Information about http://www.rrcap.unep.org/semis/pubs/consultancyreportofMike.pdf
REVIEW OF AVAILABILITY AND ACCESSIBILITY OF GEOSPATIAL DATA IN THE…
Tags: consultancy report, core data, data exchange format, data layers, database design, environmental monitoring, geospatial data, information system, information systems, introduction 1, mekong, metadata standards, regional databases, semis, spatial data exchange, spatial database, starbuck, topography, united states geological, united states geological survey,Pages: 26
Language: english
Created: Fri May 11 15:40:29 2001
Display cached document
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Page 11
Page 12
Page 13
Page 14
Page 15
Page 16
Page 17
Page 18
Page 19
Page 20
Page 21
Page 22
Page 23
Page 24
Page 25
Page 26
REVIEW OF AVAILABILITY AND ACCESSIBILITY OF
GEOSPATIAL DATA IN THE GREATER MEKONG SUBREGION
A consultancy report in support of the Subregional Environmental
Monitoring and Information Systems II Project (SEMIS II)
Prepared by Michael Starbuck
United States Geological Survey
March 2001
TABLE OF CONTENTS
page
1.0 INTRODUCTION 1
2.0 BACKGROUND AND OBJECTIVES 2
2.1 Objective of this activity 3
2.2 Approach 3
3.0 REVIEW OF EARLIER PROJECTS 3
3.1 Sub-regional Environmental Monitoring and Information
System (SEMIS I) 3
3.1.1 Output 1 Core dataset definition 4
Topography and hydrography 5
Data collection issues 5
Framework perspective 5
Additional core data layers 6
Responsibility for data layers 6
3.1.2 Output 2 Spatial database design 7
3.1.3 Output 3 Data standard for information to be held in
national and sub-regional databases 8
Spatial data exchange format standard 9
3.1.4 Output 4 Metadata standards for information to be
held in National and sub-regional databases 9
3.1.5 Output 5 Catalogue of existing environmental and
natural resources data holding amongst countries
of the GMS 9
3.2 Strategic Environment Framework for the GMS (SEF project) 10
3.2.1 Early Warning and Information System (EWIS) 11
4.0 REVIEW OF AVAILABLE DIGITAL GEOSPATIAL DATASETS 12
4.1 Map accuracy issues 12
4.2 Map projections and coordinate systems 12
4.3 Recommended data scales/resolutions 13
4.4 Data accessibility 13
5.0 DATA GAPS 14
6.0 CONCLUSION / RECOMMENDATIONS 14
TABLES
page
Table 1 Datasets used by EWIS 11
APPENDICES
APPENDIX 1: Thailand Available Datasets 16
APPENDIX 2: Vietnam Available Datasets 17
APPENDIX 3: Cambodia Available Datasets 19
APPENDIX 4: Lao Available Datasets 21
APPENDIX 5: United States National Map Accuracy Standards 23
Abbreviations
ADB Asian Development Bank
ASCII American Standard Code for Information Interchange
CD-ROM Compact Disc Read Only Memory
CIESIN Center for International Earth Science Information
DEQP Department of Environmental Quality Promotion
ESRI Environmental Systems Research Institute
EWIS Early Warning Information System
FAO Food and Agriculture Organization
FGDC Federal Geographic Data Commission
GIS Geographic Information System
GMS Greater Mekong Subregion
GRID Global Resource Information Database
ICIMOD International Center for Integrated Mountain Development
IIMI International Irrigation Management Institute
ISO International Standards Organization
LIDAR Light Detection and Ranging
Landsat ETM Earth Resources Satellite (Enhanced Thematic Mapper)
MRC Mekong River Commission
NAMRIA National Mapping and Research Information Authority
NASA National Aeronautics and Space Administration
NEA National Environment Agency
RRC.AP Regional Resource Center for Asia and the Pacific
SDTS Spatial Data Transfer Standard
SEF Strategic Environmental Framework
SEMIS Subregional Environmental Monitoring and Information System
SPREP South Pacific Regional Environmental Programme
TA Technical Assistance
TIFF Tagged Image File Format
UNEP United Nations Environment Programme
USGS United States Geological Survey
WWF World Wide Fund for Nature
1.0 Introduction
The Asian Development Bank (ADB) is implementing a Technical Assistance (TA) for the
Subregional Environmental Monitoring and Information Systems Phase II (SEMIS II) in
collaboration with the United Nations Environment Programme Regional Resource Center
for Asia and the Pacific (referred to as the Center). The project is co-financed by ADB
(through the Japan Special Fund and the Government of Norway) and the Center.
