will be invaluable in national planning for sustainable land and water resources.

Key words: Irrigated area mapping, Irrigation sources, MODIS, National statistics, NOAA AVHRR, NDVI indices

Corresponding author: obireddy@nbsslup.ernet.in

Obi Reddy P. Gangalakunta
Natioanl Bureau of Soil Survey and land Use Planning (ICAR), Amravati Road, Nagpur, India

1. Introduction

2. General Description and Background
2.1 Geographical settings of India
2.2 Climatic and rainfall patterns in India
2.3 A history of irrigated areas of India

3. General land use pattern: Existing knowledge base
3.1 Land use pattern in the country
3.2 Major cropping patterns in the country
3.3 Irrigated areas in the global context
            3.4 Historical background of irrigation system
3.5 Major Irrigation sources and their dynamics
3.6 Trends in irrigated areas

 4. Methodology
4.1 Collection of National statistics
4.2 MODIS, LANDSAT and IRS data analysis
4.3 Sensors data and national statistics

5. Satellite sensors data and National Statistics
5.1 National statistics on irrigated areas
5.2 Satellite sensor data at national scale
5.3 Validation of Irrigated area maps with national statistics
5.4 Refinement of irrigated area maps

6. Issues and way forward
6.1 Irrigated area mapping and scales
6.2 Analysis of dynamics and composition of irrigation sources
6.3 Analysis of Irrigated areas in association with LGP 
6.4 Analysis of changes in irrigated areas
6.5 Irrigated areas and water use efficiency
6.6 Analysis of cropping systems and cropping intensity
6.7 Irrigated areas and yield gap analysis.
6.8 Prediction of near future Irrigated areas changes in GIS.
6.9 Simulation of scenarios of irrigated areas
6.10 Analysis of impacts on Irrigated areas
6. 11 Identification of critical areas for judicious use of irrigation water
6. 12 Assess and estimate the potential irrigated areas

Corresponding author: obireddy@nbsslup.ernet.in

Mapping the spatial distribution of irrigated areas and volumetric water demand in a temperate climate: a case study in England and Wales
J.W. Knox * and E.K. Weatherhead
Institute of Water and Environment, Cranfield University, Silsoe, Bedford, MK45 4DT, UK

Abstract
In England and Wales, irrigation is supplemental to rainfall but concentrated on high value crops such as potatoes, vegetables and soft fruit, where reliable, continuous supplies of premium quality produce are demanded by the major supermarkets. Demand for irrigation is rising steadily, peaking in the driest years and in the driest catchments when water resources are most limited. New water regulation and changes in resource management and abstraction licensing are currently being implemented by the water regulator (Environment Agency) to address concerns regarding the sustainability of water withdrawals, and the potential impact that over-abstraction is having on the water environment. Increasing competition between sectors (industry, public water supply and agriculture) coupled with the longer-term threat of climate change will exacerbate the situation, with reduced summer river flows and increases in water demand being predicted. New detailed baseline information showing the temporal and spatial the extent of the areas irrigated and volumes of water abstracted for agricultural irrigation would therefore be useful for informing decision-making by water resources planners and water policy advisors seeking to redress the balance between the needs of abstractors with that of the water environment.

A procedure has been developed to model and map the spatial distribution of the irrigated areas and the volumetric irrigation water demand in England and Wales. The methodology uses a data-bridge approach to combine high resolution digital datasets relating to soils, agroclimate and land use within a geographic information system (GIS), with outputs from a daily water balance irrigation scheduling model and data derived from a national survey of irrigation. Maps showing the spatial extent of the areas irrigated (ha) and the theoretical irrigation water demand (m³), by crop type and in total, in a ‘design’ dry year have been produced.

The GIS mapping outputs have been aggregated for use at a range of levels, from region to local (sub-catchment) scales. A brief description of the individual components of the modelling procedure are provided. The integrity of the datasets used, the problems relating to error propagation and the statistical accuracy of the irrigation survey data are then discussed. The future application of the GIS modelling outputs in support of improving the management, allocation and planning of water resources for irrigated agriculture (with and without climate change) are highlighted.

