You are on page 1of 18

Module 8




Spatial Data
Point Data
Linear Data
8.3.1 The River
8.3.2 The drainage network
8.3.3 Stream Profile
Area Data
8.4.1 River Cross-section
Volumetric Data
Topographic Maps
8.6.1 Soil and Land Use Maps
Emerging Technologies for Spatial Data Acquisition and Management
Remote Sensing
8.8.1 Physical Properties of Electromagnetic Energy
8.8.2 Interaction of Electromagnetic Energy with Objects
8.8.3 Basic Components of a Remote Sensing System
8.8.4 Reflectance Characteristics of Different Objects

8.8.5 Remote Sensing Platforms

8.8.6 Digital Image Processing

8.8.7 Indian Remote Sensing Satellite Program

Geographical Information Systems
8.9.1 Advantages of GIS
8.9.2 Spatial data representations
8.9.3 Map projection
8.9.4 GIS Operations
8.9.5 Spatial Data Analysis
8.9.6 Applications
8.10 Closure
Keywords: Spatial Data, Point, Linear, Topography, Remote Sensing, Geographical Information

Spatial Data
Spatial data are those data that have a spatial or space dimension, for example, command area of
a project. A wide variety of spatial data are used in water resources development and
management. Maps are the most effective means of visualizing the spatial data. In water
resources, both time-varying and time-invariant data are used. The spatial data that do not vary
with time (at least on time scales of interest here) include the catchment/command boundary,
topography, soil map, stream network, and geological features. Some spatial data change with
respect to time and such data include land use, cropping pattern, etc. Many times, variation of a
hydrological variable is displayed through a map, e.g., isohyetal map.

Spatial data pertaining to a river such as channel cross sections, longitudinal profiles and
bed characteristics are needed for many hydrological studies. These data slowly vary with time
and they can, therefore, be considered as semi-static. In alluvial areas, channel cross section and
bed characteristics may significantly change after a major flood event. River cross section area
and its properties have important bearing in stage-discharge relation. It is important that these
data are collected at the desired locations with appropriate frequency. Conventionally, such data
are stored in the form of paper maps and analyzed using manual means.
Large infrastructure is needed to capture, manage, and analyze spatial data. Spatial data
infrastructure consists of technology, hardware and software, policies and procedure/rules, and
people that are necessary to acquire, store and manage, analyze, and share the spatial data among
the users. It provides a structure of practices and relationships among data producers and users
that facilitates data sharing and use. A rigorous national data framework facilitates information
exchange and significantly reduces duplication of effort.
In GIS, maps are called spatial data. Information shown on paper maps can be input is
GIS as spatial data. Example of spatial data are stream network, well locations, villages, cities,
topographic contours, spot elevations, roads, land use, soil, geology, hydrological investigation
locations, hydrological response units etc. Spatial data are classified in to three types, namely
area, line and point. Areas are spatial data that are represented as closed figures, e.g. forests,
lakes, Thiessen polygons etc. Lines are spatial data that are represented as curves e.g. forest
boundaries, lake boundaries, contours, stream network, roads etc. Point spatial data are
represented as points on maps e.g. well locations, rain gauge stations, villages, etc. The data is
scale dependent in some instances, e.g., on small-scale maps a city will be represented as point
data, whereas on large-scale map, it will be represented as area data. A closed line data can be
converted easily to area data in GIS. For example forest boundary data can be converted to forest
land-use map. Point data cannot be converted in to area data. But reverse is true, i.e. area data
can be converted as point data.
Point Data
In the context of spatial data, a point is a unique location on a line, in an area, or in a volume. It
has no spatial extent. A point may represent the location of a rain gauge, the location of a well,
the outlet of a basin. The attribute of a point may be simple or complex. A simple attribute may
be, for example, its coordinates which uniquely identify its location in the three dimensional
space. A complex attribute may be the geological profile beneath the point.
The physiographic description of a point covers its geometric properties (form, relief,
slope, etc.) and its permanent physical properties (permeability, nature of rocks, soil structure,
land-use type, etc.). The former are limited to the local slope, while the latter comprise a whole
range of possible physical properties, expressed in scalar form for a point on a horizontal surface

or in vectorial form for a profile, for example, geological core.