A follow-up of SEMIS I was requested by the Greater Mekong Subregion (GMS) countries at
the Fourth Meeting of the GMS Working Group on Environment held in Hanoi in March
1998. GMS countries are Cambodia, Yunan province of the People's Republic of China, Lao
People's Democratic Republic, Myanmar, Thailand, and Vietnam. The GMS Ministerial
Meeting held in Manila in September 1998 endorsed the request. The project was approved
by ADB on 29 December 1999.
The objective of the Technical Assistance is to build upon the achievements of SEMIS I
which include: (1) a defined core dataset, (2) a conceptual spatial database design, and (3)
technical capacity for the exchange of data. The overall goal of SEMIS II is to help GMS
governments to make informed decisions regarding sustainable development through
integrated economic and environmental planning.
In January 2001, the United States Geological Survey (USGS) was asked to provide expertise
for the SEMIS II project in terms of geographic information systems and spatial databases.
Mr. Michael Starbuck, USGS, spent 8 weeks at the Center from January 29 to March 23,
2001. Terms of reference included: (1) Review and analyze available information at the
Center to determine the usefulness and relevance for operational level development planning
for the GMS Hotspot areas, (2) Review the identified gaps and suggest additional
data/information needs that may be required for the operational level planning purposes, (3)
Suggest the appropriate scale and format for GMS spatial data used for operational planning,
and (4) Review the data collection and management guidelines prepared by the SEMIS I team
and suggest revision and refinements. This report summarizes items 1 through 3, with a
separate report, entitled "Draft Data Collection and Management Guidelines", covering item
4.
Due to time constraints and scheduling difficulties, the following items from the terms of
reference were not completed: (1) Assist in analyzing temporal land use/land cover changes
in the GMS countries using available software packages, (2) Assist in organizing national
training seminars to be conducted in Kumming-Yunnan, PRC, Vientiane, Lao, PDR, Phnom
Penh, Cambodia, Hanoi, Vietnam, and Bangkok, Thailand. At these seminars, make
presentations on collection and processing of data/information, and (3) Assist in evaluating
the status of data processing in the GMS countries and suggest personnel and material
(hardware/software) needs to establish or strengthen to data centers, as appropriate. The
SEMIS II project is not advanced sufficiently to yet schedule national training seminars. A
trip was made to Phnom Penh, 1- 3 March 2001, to attend a Strategic Environmental
Framework project meeting, and to visit the Mekong River Commission office.
2.0 Background and objectives
The overall goal of the SEMIS II project is to help the GMS governments to make informed
decisions regarding sustainable development through integrated economic and environmental
planning. A key component of informed decision making is having access to reliable
information, including spatial data. The GMS countries, because of their geographic
locations, share common environmental problems and therefore have a common need to
share environmental information in a timely manner. SEMIS I, the Technical Assistance
project that was the precursor to SEMIS II, established the groundwork for the sharing of
information on environmental and natural resources issues.
A related on-going Technical Assistance project is the Strategic Environment Framework
(SEF) for the GMS (TA No. 5783). This project has similar goals of integrating economic
and environmental planning and is also collecting geospatial data for selected areas in the
GMS.
SEMIS II aims to build upon the achievements of SEMIS I by undertaking pilot
demonstration projects, further developing subregional/national databases, reviewing the
current mechanism for collection of identified core data, and defining the best approach to
performing data collection, storage, manipulation, and transmission.
The key objectives of SEMIS II include:
1) Assess the availability of useful and relevant data for planning purposes;
2) Increase and strengthen the capacity of national governments to collect and
process the information/data;
3) Increase the capacity of national governments to make informed decisions
regarding development investments relating to sustainable use of natural
resources;
4) Enhance the ability of GMS national governments to conduct integrated
economic and environmental planning with relevant data; and
5) Conduct, store, manipulate and share actual integrated planning information
using the data collected in pilot projects for some "Hotspot" areas, such as
those identified in TA 5783-REG: Strategic Environmental Framework for the
GMS.
The following are the specific outputs planned for the SEMIS II project (see the SEMIS II
Implementation Plan for detailed information):
1) Available data and data gaps,
2) Integrated economic and environmental planning procedures and background papers,
3) Capacity building plan on hardware/software support and training needed,
4) Guidelines for data collection and data management,
5) Hardware and software support,
6) Internship-cum-training for six national coordinators,
7) Enhancement of a sub-regional network,
8) Study tour of national coordinators/finance personnel,
9) "Hot spot" database (1:50K),
10) Baseline data of GMS (1:250-500K),
11) Case studies,
12) Sub-regional and national training/seminars/workshops and study tours,
13) Project reporting and management.