Keywords: England and Wales; GIS; Irrigation; Maps; Prediction; Water demand

 * corresponding author: j.knox@cranfield.ac.uk

 Irrigated area in India over the last 50 years: Past expansion and present trends
A. Narayanamoorthy
Gokhale Institute of Politics and Economics (Deemed University), Pune, India

 Abstract
India’s irrigation sector is one of the largest in the world, both in terms of irrigated area and the volume of investment.  Government alone has invested about Rs. 1556.25 billion (in current prices) up to 2001-02 in irrigation sector.  As a result, the irrigated area of the country increased from 20.85 million hectares in 1950-51 to 78.38 million hectares in 2001-02, an increase of nearly four times. Though a large number of studies have analysed the impact of irrigation on various developmental parameters, not many studies have looked at the development of irrigation across major sources as well as across states using time series data from 1950-51 to 2002-03.  The trends in development of irrigated area at the country level may not be the same across the states owing to various reasons.  Similarly, among the three major sources (tank, canal and groundwater) of irrigation, the development may not be the same across the states.  In this study, therefore, an attempt will be made to study the past development of irrigated area across the states using time series data.  Specifically, the study will address the following aspects of irrigation: (a) Irrigation potential created up to 2001-02 in relation to ultimate irrigation potential of the country as well as different states, (b) Expansion in irrigated area in absolute term as well as its share in terms of net sown area and gross sown area, (c) Expansion in irrigated area by source (d) Growth pattern of irrigated area by source at different periods. (d) Impact of irrigation (by source) expansion on cropping intensity.   While descriptive and growth analysis will be used for this study, the study would cover all the major states for its analysis.

Corresponding author: na_narayana@hotmail.com

 Abstract
Irrigation development has been accepted as a major factor in increasing agricultural production. Irrigation forms the datum line for sustained successful agriculture. The massive development of a vast irrigation network in India has been recognised as a landmark in the history of agricultural development anywhere in the world. Ultimate irrigation potential in the country is estimated at 139.89 million ha, comprising of 58.46 M ha through Major and Medium irrigation and 81.43 M ha(  64.05 M ha from groundwater and 17.38 M ha from surface water) from Minor irrigation. Additionally, the inter-basin transfer of water from surplus to deficit basins is being pursued by Government of India is expected to bring 35 M ha under irrigation.  From 1951 to 2002, gross irrigated area (GIA) expanded more than four fold, from 22.6 M Ha to 93.96 M ha contributing to the growth in the overall cropping intensity from 111 % in 1950-51 to 132% in 2001-02 . While enormous irrigation potential of  99.36 M ha has been created at huge cost, the gap between created potential and utilisation is significantly large around 14 M. ha till the end of March,2005.

In India, estimates of UIP are not based on any river basin wide planning or survey .Seasonal imbalance in flow of rivers, geographic incongruity between regions with undeveloped water potential and those with irrigable lands; increasing competition for land and water from non irrigation sector, and due to non - conjunctive assessment of surface and groundwater are some of the factors leading to such questionable estimates. In addition to these, there are differences in the interpretation of the concept of IP creation and utilisation by the reporting agencies in the country. The system of monitoring and verifying information provided by executing agencies on both potential and actual irrigation is inadequate and casual and there are substantial reporting and compellation errors in the data.

Poor water use efficiency, low productivity and poor sustainability and land degradation (salinity and water logging) are limiting factors in water utilisation pattern in India. Conventional methods of irrigation management are based primarily on unreliable data bases, with considerable time lag in their generation, trial and error approaches and operator experience.   Now methods which use computer based capabilities of data collection, management, analysis and decision support are, therefore, needed to increase the efficiency of irrigation system operation.

A well-developed standardized national information system, with databases and networked data banks, integrating and strengthening the existing Central and State level agencies, helps ratoinalising  the irrigation data bases across the country is the need of the hour.

A conceptual frame work is proposed in this paper taking into account the developments in hydrologic, irrigation and agricultural sciences, database management systems, GIS, remote sensing and GPS facilitating aggregation of data across spatial scales. There should be a National agenda using the frame work suggested to rationalise the irrigation potential created and utilized. The recent initiatives in India in generating irrigated data bases hold promise to realize such a frame work.

