ANALYZING WATER SYSTEM IN PLAIN AREA BY REMOTE SENSING AND DEM

Abstract

Remote sensing and digital elevation model data are all important data sources in the research on geographical information. Geographical information system has the ability of providing powerful functions to support our research. However, we found it was difficult to study water system in plain area by routine methods because the topography is too flat. A method for analyzing the surface water system in plain area with the help of remote sensing, geographical information system and digital elevation model was given in this paper. Based on remote sensing images and the vertical exaggeration of digital elevation model, it was discovered that the relationship between the distribution of water and digital elevation model could provide the evidence for the cause of water system formation and the process of water system evolvement.

Full text

Page 1

ANALYZING WATER SYSTEM IN PLAIN AREA BY REMOTE SENSING AND DEM


HU Zhuowei a b *, LI Xiaojuan a c, GONG Huili a b, ZHAO Wenji a c, LIU Xiaomeng a c

a College of Resources Environment and Tourism, Capital Normal University, #105 West Sanhuan North RD., Beijing,
100037, China – forevergis@hotmail.com, xiaojuanli@vip.sina.com, gonghl@263.net, zhwenj1215@163.com,
liuxm@sina.com
b Beijing Key Lab of Resources Environment and GIS, #105 West Sanhuan North RD., Beijing, 100037, China
c Key Lab of 3D Information Acquisition and Application, MOE, #105 West Sanhuan North RD., Beijing, 100037,
China


KEY WORDS: Remote Sensing, Digital Elevation Model, Geographical Information System, Water System, Geological Survey


ABSTRACT:

Remote sensing and digital elevation model data are all important data sources in the research on geographical information.
Geographical information system has the ability of providing powerful functions to support our research. However, we found it was
difficult to study water system in plain area by routine methods because the topography is too flat. A method for analyzing the
surface water system in plain area with the help of remote sensing, geographical information system and digital elevation model was
given in this paper. Based on remote sensing images and the vertical exaggeration of digital elevation model, it was discovered that
the relationship between the distribution of water and digital elevation model could provide the evidence for the cause of water
system formation and the process of water system evolvement.


1. INTRODUCTION

* Corresponding author.
With the development of digital elevation model and remote
sensing technology and their extension to other research field,
they have been put more and more attention on and adopted by
many users and researchers. Recently, in China, the
construction of national digital elevation library, which is an
important part of NSDI (National Spatial Data Infrastructure), is
carrying through and stepping toward practicability. At the
same time, remote sensing got the ability to provide massive
earth observing data having many kinds of spatial resolution,
temporal resolution and spectral resolution. Additionally, the
ability of geographical information system is reinforcing
constantly. As a result, these three technologies have been
recognized as indispensable tools in the research on resources
and environment. Moreover, geographical information system
is utilized as supporting method of management and decision-
making.
The fresh water resource is scarce and does not distribute
evenly in China. It is the scientists’ responsibility that service to
the scientific decision-making by macro-investigation and
research on the water resources. In plain area, considering the
special topographical condition, it is very important to study the
method of research on water resources and surface water by
digital elevation model, remote sensing and geographical
information system technologies. Combining these advanced
technologies with the geological research result, the distribution,
cause of formation and evolvement of water system will be
understood. As a result, the judgment or decision about
reasonable utilizing and protection of water resources will be
worked out.
Because the topography of plain area is flat, the effect is not
good when we use digital elevation model to study water
system. However, with the aid of other processing technologies
and combination remote sensing with GIS, good result will be
achieved. In following sections, we introduced measures can be
used to integrate these technologies. With the evidence
supported by geological surveying results, these measures have
been utilized to the ground water research works in the west of
Jilin, China. The achievement in the application was introduced
also.


2. THE MEASURE OF STUDYING SURFACE WATER
SYSTEMS INTEGRATING DIGITAL ELEVATION
MODEL AND REMOTE SENSING
2.1 Technical Route
Remote sensing images contain plenty of information. Digital
elevation model can be used to display the zonal topography.
Although the topography of plain area is flat, it has variety not
only locally but also as a whole. Geographical information
system assists us to integrate and analyze spatial or attribute
data. Geological surveying datum explains the cause of
formation of water systems.
In our research work, various remote sensing datum and non-
remote sensing datum were utilized. By data processing,
integration, information extraction, information compare, we
discovered the distribution and the cause of formation of water
systems. And then, the formation mechanism was analyzed and
the qualitative research was pushed to the quantitative research.
The more precise research result was obtained. Figure 1
explains the technical route.