2.1 Objective of this activity
This report addresses the activities supporting the SEMIS II Objective 1 (Assess the
availability of useful and relevant data for planning purposes), and specifically Output 1
Available Data and Data Gaps. The product is this report on the availability and accessibility
of data and data gaps.
2.2 Appro ach
The following steps were used in determining the availability and accessibility of data and
data gaps:
1) Review outputs of earlier projects (SEMIS I and SEF);
2) Identification of core datasets required for analysis;
3) Inventory of available datasets using the Data Catalogue; and
4) Identification of data gaps.
3.0 Review of earlier projects
This section summarizes the main points of the earlier projects and presents
recommendations on ways to improve upon the results.
3.1 Subregional Environmental Monitoring and Information System (SEMIS I)
SEMIS I is the Technical Assistance project financed by the Asian Development Bank (T.A.
No. 5622-REG), which was the precursor to the current SEMIS II project.
SEMIS I was approved in January 1995 and was completed in November 1999. There were
17 different ouputs from the SEMIS I project, with responsibility for individual outputs
varying between the Center, the ADB project team headed by Roche International of Quebec,
and the Mekong River Commission. This review will focus primarily on the following
outputs: (1) core dataset definition, (2) spatial database design, (3) data standards, (4)
metadata standards, and (5) catalogue of data holdings.
3.1.1 Output 1 Core Dataset Definition
A core dataset is defined as,
"Identification and description of the `core' or `minimum' set of spatial information needed
to support national and subregional environmental assessment, decision making, and
environmental reporting."
Another definition used is,
"The basic, frequently required data necessary for the range of environmental decisions
which will arise in subsequent years."
The core dataset definition resulting from the SEMIS I activities was designed to support a
wide range of national and subregional environmental decision making and analysis. By
definition, the core data sets would be collected and integrated across the GMS countries
according to established guidelines and standards, thus making them useful to a variety of
decision makers in a timely fashion.
The SEMIS I activities which lead to the core dataset definition were:
· review of previous studies of relevance in the region
· preparation of a draft set of core datasets
· consultation with the six countries, and
· consolidation of the findings into the final core dataset definition.
The following are the 13 core datasets as defined by the SEMIS I team:
1. Infrastructure
2. Soil Class
3. Vegetation Cover
4. Air Quality Measurements
5. Demography
6. Climate Zonation
7. Administrative Boundaries
8. Topography (and Hydrography)
9. Land Use
10. Geology
11. Major Harvesting Activities
12. Water Quality Measurements
13. Soil Analysis Samples
Topography and hydrography
While definitely related, topography and hydrography, in my opinion should be treated as
separate data layers. These data layers can be derived independently using modern
technology (i.e. LIDAR and interferometry) and many users may wish to use one or the other
independently.
Data collection issues
Infrastructure, administrative boundaries, topography, and to some extent, vegetative cover
and land use, are all data layers which historically can be found on traditional topographic
maps. Initial data collection efforts can concentrate on the digitization of existing map
features from paper maps. A shortfall to this approach, however, is that you are constrained
by the interpretations performed by the original mapmaker. Questions that must be
considered include: what features did they actually collect, and what classification scheme
did they use and does it match your own? The best approach is to clearly document the map
features collected and the classification scheme used, preferably in a standard metadata
format, so that subsequent users know exactly what kind of information is present in the data
sets.
The standards for the core datasets as defined by the SEMIS I project have established
guidelines for feature collection (at least in the case of infrastructure), which have not
considered the original data source. For example, the attributes available to the map feature
"Main Road", include the following possibilities: international, national, and secondary. This
is a very generic, simple classification scheme. However, the most likely data source for
these map features will be the national 1:250,000 scale topographic maps. Roads on these
maps are classified according to surface type and number of lanes. There is no clear and
reliable translation from that scheme to the other. Determination of whether a road is
national or international may be very subjective. Besides taking advantage of the existing
map symbolization (and classification scheme), a scheme based on observable characteristics
is preferable to one based on interpretation of a features' use.
Framework perspective
Another perspective to consider is that from the Federal Geographic Data Commission
(FGDC). The FGDC has identified a set of data layers it has named the "Framework". These
data layers are the common themes needed by most data users and include transportation,
hydrography, geodetic control, digital imagery, boundaries, elevation, and cadastral layers.