Corresponding author: vvrao@nrsa.gov.in

Mapping irrigated areas of the world using remote sensing data
Abdolreza Ansari Amoli
Iran Space Agency,Tehran, Iran 

Crop yield mapping includes Irrigated and non-Irrigated area mapping . Geographic Information Systems (GIS) and remote sensing are two important technologies that enable mapping of Irrigated and non-irrigated area. The capability of data acquiring in a large area, highly potential in time-series and online data producing, are the most important and useful characteristics of remote sensing technology. In this regard, hyperspectral satellite data like MODIS are specially suitable for mapping of Irrigated area.By using MODIS data and applying different algorithms based upon plant phenology and growing stage (Multitemporal Methods) we can discriminate the agricultural crops by a very good precision.

I Suggest following  subsections ( as Draft)  for the section entitled “Mapping irrigated areas of the world using remote sensing data”:

1.Irrigatd Area Mapping

          1-1.Classical Methods
          1-2.New technologies
              1-2-1.Remote Sensing
                  1-2-1-1   Type of  Data
                             1-2-1-1-1  10 Km  Resolution
                             1-2-1-1-2    500m Resolution
                             1-2-1-1-3   30m Resolution
                   1-2-1-2   Methodologies
                             1-2-1-2-1   Classification
                                  Old methods (include Maximum Likelihood,Minimum Distance,…)
                                  Genetic Methods (include Fuzzy Classification,Neural Network,….)
                                  Multitemporal Methods

Corresponding author: aansari@isa.ir

Abstract
Accurate crop performance monitoring and production estimation are critical for timely assessment of the food balance of several countries in the world. Recently, the Famine Early Warning Systems Network (FEWS NET) has been monitoring crop performance and to some extent relative production using satellite derived data and simulation models in Africa, Central America and Afghanistan where ground-based monitoring is limited due to a scarcity of weather stations. The commonly used crop monitoring models use a crop water balance algorithm with inputs from satellite-derived rainfall. While these models provide useful monitoring for rain-fed agriculture, they are ineffective for irrigated areas. This study has focused on Afghanistan where over 80% of the agricultural production comes from irrigated agriculture.  We implemented a simplified energy balance approach to monitor and assess the performance of irrigated agriculture in Afghanistan using the combination of 1-km thermal data and 250-m NDVI from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor. Up to 19 cloud free thermal and NDVI images were used for each year to estimate seasonal actual evapotranspiration (AET) for two major irrigation river basins (Kabul and Helmand) over 6 years (2000-2005). Seasonal AET estimates were used as relative indicators of year-to-year production magnitude differences. The temporal water-use pattern of the different irrigated basins was indicative of the cropping patterns specific to the region. The results were comparable to field reports and watershed-wide crop water balance based estimates in that the 2003 seasonal AET was the highest of all six years. The advantage of this method over crop water balance methods is that the energy balance approach also helps identify spatial extents of irrigated fields and their spatial variability as opposed to a lumped watershed-wide assessment that can be obtained from large-scale water-balance models.

Corresponding author: senay@usgs.gov

Conjunctive Water use Area Dynamics and its Mapping using Multi-temporal IRS Data
Praveen Gupta¹, Sushma Panigrahy¹, Parihar, J.S¹ and Subashisa Dutta^2
1  Space Application Centre, Ahmedabad, India
² Department of Civil Engineering, Indian Institute of Technology, Guwahati, India

Abstract
Conjunctive Water use Area Dynamics and its Mapping usingMulti-temporal IRS Data P. K. GUPTA1, S. DUTTA2, S. PANIGRAHY1 and J. S. PARIHAR1 ABSTARCT Irrigated agriculture in many areas of the world, is currently being practiced from multiple water sources such as precipitation, canal, wetlands, ground aquifer, etc. Analyzing this conjunctive water use and its dynamics is an important aspect for sustainable water resource management. This chapter highlights the use of multi-scalen, multi-temporal remote sensing data and Geographic Information System (GIS) to analyzing the conjunctive water use in rice cropping system area in West Bengal, India. The Methodology has been developed and validated for the upper canal Damodar irrigation command area, for the summer season of year 2000 and 2001. The prime objective of the work is first to develop methodologies to derive the controlling variables of the water use from a multi-date image stack. The controlling variables used are relevant land use, crops grown, crop calendar, growth stages, biometric parameters like leaf area index (LAI). High temporal resolution data like IRS WiFS was found useful to derive most of the crop dynamic features, while high spatial resolution data of IRS LISS III has been used to map the crop type and landuse. A correlationship analysis between satellite-measured radiances and field-collected LAI has been used to derive a spatial LAI distribution of rice crop. Other controlling variables such as canal discharge schedules, daily rainfall, wetland irrigated were spatially integrated in a geographic information system. A decision rule based classifier has been used to identify irrigation source and its spatial distribution in the study area. Results show that deviation between the mapped and field-collected water use locations is less than 5 percent. Further using all the information, daily water balance was computed to estimate actual irrigation water use and its productive in terms of LAI. It was found that irrigation productive of ground aquifer source is high.