2.2 Data Processing and Integration
Because the data source is different, the data must be processed
to satisfy the requirements of spatial analysis and demonstration
at the beginning of our work. ArcGIS 9, which was developed
by ESRI (Environmental System Research Institute), is a new
generation of geographical information system software. It

Page 2

provided various powerful functions including data storage,
data management, cartography, map publication, network
analysis and 3D spatial analysis. We made use of ArcGIS 9 in
the process of water system analysis.


Figure 1. Technical route of our research

The mode of data storage is synthetic that put vector layers and
raster data together. Raster data include 1:250,000 digital
elevation model data and Landsat TM images. Vector layers
include the distribution of water resources, political regions,
roads network, relief maps and other geological surveying maps.
In the process of data processing, data transformation and
vector layer rectification are achieved with the aid of powerful
coverage data processing functions provided by Arctoolbox
module.
Before the overlapping and integration of images and vector,
projective transformation must be executed on them because
their geographical references are different. We chose affine
transformation. The result can meet the requirement of our
work (Figure 2).



(a) (b)
Figure 2. Projective transformation. (a) before transformation;
(b) after transformation

Via remote sensing image interpretation, the related information
of water system can be acquired. However, original remote
sensing image itself did not contain geographical positional
information, which we can understand easily. In order to let
them match other spatial data, we must execute georeferencing
on remote sensing images and digital elevation model data.
Firstly, we found surface points those can be recognized not
only on images but also on other geographical data. This type of
surface point was called ground control point. Some links were
established between those ground control point pairs.
According those links, the match from remote sensing images to
the corresponding geographical was achieved.
Using the 3D analyst module in ArcGIS, zonal terrain and water
system distribution can be demonstrated directly. To do it, 2D
water system layers must be converted to 3D feature layers
according digital elevation model data. It was accomplished by
Convert / Feature to 3D command in 3D Analyst module.

2.3 Information Visualization
Remote Sensing Images
Digital Elevation Model
Data Collection
(1) In order to provide better visual effect, we design and
modified styles or symbols for all the map features. For digital
elevation model, height-rendering function was adopted. The
type of symbol includes point symbol, linear symbol and
polygon symbol. Arcmap, which is an import module of
ArcGIS, provided plenty of symbols in the form of style library
to meet our working requirements. Furthermore, we can design
and make symbols conveniently to meet special requirements.
We can change the font of labels and let the size of labels
change according to the map scales. If there is mass of data to
be treated, the running burden of software will be mitigated.
Using Arcmap, we also implemented the map output. After we
set the map’s visual properties, we inserted map title, scale, and
legend on map. High quality water resources distribution maps
were output via powerful cartography module.
(2) 3D visualization of information was implemented in
ArcScene module. The viewer’s position, observing object and
camera parameter can be modified to fulfil omni directional
observation to research area. Because the terrain of research
area is so flat that routine visualization measure could not meet
the analyzing requirements. So, height exaggeration function
was adopted. There is a vertical exaggeration setting option in
general tab of dialog named scene layers. Via this option,
appropriate vertical (height) scale was set.
To facilitate the user’s operation, we used some developing
interfaces provided by ArcObjects, which is a set of COM
(Component Object Model) objects. From these developing
interfaces, we developed a simpler user interface. Vertical
exaggeration scale setting, fly-through demonstration and
section building can be achieved easily by this compact user
interface.
(3) To implement fly-through function, a flying path must be
selected by users at first. According to the digital elevation
model and precision assigned in advance, some points would be
interpolated along the flying path. We modified the viewer’s
position, camera parameter and observing point continually by
programming. Animation was implemented by switching
special number of frames in one second. The key point of
programming is the height value interpolation of surface point
according to the digital elevation model. There is an
InterpolateShape function in ISurface interface. ISurface object
can be instantiated from raster or TIN (Triangulated Irregular
Network) dataset. InterpolateShape has three parameters, which
can be used to assign input data location, output data location
and interpolation step respectively.
(4) Section building need height value interpolation according
to the digital elevation model also. And then, we connected
these points having 3D coordinate in the form of polyline,
which is the principal content of section map.
Geological Surveying
Datum
Topographical Maps
Other Vector Maps
Projection
Data
Processing by
GIS
Correction
Overlap & Integration
Water System Information Interpretation on
Thematic Maps
Comparison with Geological Surveying Datum
Distribution, Cause of Formation and
Evolvement of Water Systems

Page 3

Figure 3 demonstrates the fly-through and section building.