To make the use and distribution of framework data easier, the FGDC is proposing certain
technical, operational, and institutional contexts, such as a feature-based data model,
permanent feature identification codes, references to datums, and metadata.
While the core data list proposed by SEMIS I is more inclusive, having been designed with a
more specific user base in mind, many of the concepts behind the framework design would
apply and it may be worth examining in more detail. See the FGDC Framework website at
http://www.fgdc.gov/framework/framework.html.
Additional core data layers
Another data layer usually listed as a core dataset is imagery. Imagery can be in many forms,
including digitized air photos, digital orthoimagery (airphotos processed to remove
distortions due to terrain relief), satellite imagery, and scanned paper maps. Landsat 7 ETM+
scenes are relatively low cost and have no re-distribution restrictions (the data is not
copyrighted). Landsat scenes are an excellent data source for a variety of map features, from
basic infrastructure to land cover.
Scanned paper maps are an inexpensive alternative to performing vector digitizing of
topographic maps. The original paper maps can be scanned wherever a service provider has
large format scanning capability and then the raster image can be georeferenced using
standard geoprocessing software. The result is a georeferenced im age of the topographic map
that can be used for a number of applications, ranging from performing heads-up digitizing of
map features to serving as an inexpensive base map for other project data.
Slope and aspect are two data layers often critical to certain kinds of analysis. While these
layers are derivable from the elevation dataset, some consideration might be given to having
these layers pre-existing, for those users who may not have the capacity to create them
themselves.
While the data layer, Water Quality Measurements, is one of the core datasets, environmental
impact analysis could benefit from more detailed surface and subsurface hydrologic data.
Currently only water sample sites are specified, listing various characteristics of the sample
(pH, conductivity, chemical analysis, etc.). Additional measurements like depth to
groundwater, surface water hydrographs, basin delineation, and subsurface flows would be
valuable to performing analysis of potential impacts of development projects.
Responsibility for data layers
Creating lists of core, required data sets is a useful exercise to perform. It gets the
stakeholders thinking about what kinds of datasets they may regularly require in their
decision-making activities. However, for these data layers to become more than "wish lists"
on paper, responsibility must be assumed for their creation, maintenance, and distribution.
One of the weaknesses of the SEMIS I output concerning core datasets is the lack of a strong
assignment of responsibility for the individual datasets. One approach to this is to get the
appropriate agency or group most aligned with a particular data layer to sign on as the data
supplier for that data set. For example, the government agency responsible for forestry may
be the best equipped to provide the vegetation cover data layer. The transportation
department would be most likely to provide at least some of the infrastructure layer. Of
course, this will not work in many cases, and in the GMS countries, only a handful of
agencies may be capable of creating and distributing geospatial data. This brings us back to
the basic underlying premise for the SEMIS projects establish among the GMS countries a
mechanism for the efficient exchange of geospatial data for timely economic and
environmental planning.
3.1.2 Output 2 Spatial database design
From the SEMIS I implementation document A conceptual level spatial database design for
a hierarchical subregion wide GIS to support national and sub-regional environmental
assessment, decision making and environmental reporting. The design should consider a
distributed system of inter-linked spatial databases at the national level (target scale 1:50K)
which can be integrated into a sub-regional GIS (target scale 1:250K). The primary function
of the database is to manage and analyse the Core Dataset defined in Output 1.
The SEMIS I team used a list of criteria to help govern the development of the conceptual
database design. The database should be:
· Decentralized,
· Hierarchical,
· Spatially based,
· Expandable and flexible,
· Easy to use and maintain,
· Built using appropriate technology,
· Compatible with UNEP State of the Environment Database, and
· Compatible with Existing Subregional Databases.
The conceptual database design is reasonable and logical in its structure. A national hub in
each country acts to obtain and exchange data with other countries using the standard
exchange formats. The subregional hub, probably an international agency such as UNEP or
MRC, will link to the national hubs and other international agencies to exchange data.
A key component of the design is the concept of decentralization and that there should be no
duplication of data. At the national level, a number of different agencies will hold various
components of the core datasets. Data exchange would occur through the hub agencies.