Key word: Conjunctive water use, remote sensing, LAI, reflectance, irrigation mapping

Corresponding author: pkgupta@sac.isro.gov.in

Semi-supervised technique to retrieve irrigated crops from Landsat ETM+ imagery for small fields and mixed cropping systems of South Asia
MSDN Gamage, Mobin-ud-Din Ahmad and Hugh Turral
International Water Management Institute, Colombo, Sri Lanka

Abstract
Precise land cover maps and crop statistics are essential for monitoring and performance evaluation of any irrigation system. Remote sensing techniques offer efficient means of obtaining land cover information with high accuracy in near real time.  Many applications developed and referenced in the literature classify large tracts of homogenous cropped area, and are usually less effective in highly mixed cropping systems in South Asia (Canisius, forthcoming). This paper presents a semi-supervised technique to map cropped area in two major irrigation commands representing the Punjab Rice-Wheat, Punjab Sugarcane-Wheat and NWFP (North West Frontier Province) Maize-Wheat cropping systems in the Indus Basin irrigation system in Pakistan. Four Landsat ETM+ images were selected to the study: two Landsat ETM+ images (Path 149 Row 038) for September 2001 and March 2002 were selected to represent Kharif (2001) and Rabi (2001-02) seasons of Rechna Doab in Punjab. Rest of two images (Path 151 Row 36) of April 2002 and September 2002 were to represent Pehur High Level Canal (PHLC) area in NWFP.

A ground truth survey was conducted in different cropping areas that covered both irrigation areas. Accordingly geo-referenced GIS coverages of land use land cover were prepared for image classification. As first step, using field data and NDVI of those images, the vegetation was segregated form non vegetation area based on threshold values. Then based on Principle Component Analysis (PCA), Red, NIR and MIR bands were selected for supervised classification. GIS coverages were overlaid on a stacked image and 50% of the collected information was used to extract training signatures for all crops. The training samples were evaluated based on their location on two scatter plots between NIR:Red and NIR:MIR which yielded to distinct crop classes. The signatures of these distinct crop classes were used to perform supervised classification using maximum likelihood parametric rule. Finally, using the classified maps and other half of the field data, contingency matrixes were prepared to evaluate the classification accuracy in each season.

The results of Rechna Doab area show an overall accuracy of 87% in Kharif (Summer-Monsoon) compared to 83% in Rabi (Winter-Dry). Fodder signature mixed up with wheat causes to comparatively low overall accuracy in Rabi even though the lesser diversity of crops within the season. The results from the PHLC area give an overall accuracy of 78% in Kharif and 85% in Rabi season. Comparatively higher classification accuracy in Rechna Doab, especially in Kharif season, is due to more representative ground truth data, and large and homogenous area under paddy crop.

Keywords: Remote Sensing, GIS, Crop Classification, Land use-Land cover, Spectral Mixture, Pattern Recognition, Mixed Cropping System, Irrigation, Pakistan.
Corresponding author: a.mobin@cgiar.org

Assessment of availabilities and demands of water resources in river basins
Geerken, R. and Smith, R.
Department of Geology and Geophysics, Yale University, New Haven (CT) 06520, U.S.A.

Abstract
The regional modeling of hydrological characteristics of watersheds is currently limited to some basic input parameters that are globally available. Though, more variables may be locally measured at a regular basis, consistency in the input parameters is advisable where comparability of results is required. For data sparse regions we developed a simple water balance model that takes account of the limited availability of data with regional coverage. Input parameters are rainfall, temperature, and topography. Soil moistures are computed at 10-day intervals to show seasonal and spatial soil moisture/snow/run off variations over a time range from 1994 to 2004 (1981-2000). Through integrated analyses of modeled soil moistures with NDVI data from 8-km AVHRR (1981/1994-2000) and SPOT (1999-2004; radiometrically adjusted and spatially degraded) we identify locations and periods with crops under irrigation as well as inter-annual changes and trends in irrigation patterns. Model outputs further allow a characterization of the landscape with regard to their ecological potential or their vulnerability.