Figure 3. Fly-through and section plot

(5) Data Analysis. After the data processing, we got various
type of data, such as remote sending images, water system
related information (ancient river way, current river way),
geographical structure and so on. Integrating these data and
comparing them, we analyzed the formation mechanism of
water systems according to the digital elevation model and
water system distribution.


3. A PRACTICAL INSTANCE

Figure 4. Integrated map of research area

We applied this method to the research on water system in the
northeast china’s Jilin province. Adopted data included
1:250,000 digital elevation model data, traffic data (polyline),
political region data (polygon), 1:100,000 topographical maps
and geological surveying datum. With the aid of powerful
functions in ArcGIS, such as data management, map production,
3D analyst and so on, we completed the research work of water
system analysis in the west of Jilin. Water system distribution
map, integrated map of remote sensing images and other vector
data (such as traffic maps and political region maps) were
produced. Making use of vertical exaggeration, images and
digital elevation models were overlapped to get 3D atlases and
other thematic maps. Although the topography of research area
is flat, excellent 3D visualization effect was achieved.
From the 3D atlases we founded there is a ridge strip from
Jingjiatai to Huaide-Changling-Taipingchuanzhanyu. The
maximum height difference between this ridge strip and its
neighboring topography is less than 70 meters. It is the
watershed between the Songhua River and the Liao River. From
the remote sensing image interpretation we got to know the
Songhua River had the same ancient river way as the Liao
River. They belonged to the same water system. What was the
reason and how did current status come into being? Combining
geological datum, we held that the Songliao watershed is the
watershed of the Songhua River watershed and the Liao River
watershed. Its eastern terminal is Jingjiatai. It passes Huaide-
Changling in northwest direction and then passes Taipingchuan-
Zhaoyu in the west direction. This watershed is evident between
Jingjiatai to Taipingchuan. It becomes illegible when it reaches
Taipingchuan-Zhanyu. The principal formation period of this
watershed is the end of middle Pleistocene and the beginning of
late Pleistocene. The Nen River intruded into the Liao River
before the middle Pleistocene (including the middle Pleistocene
era). The three most evident and integrate Nen River ancient
river ways, which were reserved nowadays, can be divided into
west branch, middle branch and east branch. Except these three
ancient river ways were formed before the middle Pleistocene,
most of the ancient river ways, which were lying on the low
plain area of the north of Songliao watershed, were come into
being in the late Pleistocene era or the Holocene era.
From the TM remote sensing images and relief maps, the
number of rivers could be distinguished by us was 44. 9 of 44
rivers’ headwaters were the eastern slope of the Great Xing’an
Mountains. The others originated from eastern upland. It is the
upheaval of the Songliao watershed and the middle sunken that
made the segmentation between the Songhua River and the Liao
River. The Nen River did not intrude into the Liao River
anymore. It became a member of the Songhua River. This water
system change promoted the transformation of the Nen River
watershed. The low plain area of the Songliao plain, which
belonged to the Nen River watershed, became to belong to the
Tao’er River watershed. The other part of the Songliao plain
became to belong to the Huolin River watershed. Nowadays,
most of the northern Tongyu city belongs to the Huolin River
watershed. The area from the south of Zhenlai city to Taonan
city belongs to the Tao’er River watershed. The area between
Qian’an city and Changling city has no principal river. There
only has backwater regions or small lakes. Figure 4 is the
integrated thematic map of research area.


4. CONCLUSIONS
It was a good measure that applying remote sensing and digital
elevation model to the water resources research, especially to
the research on water system of plain area. With the mutual
confirmation of spatial analysis result and geological surveying
datum, we got the conclusion that the formation of watershed
between the Songhua River and the Liao River was driven by
geological structural movement. It results in the two big water
systems nowadays. With the aid of vertical exaggeration, zonal
terrain was visualized clearly. Even tiny changes in the terrain
could be discovered. It is important for us to find the position of
upheaval and then provide evidence for the existence of the
Songliao watershed. Thanks to the high precision and high
temporal resolution of remote sensing and digital elevation
model, we had excellent assistant tools to make qualitative
geological survey transform to quantitative geological survey.

Page 4

5. REFERENCES
Bob B., 2000. Using ArcGIS 3D analyst. ESRI.
Michael M., 2000. Using ArcMap. ESRI.
Melita K., Steve K., 2000. Understanding map projections.
ESRI.
Practical guide of ArcGIS 3D analyst. Beijing Zhongke
Yongsheng Data Technology Co., Ltd.
ArcObjects developer help. ESRI.
~, 2000. Integrated surveying reports on human living
environment remote sensing in the west of Jilin. College of
Earth Sciences, Jilin University.