The SEMIS functional design discusses the kinds of functions that will be required of a
SEMIS system to support environmental monitoring and reporting:
· Manage Core Datasets
· Manage Non-core Datasets
· Produce SEMIS Outputs
· Manage Auxillary Datasets
· Manage Dataset Availability
· Manage Dataset Exchange
· Convert Existing GIS Datasets
The subsystem "Produce SEMIS Outputs" contains major subcomponents:
· Spatial Data Management
· Spatial Data Analysis
· Spatial Output and Display
· Non-spatial Analysis
To provide these kinds of functions, a full function GIS is required. The SEMIS report
indicates not every participating agency will need to have this functionality and that it may be
best initially to concentrate GIS capability in one or a few centers in each country. For the
majority of users, desktop computers running the latest version of Arcview will provide this
functionality.
The FGDC is developing a metadata distribution mechanism that may have implications for
data distribution in the GMS. It is called a Data Clearinghouse and is a decentralized system
of servers located on the Internet that contains descriptions of available digital spatial data
(metadata). The fundamental goal of the Clearinghouse is to provide access to digital spatial
data through metadata. For more information, see the website at
http://www.fgdc.gov/clearinghouse/clearinghouse.html
.
3.1.3 Output 3 Data standard for information to be held in national and sub-regional
databases
From the SEMIS I implementation document: Following from the Core Dataset Definition
(Output 1) these standards and guidelines are to ensure the feasibility of data exchange and
integration. They will include preferred classification systems for core data items, spatial
data frameworks and recording standards, and standard data interchange formats. They are
not meant to extend to standards for data collection, measurement methods, data coding, or
quality control procedures.
A great deal of work has gone into developing a detailed data standard for the core datasets.
The general approach taken was to rely on the organization of expertise for a given data
layer. For example, the standards adopted for the soil class data set are those defined by FAO
in the Global and National Soils and Terrain Digital Databases (SOTER) Procedures Manual
(FAO, 1993).
The attribute scheme for the infrastructure datasets is a bit rudimentary, and as discussed
earlier, the relationship between the desired classification scheme and those already in use
may present a problem.
The main point for discussion here is whether the standards are, or will be, used. There is
little evidence that the data standards are currently being used, even by the international
agencies in the region. Perhaps as the national agencies begin their data collection activities
in earnest, they will follow the proposed standards. More likely, however, is that as the
individual agencies create their own data sets, they will devise a scheme of their own design,
emphasizing the features and attributes that are important to them. Adhering to standards is a
difficult process, especially amongst agencies and countries that may not see doing so in their
best interest.
Spatial data exchange format standard
Internationally, a great deal of work is going into devising data exchange formats. Of
particular note are the USGS Spatial Data Transfer Standard (SDTS), and the efforts of ISO
Technical Committee 211. It was the recommendation of the SEMIS I group to use the
Arc/INFO Export Format for vector GIS data as an interim standard for data exchange. This
was to be in effect until the ISO 211 standard was complete. As GMS countries will be using
commercial GIS software packages, it makes sense to use an existing exchange format
supported by the software. Arc/INFO Export Format will be suitable in the vast majority of
cases.
The ASCII coded headerless files have been selected as the interchange format for raster
files. This is suitable as a basic common format, but in practice, other widely used raster
formats will probably be used more often, such as TIFF and GeoTIFF.
3.1.4 Output 4 Metadata standards for information to be held in national and sub-
regional databases
The agreed interim standard format for metadata is that developed by UNEP-GRID. This is
already used in the sub-region and is compatible with the major metadatabases of NASA,
CIESIN, WCMC and others. An existing metadata entry tool, a Microsoft Access-based
software package developed by UNEP, is available for creation of new metadata files.
This standard is a good choice for metadata, as it is already accepted and based on
international standards. Metadata creation is a difficult and time-consuming task. If more
complicated metadata standards were required, it is possible that little metadata would
actually be created. The practice of creating metadata should be reinforced at every
opportunity. When the training for the national coordinators is conducted, a standard
metadata file should be created for each spatial data file created by the trainees.
3.1.5 Output 5 Catalogue of existing environmental and natural resources data
holdings amongst countries of the GMS
The Center maintains a catalogue of datasets held by agencies in the Asian and Pacific
regions. The catalogue is updated every 6 months. The issue reviewed was dated October,
2000. Eleven institutions contribute information concerning data holdings. They are:
International Center for Integrated Mountain Development (ICIMOD) in Nepal; Mekong
River Commission (MRC) in Cambodia; South Pacific Regional Environment Programme
(SPREP) in Western Samoa; UNEP -GRID in Bangkok; Landcare Research in New Zealand;
Ministry of Environment in Cambodia; International Irrigation Management Institute (IIMI)
in Pakistan; National Mapping and Research Information Authority (NAMRIA) in
Philippines; Department of Environmental Quality Promotion (DEQP) in Thaiand; National
l
Environment Agency (NEA) in Vietnam; and SENRIC Project (SACEP) in Sri Lanka.