Linear Data
A linear feature can be represented by a line on a map, for example a road or a canal. In
hydrology, three types of linear elements are common:
a) Boundaries;
b) Isopleths of a permanent feature, for example, contours;
c) Thalwegs.
The first two types are linked to areal aspects, which will be examined later. The thalweg is itself
to be considered not only as represented in horizontal projection and longitudinal profile, but
also by the way in which it combines with other thalwegs to form a drainage network, which has
its own physiographic characteristics. Some drainage network characteristics are linear, for
example, the bifurcation ratio, while others are areal in nature, such as the drainage density.
8.3.1 The River
Depending on the scale of the map, a river may be represented by a single curve or by two curves
representing its banks. The river is represented by two lines when either the map has small scale
or the river is very wide or both. Rivers are frequently shown by their thalweg which is a line
connecting the lowest points along the stream channel. Sometimes the line midway the two
banks is taken as the thalweg. In many cases, the banks and the lowest points are not always very
distinct and the map scale does not permit these to be marked accurately.
8.3.2 The Drainage Network
Drainage network, as the name implies, is formed by orderly joining the streams in a basin. In a
network representation, the size of the streams is not important. In hydrological studies, several
systems have been proposed to classify the streams, for example, the Horton, the Stahler, and the
Shreve system. In the Hortons scheme which is frequently followed, any elementary stream is
said to be of order 1, any stream with a tributary of order 1 is said to be of order 2, and any
stream with a tributary of order x is said to be of order x + 1. While mapping the streams, usually
the information is picked up from the Survey of India maps at 1:5000 scale. The definition of the
smallest streams is often subjective.
The first-order streams are the channels without any tributary; they receive all the flow
from surface overland route. Two first order channels join to form a second-order channel. A
second-order channel receives flow from the first-order channels which form it and the overland
flow. Evidently, a second-order channel carries more flow than a first-order channel. A thirdorder channel is formed by the junction of two second-order channels; it receives flow from the
two second-order channels that form it, from direct overland flow, and from first- or secondorder channel(s) that might join it. Thus, a stream of any order has two or more tributaries of the

next lower order and so on. This scheme of ordering of streams is referred to as Horton-Strahler
ordering scheme.


8.3.3 Stream Profile

The stream profile is the variation in elevation of the points of the stream thalweg as a function
of their distance from the origin, which is generally taken as the confluence of the stream with a
larger stream or as its mouth. On such a profile, a certain number of topographical features may
be present such as rapids, waterfalls and changes of slope that frequently mark the boundary
between two reaches with different geologic controls.
The average slope of a stream is the difference in elevation between its highest point and
its confluence or mouth divided by its total length. Slopes of the various segments of the river are
required for hydraulic models, flood routing hydropower and morphological studies.
Figure 8.1 shows examples of stream profiles of a Himalyan river and its tributary. Such
a diagram gives an over view of the variation in slope in the drainage network and is very useful
in planning studies.


Fig. 8.1 Stream profiles of a typical Himalyan river and its tributary.