Using examples from the Middle East and from Central Asia we use these methods, e.g., to visualize the increasing irrigation water demand along the Euphrates River, the inter-annual response of irrigated agriculture to climate variation, or the impact of climate change on the seasonal availability of water resources.

To further improve the mostly qualitative model outputs, and add vegetation characteristics as an important component to the hydrologic modeling, we developed a Fourier based vegetation classification scheme of spatial and temporal compatibility. Class assignment is decided by a vegetation types’ phenological cycle but also measures its vegetation coverage. Classification results represent a highly consistent clustering of “phenologies” (NDVI-cycles) according to vegetation type, to land management (e.g. double/single crop), and to their coverage or, in multi-temporal analyses, quantifiable changes in these parameters. Multi-temporal analyses using classification results and hydrologic model outputs are used to analyze land cover/land use changes and inter-annual fluctuations/changes in water use efficiency.

Finally, initial steps are discussed towards a consistent parametrization of hydro-meteorological vegetation variables derived from the Fourier based classification, intended to enable the modeling of interactions and feed backs between vegetation and the hydrological cycle.

Corresponding author: roland.geerken@yale.edu

Abstract
The command areas are facing water efficiency problem because of upstream farmers siphoning out large volumes of water by adopting water intensive cropping pattern, creating land degradation problems and water scarcity at tail-end. This is the greatest hurdle in reaching possible irrigation potential of 3.5 m.Ha in Karnataka and 0.29 m.Ha in Hemavathy Command Area. The pursuit of maximizing productivity of land at the cost of optimizing productivity of scarce water resource has defeated the very objective of huge public investments on irrigation structures. Prior to independence, the local institutions had responsibility of maintaining and monitoring the resources. Now the same is managed by the Government which lacks manpower and resources, causing increased problems of maintaining and monitoring a wide network of irrigation systems. Hence, Government of Karnataka through state water policy has mooted the formation of Water Users Association (WUA) in the Command Areas to encourage the active participation of the end users and reviving the tradition of collective management during the year 2002. During this time, participatory action project was initiated through one of the identified informal WUA in Bagevalu village considering massive scarcity to share the water, managing and maintaining the structures in addition to the analysis of experiences of existing informal organizations.

Personal interviews, dialogues, village-stays, Participatory Rural Appraisal techniques, field visits and discussion with the concerned stakeholders of Hemavathy Command Area were proceeded to map the informal organisation and identify the research location for conducting participatory research. The analysis of eight informal organisations revealed that they were engaged in sharing canal irrigation water during the crisis without having their own organisational structure and resources. These organisations used only “discussion cum persuasion techniques” with fellow stakeholders to share water. Thus, they gave importance to social sustainability. “Manipulation by powerful elite farmers”, “biased bureaucratic approach”, ‘violation of recommended cropping pattern”, “ill maintenance of physical structures” and “wastage of water” were the problems faced by these organizations in sharing water. Then participatory action project started with mobilization of all stakeholders to become the members of this association by paying membership fee. After completing cent percent, farmers were facilitated for its registration. The 982 stakeholders of this association were spread along the five distributaries. The repair works of the physical structures were carried out jointly by the Irrigation Department, Command Area Development Authority and the Bagevalu WUA. “Proper and judicious use of available canal irrigation water in an equitable way among all stakeholders through empowerment of people” was the principle used. The capacity building and field level participatory extension activities were organised for the members intensively during the next two years on water use efficiency and recommended cropping pattern. The impact assessment made during the year 2004 revealed that “groups and networks”, “trust and solidarity”, “information and communication”, “collective action and cooperation”, “social cohesion and inclusion” and “empowerment and action” between the stakeholders have improved and they changed crop from paddy to light irrigated crops. Minimal efforts of social capital development have responded favourably to manage the water productively.