While the Center maintains the list, only those datasets listed under the UNEP-GRID data
holdings may be obtained from that office. For datasets of other agencies, one must contact
that agency directly using the supplied contact information. Datasets can be distributed
unconditionally, distributed with source approval, or made available for in-house use only,
depending on the particular dataset. This service was originally provided free of charge, but
a $50 per data request fee is now applied. The data listing is also available from the UNEP
website at http://www.eapap.unep.org/fs-datacat.html
.
The data listing gives the following information for each dataset: code, title, type(vector or
raster), general location, date, scale, and size of digital file. This listing is a wonderful
resource for those looking for spatial data. From the website version, users can access
metadata listings for a selected subset of the datasets listed. Metadata files need to be created
for all the datasets listed in the catalogue. Actual data requests seem to be very infrequent,
with approximately 2 requests coming in to t e Center in the past year. When a data request
h
is received, it must be made clear to the user that a fee is required. Then the appropriate files
are found on the stored CD-ROMs and copied to the medium of choice, either 3.5 inch
diskette, or CD-ROM. Occasionally the user will request some additional processing, such as
data subsetting, or conversion to other coordinate systems.
Hopefully, in the future, additional agencies will participate in this data listing. Currently, of
the GMS countries, Lao, Myanmar, and China are not represented in the Data Catalogue.
There are also undoubtedly many more agencies that have geospatial datasets that are not
currently participating in the catalogue. Also, the majority of the datasets listed in the
catalogue are small scale, at 1:250,000 or smaller. Most agencies are reluctant to make larger
scale datasets available outside their department, let alone to the general public or other
countries. Perhaps some kind of incentive can be devised to encourage more participation
and the release of larger scale datasets.
3.2 Strategic Environment Framework for the GMS (SEF project) TA No. 5783
The overall objective of the SEF Project is to promote a better understanding of
environmental and social impacts of planned development in the GMS. In particular,
emphasis is given to the energy/water resource and transportation sectors of the ADB's GMS
Programme. A key component is to help ensure that environmental and social aspects are
considered at an earlier stage in the planning process than currently takes place.
SEF outputs include:
· A report that will provide a framework of operational, policy, and institutional
recommendations designed to better ensure the environmental and social
sustainability of economic development;
· A list of recommended GMS Technical Assistance (TA) projects and environmental
investments;
· A set of maps and GIS databases on baseline conditions in the region;
· A set of maps and GIS databases on GMS environment-development "Hotspots" and
Highly Valued Areas;
· GMS development scenarios;
· A GIS-based GMS Early Warning and Information System (EWIS).
3.2.1 Early Warning and Information System (EWIS)
The EWIS is an interactive software tool, built upon the Arcview software package from
Environmental Systems Research Institute (ESRI). The tool requires Arcview to already be
installed on the system to work. Through the use of modified menus and tools in Arcview's
graphical user interface, the EWIS presents the user with an easy way of viewing and
performing simple queries on GMS datasets. The system provides information to the user on
Highly Valued Areas, the environmental and social status of the GMS, development plans in
the GMS, and priority Hotspots. It also allows the user to see, in a general way, possible
impacts to proposed development projects. The user can draw a proposed new highway, or
the location of a proposed dam, and the system will automatically calculate what other data
layers are affected.
Most of the geospatial data associated with the EWIS is at the 1:1 million scale. Table 1 lists
most of the datasets accessed by the EWIS. This level of detail is appropriate for viewing
features and relationships at the regional level, but the user must keep in mind the constraints
imposed by such a relatively small-scale database. Arcview will allow the user to zoom in
almost indefinitely, but the appropriate display scale of the data can be quickly surpassed,
perhaps leading to incorrect conclusions.
In addition to the regional datasets, the SEF project is collecting higher resolution data for
specific areas termed "hotspots". These are areas determined to represent the conflict
between economic development and environmental and social goals in the GMS. Five
hotspot areas have been delineated in the GMS region. The detailed hotspot analysis will be
undertaken at the 1:250,000 scale using data provided by the MRC, as well as information
collected by the SEF team and by UNEP.