Area Data
The basin or catchment or watershed area at a point is defined as the area that receives
precipitation and other inputs and, after hydrological processes contributes to runoff at that point.
The watershed boundary or the ridge, directs any precipitation falling within its area towards the
outlet, whereas any precipitation falling outside the boundary drains to a different outlet. The
watershed is usually defined by using contour maps small scale.
The basin perimeter is measured in a GIS or with a curvometer. On a paper map, it is
determined by planimeter. The measured perimeter is a function of the scale and accuracy of the
maps or photographs, the quality of the curvometer, and the care taken in its use Many GIS have
routines to automatically delineate catchment areas by using digital elevation models.
A basins physical characteristics are the soil types, the land use and land cover (for
example, crops lakes, swamps, or glaciers), and the type of land use (for example, rural or
urbanized areas, lakes, or swamps). These physical features may be compiled as layers within a
8.4.1 River Cross-section
The cross-section of a river at a given location shows the variation of its bed elevation across the
width of the river or in a direction perpendicular to the flow. Cross-section area data are input in
a variety of studies, such as river morphology, rating curve analysis, flood routing, and sediment
transport. Together with flow velocity, cross-section area is used to compute river discharge.
In India, cross-sections are typically measured at stream gauging sites, hydraulic
structures such as dam or barrages, and at bridges. For hydrologic studies (particularly flood
routing and forecasting), cross section data are needed at closely spaced locations and places
where the stream profile changes significantly. If a channel is subjected to high rates of erosion
/deposition, frequent cross-section measurements are required.
Volumetric Data
Volumetric data primarily pertain to water storage and in less number of cases to sediment
storage and groundwater storage. Surface storage related data either pertain to the storage
capacity of existing or proposed reservoirs or the volume of water stored in a reservoir at a given
time. Bathymetric methods are used to determine storage of an existing lake or reservoir.
Hydrographic surveys are carried out to determine the volume of sediment deposited in a
Ordinary maps do not give bathymetric data on lakes and reservoirs. Bathymetric survey
is carried out by using a boat and sounding by using echo sounders or tapes. The depths are
referenced to the national or an arbitrary datum. Volume is computed by employing the
trapezoidal, prismoidal, or any other suitable formula.

Topographic Maps
A topographic map is a detailed representation of the objects present on the surface of the earth
at a particular scale. Conventionally, topographic maps show the land features such as roads,
railway lines, power and other utilities, rivers and lakes, habited areas, etc. Elevations and
changes in it are shown by means of contour lines. Contours are the curves that connect points of
the same elevations.
Topographic maps provide the information about a terrain and thus are very useful for
water resources studies. The Survey of India has carried out extensive survey for almost the
whole India and prepared topographic maps at various scales. To identify a map of a particular
area, a map numbering system has been adopted by Survey of India. The system of identification
is as described here.
For the purpose of nomenclature or reference of the maps, an International Map Series (in
the Latitude range 4 N to 40 N and Longitude range 44 E to 124 E) at the scale of
1:1,000,000 is considered as the base. This base is divided into grids of 4 x 4 and these are
numbered from 1 (at the extreme north-west) to 136. Only land areas are covered and if any 4
square falls completely in the sea, the area is not covered.
To refer to any Indian Topographic map, each 4 x 4 region is further divided into 16
grids, each covering 1 latitude x 1 longitude. The 1 x 1 grid in the North-West corner is
assigned the letter A and the grids are assigned consecutive letters, column-wise. Thus the letter
for the last grid on the South-East corner will be P. Due to geographic coverage, these maps are
also called degree sheets and are on a scale 1:250,000. Contours are drawn in these maps at
intervals of 100m. To refer to a particular map, the number of the base map grid and the letter
code is used, for example 39N (Fig. 8.2). Approximate area covered by each such map is 11140
sq. km. Maps in this scale are highly useful for planning studies from different sectors, viz.,
water, transportation, administrative.

Fig. 8.2 Nomenclatu

ure scheme of
o 1:250,000 topographicc maps.
T 1:250,000 scale maps or degree sheets
are fuurther sub-divvided in twoo ways:
1) Each
sheet iss divided intto four partss (2 rows byy 2 columnss), each of 330' latitude x 30'
ongitude (1:100,000) deesignated by
y cardinal ddirections N
W, and SE. Such
heets are ideentified, for example
as 39
3 P/NE (Figg. 8.3).

menclature sccheme of 1:1

100,000 topoographic maaps.
Fig. 8.3 Nom

sheets are also diivided into 16
1 sheets (4 rows by 4 ccolumns), eaach 15' latituude x
2) Degree
15' longitudee (scale 1:50
0,000) and numbered
from 1 (at tthe north-weest corner oof the
particular deg
gree sheet) to
t 16 column
nwise. Thesee are identiffied, for exam
mple, as 39 B/14
Fig. 8.4).
area covereed by each map
m is 700 ssq. km. Theese are generral purpose maps
nd are used by adminisstrators, plan
nners, and eengineers. Inn fact, it is tthe most poopular
m series for all activitiees of the Gov