Corresponding author: nrganga@yahoo.co.in

Irrigated areas of Pakistan over last fifty years
Rakshan Roohi
Water Resource Research Institute, Pakistan

Abstract
Pakistan is inherently situated in a predominantly semi-arid to hyper-arid climatic zone. The Indus River System (IRS) is the main source of water for agricultural and non-agricultural water uses. It has three super dams besides 68 other large dams, 23 barrages/headworks/siphons, 12 inter-river link canals, 45 huge canal commands, more than 107,000 water courses and over 600,000 tube wells.

The sources of water are surface water (55%), groundwater (40%) and rainfall (5%). The catchment area of IRS spread over a vast tract of 500,000 sq. km out of which 56 percent lies within the country boundary. The catchment includes three major mountain ranges namely Himalaya, Karakoram and Hindukush (HKH) having variable snow and glaciated cover. Glaciers and snow melt contributes on an average 67.5% of the total IRS flows, ranging from 85% of Indus flows to little over 50% of Jhelum River basin. In an inventory, 5,218 glaciers have been identified having an area of 15,041 sq. km and estimated ice reserves of 2,738 km³. Considering an ice-water conversion factor of 1: 0.9, the water reserves in this ice mass are 1,997 MAF. The total surface water available at rim station in Indus Basin System is only 7.44% of this huge reserve. In the current scenario of climate change the glaciers are melting at a faster rate resulting in the formation of glacial lakes or increase in the size of existing lakes.

The average annual flow of IRS is approximately 144 MAF of which presently 105 MAF is being diverted for irrigation while a major portion of the balance outflows into the sea. At present, irrigation uses about 93% of the water and support agriculture which contributes about 25% of GNP, more than 60% of foreign exchange earnings and provides 90% of food and fiber requirements. About 68% of the rural population depends on the sector and 46% of the labor force is employed in it. According to the statistics (2003-04), the irrigated agriculture was practiced over an area of 18.78 mha. The area under irrigated agriculture estimated by analysis of NOAA image of October 14^th, 1992 is 17.132 mha compared to the statistics estimates of 1991-92 as 16.85 mha. The difference between two estimates needs to be further analyzed. It can be due to SRS data resolution and masking the total cropped area with rainfed zone or deficiencies in the estimate procedures. Currently using Landsat TM and ETM+ data the estimation of extent of various Landcover/landuse classes is in progress.

The population of the Pakistan was 65.3 million in1972 which has increased to 130.6 million in 1998. The per person water availability in the country has been going down and would reduce further from 1200 cubic meters per person to 1000 cubic meters per person by 2010. Furthermore, waterlogging and salinity pose a major threat to the sustainability of irrigated agriculture in about 30 percent of irrigated lands, which is directly related to the low efficiency of irrigation systems. The other threats to irrigated agriculture include uneven distribution of water, low storage capacity, excessive groundwater extraction and weak institutional linkages. Considering all these factors the need of the day is to have a "Green" as well as “Blue” revolution simultaneously for a prosperous economic growth of the country.

Corresponding author: drroohi@hotmail.com

Satellite remote sensing and GIS based irrigated area mapping in Godavari River Basin of India
Jayashree Shivaji Pachpute and Dilip Dyandeo Pawar
Mahatma Phule Krishi Vidyapeeth, Rahuri-, India

Abstract
The Mula irrigation project is a major irrigation scheme situated in Godavari river basin of India. The overall efficiency of this project is only 38 per cent under surface irrigation systems. This low efficiency is due to conveyance and seepage losses through canal distributaries, minors and field channels as well as adoption of unscientific water management practices such as over irrigation, absence of suitable layouts for crops and lack of improved package of practices such as timely sowing, high yielding cultivars and balanced fertilizer use in relation to water availability. Due to lower efficiencies in water conveyance, storage and distribution the area actually irrigated through canals is less than the area recorded by irrigation department. The tail end farmers do not get adequate or remain deprived of irrigation water. Therefore remote sensing and GIS technology was used to assess the actual area irrigated with its spatial distribution in head and tail commands. The Indian Remote Sensing Data of RESOURCESAT P6 having resolution of 23.5 m was used for irrigated area assessment in the command. The inventory of irrigated area was rapidly prepared by using multi-spectral satellite images of IRS LISS III sensor for different overpass dates. The dates of overpass were falling in rabi season during which the moisture content of irrigated land in the study area was quite higher than the surrounding non-irrigated land. Due to this difference in moisture content, there was a large contrast between the spectral reflectance from the irrigated crops and from the surrounding areas. By simple unsupervised image classification technique, the irrigated areas were identified and their statistics were determined. In rabi 2004-05, the irrigated command area (ICA) of right bank canal of Mula irrigation project was 72832 ha indicating irrigation intensity of 70.2 per cent. The irrigated area as per the official records was 80810 ha. So a reasonable difference of 7978 ha was found in recorded and actually irrigated area. The accuracy of irrigated area assessment was more than 85 per cent. The NDVI profile generated from multi-spectral imageries of different overpass dates indicated inadequacy of irrigation at the tail end of most of the distributories in the command.