File Name Data Description Scale Source
GMS_bnd.shp Country Boundaries for the Greater Mekong Subregion 1:1M UNEP
Asia_bnd.shp Regional Boundaries for other countries in Asia 1:5M ESRI
Gms-riv.shp Greater Mekong River System 1:1M UNEP
Gms-wet.shp Waterbodies in the GMS 1:1M MRC, UNEP,China
Hotspot.shp SEF Priority Hotspots 1:1M SEF
Pov_cam.shp Poverty data for Cambodia - MoE, Cambodia
Pov_lao.shp Poverty data for Lao PDR - STEA, Nat. Stat.
Pov_mya.shp Poverty data for Myanmar - For. Dept. Myanmar
Pov_thai.shp Poverty data for Thailand - UNEP
Pov_viet.shp Poverty data for Vietnam - -
Pov_yun.shp Poverty data for Yunnan 1:1M YEPB
Gms_hva.shp Highly Valued Areas in the GMS 1:1M SEF
Mrc_forest.shp MRC forest cover for the lower Mekong Basin 1:250K MRC
Wwf_bio.shp WWF for Nature Biodiversity Priorities for the fo rests 1:1M WWF
Mrc_wet.shp MRC data for wetlands in the lower Mekong Basin 1:250K MRC
Unep_land.shp UNEP Landcover data for SE Asia 1:1M UNEP
Gms_pa.shp Protected areas in the GMS 1:1M UNEP
Gms_road.shp Roads in the GMS 1:1M UNEP
Gms_road_prj.shp Proposed ADB road projects 1:1M ADB
Gms_dam.shp Current Dam locations 1:1M UNEP
Gms_city.shp Locations of cities 1:1M UNEP
Gms_prov.shp Provincial boundaries in the GMS 1:1M UNEP
Gms_dist.shp District and county boundaries in the GMS 1:250K MRC
Table 1 Datasets used by the EWIS
4.0 Review of Available Digital Geospatial Datasets
The primary document used for this review was the Data Catalogue, published by Center,
October 2000. The procedure used for review was to examine the list of Core Datasets and
then populate a table, indicating a dataset file name where it was determined that it satisfied
the Core Dataset definition. In this way, the tables in appendix 1 were constructed, showing
where datasets were avai able, and where they were not. A limitation to this method was that
l
the actual datasets were not examined, only the short description in the Data Catalogue, or in
some cases, an actual metadata listing, was all that was available for determining suitable
matches to the Core Data list.
A series of asterisks (*) in the file code field indicates no match was found in the Data
Catalogue for this entity. No tables were generated for China or Myanmar, as very little data
at the larger scales is listed in the Catalogue for these countries.
More datasets exist than what is listed in the Data Catalogue, especially for Thailand. These
additional datasets need to be documented as to their existence, either in the Data Catalogue,
or in a separate listing.
4.1 Map accuracy issues
Throughout the SEMIS I reports and other related documents, there has been little mention of
map and data accuracies. It is usually desirable to determine the level of accuracy of a given
map or series of maps. There are many kinds of accuracy assessment, including how
complete a map is (i.e. for a road map, are all the existing roads depicted on the map?), the
attribution (does that road have the correct identifier?), to positional accuracy (is the road
shown in the right place). There is also absolute and relative positional accuracy (not only is
the road in the right place, but is that house on the correct side of the road?). Accuracy
assessment can vary from a rigorous examination of each dataset, to a statistical sampling of
selected datasets. Generally you can rely on the process (techniques and methods used to
create the map) to retain the desired quality and accuracy in the product (final dataset), but
occasionally verification is desired to be certain the required accuracy is indeed being
achieved. I recommend creating some data validation plan that includes assessment of map
accuracy and applying it to datasets generated by the project. Appendix 2 discusses how U.S.
National Map Accuracy is defined and measured.
4.2 Map projections and coordinate systems
Some mention needs to be made of which map projections, datums, and coordinate systems
are to be used for the SEMIS II project. As most of the map datasets are being digitized from
existing map sources, they will already have some projection and be cast on a particular
datum. Will all datasets be transformed into a common set of datum, projection and
coordinate system? Geographic coordinates are a good general purpose coordinate system,
but the user often must transform the dataset into a projected map before they can easily
make measurements. Arcview will allow the dataset to remain in geographic coordinates and
automatically perform a transformation to a specified projection so that distance and area
measurements can be taken. In Arcview, if some of your data is in geographic, all other data
used in that view must also be in geographic for proper registration to occur.