menclature sccheme of 1:5

50,000 topoggraphic maps.
Fig. 8.4 Nom
00 scale sheeet contains four (2 row
ws by 2 coluumns) 1:25,0000 sheet (77' 1/2
latitude x 7' 1/2 longitude) which
h are numberred NW, NE
E, SW, and S
SE. Such sheeets are idenntified
as 53 O/1
14/NE. Apprroximate areea covered by
y each such map is 175 sq. km. Thiss is more dettailed
map and is used for planning
d developmeent purposes..
topograaphic maps are fairly old these maay have to bbe updated aas required for a
particularr project. Fo
or large scalee maps, furth
her surveyingg needs to be carried outt.
A recent deveelopment pro
oviding topo
ographic datta is the digittal elevationn model prodduced
at 90 m resolution by
b the Shutttle Radar To
opography M
Mission (SR
M consistedd of a
y built radar system that was sent to
o space onbooard the Spaace Shuttle E
Endeavour dduring
an 11-day mission in
n February of
o 2000. SRT
TM obtainedd elevation ddata on a neaar-global scaale to

generate the most complete high-resolution digital topographic database of Earth

( The level 2 digital elevation model, currently available only
for the United States, has a horizontal accuracy of 30 m and vertical accuracy of 18 m.
CARTOSAT 1 is the first Indian Remote Sensing Satellite capable of providing in-orbit
stereo images. The images can be used for cartographic applications. Cameras of this satellite
have a resolution of 2.5m ( The
Cartosat 1 provides stereo pairs required for generating Digital Elevation Models, Ortho Image
products, and Value added products for various applications of Geographical Information System
8.6.1 Soil and Land Use Maps
Maps of soil types and hydrologic properties and land-use/land cover are another important
spatial data for hydrologic analysis and design.
An institute under the Indian Council of Agricultural Research (ICAR), the National
Bureau of Soil Survey and Land Use Planning (NBSS&LUP) situated at Nagpur is the premier
institute for soil related activities such as classification, mapping, uniform nomenclature, and
interpretation. Publications of NBSS&LUP include State soil maps (1:250,000 scale), District
soil resource maps (1:50,000 scale), and District land use atlas. More details can be obtained at
Maps of land use are prepared and maintained by Agriculture Department and revenue
authorities. These maps are extensively used in planning activities as well as in many other
sectors including water.
Emerging Technologies for Spatial Data Acquisition and Management
The acquisition and management of spatial data using the conventional means (paper maps) is
difficult, tedious and time consuming. It suffers from the following drawbacks:

Even relatively simple tasks such as making a copy of a big size coloured map are
difficult. Distortions may be introduced in photocopying a large map.

When the number of maps increases, storage of maps becomes difficult and requires
large infrastructure.

Hardcopy maps tend to deteriorate with time.

Large resources are needed to update the paper maps at desired frequency.

Analysis of spatial data in hardcopy map form is difficult.

We will now briefly describe two proven technologies that are being increasingly used to
collect and manage spatial data. These are geographical information systems and remote sensing.

Geographical Information Systems
A Geographic Information System (GIS) is a computer-based system that is used in input,
storage, analysis manipulation, retrieval, and output, of spatial data. These systems consist of
computer hardware and software. GIS are increasingly being used in applications in natural
resources management. These days GIS are also being integrated with remote sensing and GPS.
GIS has origin in manual overlay operations done as early as in 1920s. In 1958, computer
based cartography initiated in USA which culminated in development of first general purpose
mapping software in 1960s. Canada GIS is also cited as first GIS and was developed around the
same time. Present day popular commercial/ open source GIS are: ArcGIS, GRASS, MapInfo,
ERDAS, IDRISI, ILWIS etc. GRASS (Geographic Resource Analysis Support System) is highend open source software. Arc GIS is modular high-end commercial software. ERDAS and
ILWIS packages have image processing and GIS capabilities.
8.9.1 Advantages of GIS
After a map is available in digital form, it is easy to update or modify it.
Conventionally, overlays are manually prepared. This process is cumbersome and error
prone. GIS allows digital overlaying a number of maps.
Conventionally, maps are browsed to retrieve information. In GIS retrieval of
information is very fast, and much easier.
In conventional method, hardcopy output is prepared. Updating of such maps is difficult
and the map is to be redrawn. This process is simple in GIS- the map can be easily edited
and printed.
Annotation is clumsy in hardcopy maps. Thus, while retrieving information ambiguity
may arise. Also all features may not be annotated in paper maps. Thus, attribute
information, e.g. names for some of the features are lost in paper products. In GIS,
information is stored in tables and is linked to geographic features and thus is not limited
by availability of annotation space/ color/ symbol, etc.
If multiple maps are prepared for same area, e.g. watershed, land use, geomorphology,
common boundaries are drawn manually and may not match in different maps. In GIS
common boundaries are digitized once and are available to all layers. Once GIS map
layers are prepared, any number of maps can be designed.
Storing of a large number of paper maps is difficult; maps deteriorate with time. Digital
maps can be stored in a much compact way.
GIS systems are now available for standard computers in practical, low-cost formats. The
main cost factor now resides in database compilation, and training and updating of technical
staff. Data capture or input in GIS is costly but it is one-time affair. Commercially available
paper maps may be cheaper than GIS layer. Use of GIS requires investment in computer
hardware, software, and training. GIS handling requires trained manpower. GIS software should