Corresponding author: jaishree_kumkar@rediffmail.com

Multi-angle spectral measurements for classifying cropping areas
Francis Canicius
University of West Indies

Abstract
Vegetation covers on the earth surface have different structures depending on their cover type. Forests may be the most highly organized structures with shoots, branches, tree crowns and tree groups, while grasses and crops are lacking such similar structures. Though the crop lands sometimes have similar greening process as herbaceous vegetation, they are homogenous and the distribution of crops is spatially arranged and therefore not random.

Multi-angle satellite observations provide means to characterize the anisotropy of surface reflectance, which has been shown to contain information on the structure of vegetated surfaces.  The total reflectance measured by a single observation at nadir is largely determined by the sunlit components of foliage and it overlaps between the natural vegetation and crop lands leads to misclassification. Therefore, additional information, preferably angular as it characterizes vegetation structure, is needed for the classification. Utilizing spectral measurements over a range of observation angles, however, allow discriminating cropping areas from the pattern of variation.

The Multi-angle Imaging SpectroRadiometer (MISR) instrument on board the Terra platform offers the capability of acquiring reflectance data on any earth target in four spectral bands, from nine different directions, at a multiple spatial resolution of 250m and 1000m. One dimension fully exploits the spectral information in blue, red and near infrared bands while the other dimension capitalizes on multi-angle capability of MISR to assess the anisotropic behavior of terrestrial surfaces with respect to solar radiation.

This investigation uses MISR data to estimate land cover characteristics and classification of cropping area particularly irrigated winter wheat in northern part of India. The analyses indicate that both spectral and angular variables are significantly different for different land cover types and they all convey the information valuable for identifying land covers. In addition, the incorporation of both multi-angle and multi-spectral 250m resolution data substantially improves the classification accuracy. The emphasis of the work also assesses to what extent multi-angle data can be used to reduce the difficulties in land cover classification mainly in cropping area classification process.  These new findings provide evidence to the improved capabilities for the applications of MISR data. The resolution of MISR is expected to be sufficient to classify the cropping area. Further testing has to be done for detailed discrimination of cropping systems and crop types for better understanding.

Corresponding author: franciscanicius@yahoo.com

Mapping the irrigation area for estimation of agricultural water demand in North China Plain using MODIS remote sensing data.
Zhihao Qin^1,2, Maofang Gao¹ and Wilko Schweers^1
1 Institute of Agro-Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Zhongguancun South Street 12, Beijing 100081, China.
² International Institute for Earth System Sciences, Nanjing University, Hankou Road 22, Nanjing 210093, China.

Abstract
The North China Plain is one of the most important agricultural regions in China with severe challenges of water shortage. Agriculture in the plain is very intensive. Farming in the region is a typical irrigation-supported system of winter wheat, followed by summer maize. Other important crops include cotton, potatoes, soybean and various vegetables. Cultivation of these crops requires large amounts of irrigation water to support the harvest. However, water resources are very limited due to high evaporation and unbalanced precipitation. Although the average annual precipitation in the region varies between 500-800mm, water shortage has been a common social-economic problem, resulting from a rapid increase of economic development and intensive farming. Both surface and underground water resources in the region have been over-extracted to meet an ever-increasing demand, leading to omens of water scarcity disasters in the region. For example, each year the Yellow River, the 2nd largest river in China and the largest river in the region, had no water to enter the Bohai Sea for several months. The ground water level has fallen at a rate of abou