have proper functionality as desired in an application. For example, for hydrological modeling,
DEM analysis should be available in the package.
8.9.2 Spatial data representations
Spatial data are represented mainly in two ways in GIS: raster and vector. These data
representations can be transformed from one form to the other, albeit with some information loss.
In raster, spatial data are structured as grid of cells or pixels. Their row and column numbers
addresses the cells. In many distributed hydrological models, spatial data and hydrological
computations are done in this form. This is a native representation for remotely sensed data.
Satellite data are captured/ resampled as pixels (picture elements) and information is extracted
through digital image processing. In vector model, spatial data are represented as coordinate
points. For example, point data is represented as a pair of coordinates. A straight-line is
represented as two pair of coordinates, representing end points of the line. A curved line is
represented as finite line segments. Area data are represented as line data with some additional
information e.g. centroid, adjacent areas, etc.
In raster data, points and lines are represented with finite area and finite width and thus is
not a natural representation. Lines have stepped or zagged appearance. In vector model, points
and lines have infinitesimal area and width respectively. Lines are smooth curves. Raster data
require large storage space. Vector data require small storage space. Thematic maps prepared
from remotely sensed data are available is raster form and are often processed as such. Many
hydrological models use both the representations. For example, thematic maps of catchment
variables and hydrometeorolgical measurements are prepared in raster form. Stream network is
processed in vector form etc. In raster form, value of many catchment variables is scale
dependent. For example, average slope of catchment reduces with increase in raster grid size. In
most GIS, the representations coexist. For example, it is better to capture spatial data from
conventional thematic maps, through visual interpretation of remotely sensed data in vector
form. Thematic maps from digital processing of satellite data may be obtained in raster form.
Topology: Method of representing vector data is called its topology. A line consists of two nodes
and one or more vertices. Nodes are end points of the line. Lines also have directions. Thus,
nodes are referred as from node and to node depending on direction of the line. Areas are
represented by left area and right area of each line.
Digital Elevation Model (DEM)
Topographic elevation data in GIS are called DEM. These are represented in GIS in various
manners namely contours, raster, and TIN (Triangulated Irregular Network). Contours are
conventional representations of DEM and are used in topographic maps. Contours are equal
elevation lines. Normally, equal interval contours are drawn in topographic maps to represent
topography. For example Survey of India (SOI) maps at 1:250,000, 1:50,000 and 1:25,000 scale

have contours at 100, 20 and 10 m elevation interval. DEMs are used to derive topographic
information such as slope, aspect and are also used in hydrological calculations, e.g. stream
network delineation, topographic index, delineation of catchment area, etc.
In TIN model, elevations at the vertices of triangles are used to compute elevation at interior
points of the triangles. Using elevation of the vertices of a triangle, a planner or higher order
surface can be fitted. The surface can be used to derive elevation at points inside the triangle.
TIN model requires Delaunay triangulation. In this, constituent triangles are as equilateral as
possible. Circum-circles of the triangles include no other point of the triangulation. Triangulation
is performed first by constructing Voronoi diagram (Thiessen polygons). Points included in
adjacent polygons are joined to create Delaunay triangulation. Voronoi diagram is drawn using
proximity analysis.
Interpolation is a technique of determining unknown value of a variable at location from known
values at other locations. Interpolation can be used for any spatial variable, e.g. topographic
elevation, pH, SAR, pollutant concentration, groundwater depth and level, population etc.
Known values can be at point, line of area locations. Point data can be spot heights, pH, pollutant
concentration etc. Line data can be topographic contours, etc. Area data can be population
density in regions, etc.
Thiessen polygons or nearest neighbor
This is popular method of interpolating rainfall values from point rainfall measurement at rain
gauges. Generally, point rainfall stations are limited in number. To determine basin wide average
rainfall, the method of interpolation is used. The Thiessen polygon diagram is prepared by
proximity analysis. For measuring average rainfall in the catchment, weights for each rain gauge
station are area of Thiessen polygon surrounding that station divided by the total catchment area.
Distance weighted averaging
In this averaging, a weight of inverse of distance function is used. Distance function is nth power
of distance. Thus, more weight is assigned to stations closer to the interpolation location.
Surface fitting: Here, n-degree polynomial surface is fit between selected known values. The
surface can be used, among the other application, to interpolate values.
Krigging: Variation of spatial variables can be partitioned in three components, namely drift or
structure, small variations and random noise. First component depicts general trend of the data.
Second component represents small variations from the general trend. These variations are
random but spatially autocorrelated. Third component depicts random values that are not

spatially autocorrelated. Kriging is best suited for interpolation of pollutant concentration,

geological and mining variables, e.g. grade of ores, etc. For these data, single smooth
mathematical equations are not suitable. The technique is based on assumption that values in
neighbourhood have generally higher correlated. Apart from the estimate of values, error
estimates are also provided in kriging technique. In presence of large random noise in data, good
semivariogram is not obtained and this results in deterioration in interpolation quality.
Georeferencing: Earth is a three dimensional surface. In maps, this three-dimensional surface is
transformed in to flat surface. For the transformation, map projections are employed. Locations
on the map are drawn using Cartesian coordinates obtained through map projections. Geographic
graticules (latitudes and longitudes) are later drawn in maps. Sometimes, Cartesian coordinate
grids are also drawn on maps. In georeferencing earth coordinates are assigned to spatial data.
Either lat/long or Cartesian coordinates can be used in georeferencing maps in GIS. Cartesian
coordinates allow measurements, e.g. area and lengths and are thus frequently used. Geographic
coordinates do not allow measurements. A map can also be referenced without using a map
projection. In such case, it is difficult to integrate GIS layers obtained from different sources.
A map which is to be geo-referenced is called the source map and the reference map is
the map which has known coordinates. Points whose reference coordinates are known and which
are clearly identifiable on both the source and reference maps are known as control points. For
coordinates of control points in two maps, coefficients of a polynomial transformation equation
are estimated.
8.9.3 Map Projection
Map projection is transfer of positions on earth to corresponding points on a flat sheet of paper.
Because of the shape of the earth, this transformation involves approximations and is not
distortion free. Distortions occur in lengths, angles, shapes and areas. Scale is a ratio of length on
map to its counterpart on the earth. Since large size of features on earth surface, a scale is needed
to draw these features on a small sheet of paper. Earth shape is assumed spherical or spheroidal.
An intermediate plotting surfaces namely cylinder, cone or plane is used in projections.
Corresponding projections are called cylindrical, conical and azimuthal respectively.
Distortions occur in projecting earth surface on to intermediate plotting surface. Ideally,
areas distances, directions, angles and shapes should be preserved. In reality, few of these
properties are preserved. Based on application, choice is made as to which propertied are to be
preserved and appropriate map projection is selected. In areal distortions area of a figure may
increase or reduce. In linear distortion length and its curvature may change. In angular distortion
an angle may increase or decrease. In shape distortion, a square may become parallelogram,
rectangle or may have curved boundaries or both. A point may be distorted in to a line. In equal
area projections, area of a figure is preserved. In the process distortions are introduced in

distances and angles or shape of figures. In conformal projection, shape is preserved. In this
process, areas figures are distorted. Projections with these contrasting properties are called equal
area and conformal projections respectively. With different orientation of intermediate plotting
surface, it is possible to obtain different projections. Azimuthal projections are called polar,
equatorial or oblique depending on point of contact of plane falling on poles, equator or at
intermediate latitude. For conic and cylindrical projection based on orientation of axis of the
plotting surfaces, the projections are classified. When the axis coincides with earths polar axis,
perpendicular to it and lying in equatorial plane and is oblique to it, the projections are called
regular or equatorial, transverse and oblique respectively. The plotting surfaces can also be
tangential or secant to the earth surface. Normally for projection, mathematical approach is used.
8.9.4 GIS Operations
Input of data in a GIS database is either by digitization or import. Digitization can be done onscreen or by a digitizer to create/ edit GIS objects in vector format. Input data are, sometimes,
available in GIS image formats. These data are converted to native format of GIS.
Storage: Geographic data are stored in GIS the native format of GIS. For one spatial data, many
computer files are created which contain different information. Attribute data are stored in Data
Base Management System and are linked to geographic objects. Storage of data in the form of
layers looks very attractive from water resources data. Different types of data, such as soil and
land cover, are stored in different map layers. GIS permits analysis of single or multiple layers
and various layers can be overlaid, one on top of another. From a water resources point of view,
spatial variation of data is important, e.g., the variation of soil hydraulic properties.
8.9.5 Spatial Data Analysis
Data analysis involves operations with geographic data and their attributes to obtain derived
information, generate query, statistics etc. broad categories and operations therein are as follows.
Statistics: for example, count, length, area, perimeter, shape, centroid, etc. for geographic objects
can be derived in GIS. For continuous surfaces, average, standard deviation, maximum,
minimum, etc. are derived. Summary operation produces zonal statistics for a map. For example,
land use statistics for watershed in a basin can be generated.
Mathematical operations: Mathematical operations, e.g., addition, subtraction, multiplication,
division, exponential, logarithm, absolute, truncation, round off, negative, trigonometric
operations can be performed in GIS. For example various component maps in the universal soil
loss equation, namely R, K, L, S, C and P can be prepared as different largess and multiplied
using multiplication operation.
Logical operations: Logical operations, namely or, and, not, xor can be performed on maps.

Figure 8..8 shows log

gical operatiions. For exaample, landuuse= agricullture and pH
H >= 8 will rresult
in salt aff
ffected agricu
ulture area.
nal: If-then--else conditio
onal operatio
onal can be performed oon maps. Foor example, if 50
< return period
00 and land use=
ntial, then vuulnerability=
= high else vuulnerability=
n gives flood
d vulnerability map.

Figure 8.8 Lo
ogical operaations commonly used inn GIS.
nal: In this operation, all combinaations of claasses in tw
wo maps aree obtained inn the
resulted map.
m For ex
xample, overrlay of soil hydrological
soil group aand land usee/ cover mapp will
provide soil-cover
omplex map..
nverts valuees into interv
val. A contiinuous surfaace is input and area m
map is
Classification: It con
or the operattion. In the output area map, isolinees, i.e., line of equal vaalues, enclosse the
output fo
area. Exaamples of vaarious isolin
nes are conto
ours, isohyette, isotherm,, isobar etc.,, which reprresent
topographic elevatiion, rainfalll, temperatture, pressuure, respecctively. In reclassificaation;
informatiion of geogrraphic objectt is changed.. For examplle soil seriess map may bbe changed too soil
pH map.
The operation iss similar to distance, exxcept that at a specified distance ann area
Search/ buffer:
hic object is created.
Neighborrhood: Inforrmation in eight
bor, their loccations and statistics, ee.g. mean, m
median, minimum,
standard deviaation, coeffi cient of variiation, etc. arre extracted..

Aggregate: Cell size of raster maps can be changed in fractions of half, one fourth etc. using
functions e.g. mean, predominant, minimum and maximum.
Query: Query is done by attributes or geometry. In query by attribute, a logical expression is
written in attributes and result is obtained. For example land use=agriculture will select/ display
agriculture areas. In query by geometry, objects are selected on screen to view their attributes.
After the requisite data are stored in a GIS database, it is easy to answer complex queries like
what areas in the catchment have forest on shallow soil with 3% slope?
Output: GIS output may be obtained as paper maps or as a picture which may be shared or
printed or inserted in a report or presentation. The output maps may contain various cartographic
elements namely title, legend, graticules or grids, north arrow, scale, annotations, notes, etc. In
one output more than one GIS layer may be included apart from cartographic elements. When
design is saved, it only contains reference to the layers. Thus, if a layer is modified and designed
output map is opened at a later time, the changes are reflected in the output.
The topography of a river basin may be represented in two different ways: as a digital
elevation model or as a triangulated irregular network (TIN). The digital elevation model is a
grid of elevation values that has regular spacing while TIN is a series of points linked into
triangular surfaces that approximate the surface. The spacing of points in TIN are non-uniform,
which allows points to be located on critical terrain features, roads or river banks. The accuracy
of such digital terrain models depends on the source of the data, the point density and
distribution, and other related data used in their development. Conventional contour maps may
be prepared from a digital elevation model or TIN. Orthophotos are images of the landscape
from which features can be referenced to one another. They are digital images produced by
processing aerial photography to geodetic control elevation data to remove all sources of
distortion. The image has the properties of scale and accuracy associated with a map. Such
images can be derived using airborne or satellite sensors.
8.9.6 Applications
Several hydrological software now have a GIS interface. GIS interfaces have been developed for
hydrological models such as HEC models, SWAT, etc. ArcInfo extensions such as Spatial
Analyst, 3D analyst, are useful for hydrological tasks. Script languages are also available to write
interfaces in GIS, for example, Arc Macro Language is the script language in ArcInfo.
GIS techniques facilitate input of spatial data to hydrological model. GIS are being
incorporated in hydrological models to extract and format distributed watershed data. Use of
DEM permits complete physiographic and hydrological depiction of basins. The efficiency of
handling large volumes of data means that more comprehensive and detailed maps, isolines and

themes can be prepared. This is a significant improvement in water resources management as

map preparation is often time-consuming and expensive.
GIS are becoming very common in the field of water resources assessment and
management. Many tasks of data collection, compilation, and interpretation can be facilitated by
means of GIS. In network planning and design, the ability to map quickly and display surface
water and related stations enables a more effective integration to take place. Network maps,
showing basins or stations selected according to record quality, watershed, or operational
characteristics, can be used for both short-term and long-term planning.
Groundwater potential and quality can be studied in GIS environment. Various layers
namely slope, geology, distances to drainage channel, tanks and lineaments, depth to water table,
depth of weathered zone can be overlaid and integrated on GIS environment to obtain
groundwater potential map. Similarly, layers of water quality variables may be created to obtain
quality map.
8.10 Closure
RS and GIS are powerful tools for spatial data collection and management. Spatial information
and its attributes can be stored, analysed and output efficiently through GIS. It has many
applications in natural resources and infrastructure management. GIS also has application in
varied industries namely power, transportation etc. Whereas in many applications, it can be
independently used, it also helpful in hydrological modeling in terms of data preparation. For
such applications, varieties of interfaces, stand-alone programs, embedded programs exist. These
increase productivity and reproducibility in modeling.
Table 8.1 Map projections
Equal area
Albers, one standard parallel
Albers, two standard parallels
Lambert zenithal
Cylindrical equal area

Type (Plotting surface)

Conic, tangent
Conic, secant

Lambert conformal, one standard parallel
Lambert conformal, two standard parallel
Mercator (e.g. Universal Transverse Mercator or UTM)

Conic, tangent
Conic, secant

Engman, E.T. and R.J. Gurney (1991). Remote Sensing in Hydrology, Chapman and Hall, London.
Lillesand, T.M. and Kiefer, R.W. (1994). Remote Sensing and Image Interpretation. Wiley, New

Sabins, F.F. (1987). Remote Sensing Principle and Interpretation, W.H. Freeman & Co., San