4.3 Recommended data scales/resolutions
The key to establishing recommended data scales is to examine the kinds of measurements or
analysis one hopes to perform with the data. To use the EWIS from the SEF project as an
example, its primary objective is to provide users with an overview of various developmental
and environmental aspects of the GMS region and allow for identification of potential
impacts of proposed developments. Most of the data layers being used by the EWIS are at
the 1:1 million scale. If we allow that the datasets meet a level of horizontal accuracy
commensurate with the U.S. National Map Accuracy Standard (which may not be the case), a
well-defined point in the dataset can only be expected to be within approximately 500 meters
of its true horizontal position. This is only a rough estimate and will vary according to a
number of factors. For the EWIS application, this is an acceptable value, as users will not
(hopefully) be trying to make measurements at a resolution finer than this. The horizontal
accuracy of curvilinear features will probably not be as reliable as well-defined points, and
when you overlay multiple layers, you must account for the possible accumulation of errors.
Scale also implies a certain level of content detail. A road layer created at the 1:50,000 scale
will have more detailed roads (both in showing smaller roads and more detail of the
linework) than a road map at a scale of 1:250,000.
For analysis and planning purposes at the regional scale (the extent being across several GMS
countries), a map scale from 1:500,000 to 1:1 million is reasonable. For more detailed
analysis at the level of area hotspots (extents of about 300 kilometers), a map scale of
1:250,000 or larger is desirable. This scale implies a maximum positional error of well-
defined points of about 130 meters. Again, these positional error calculations are rough
estimates, and there are many variables that should be considered.
4.4 Data accessibility
For the datasets listed in the Data Catalogue, access can be considered fair to good.
According to the documentation associated with the Catalogue, the user simply mails a data
request form to the appropriate agency indicating the datasets desired, distribution media, and
any special requirements. Depending on the data requested (and possibly on the agency
being dealt with), the process of requesting and then receiving a particular dataset may take
anywhere from a few weeks to several months. An added complication is the fact that some
agencies are charging distribution fees and are not advertising this fact. Also, the dataset
description should be more detailed, so that the users can be sure of getting the datasets they
require.
While this dataset distribution scheme is not as convenient as online Internet access, it is
better than nothing and has the added convenience of bringing together datasets from a
variety of agencies in the region. Possible weak links include the need for continuity of
communication between a coordinator at the UNEP office and contacts at the respective
agencies and the establishment of efficient local data archives at each agency. As the
technological infrastructure improves in the region, migration to some Internet-based data
distribution will be easier because of the current scheme.
5.0 Data gaps
In reviewing the available data compared to the desired core datasets, several data gaps are
noticed. Consistent among the various GMS countries is the apparent lack of certain
infrastructure layers, such as electric transmission lines, pipelines, dams, ports, and airports.
The SEF project does use a ayer in its EWIS software for dams in the GMS region, but the
l
source scale is 1:1 million. Given adequate topographic maps, these data layers could easily
be collected in a minimal amount of time.
Another layer typically not available is the air quality measurements. These would most
likely be large amounts of tabular data linked to a point location. If the source for the tabular
data can be obtained, creation of spatial data layers would be fairly easy. The same is true for
the demography and water quality data layers.
The last group of data layers consistently not available are the Major Harvesting Activities
datasets, including agriculture, forestry, mining, and fisheries. These data layers are defined
primarily as attributed polygons at the province or district level. If the tabular data can be
obtained, linking it to existing provincial or district spatial data would be fairly
straightforward.
6.0 Conclusion / Recommendations
This report addresses Output 1 of the SEMIS II project Available data and data gaps. The
approach used was to review the outputs of the SEMIS I and SEF projects, identify those core
datasets required for analysis, inventory available datasets using the Data Catalogue, and
identify the data gaps. Areas of apparent data gaps include:
· Infrastructure (electric transmission lines, pipelines, dams, ports, and airports),
· Air quality measurements,
· Demography,
· Water quality measurements, and
· Major harvesting activities.
The core dataset list, data standards, and metadata standards as defined by the SEMIS I team
were examined and the following are suggested improvements:
· Separate topography and hydrography into separate layers or themes;
· Consider feature classification schemes and tailor to meet requirements and existing
data sources;
· Consider applying a framework approach to the SEMIS II database activities;
· Possible additional core data layers include:
o Remotely sensed imagery,
o Raster topographic maps,
o Slope and aspect, and
o Detailed hydrologic information;
· Assign responsibility for data layers;
· Use an easily populated metadata standard;
· Arc/Info export files are a suitable interchange format;
· Devise and use a data validation and accuracy assessment plan;
· Address the issues of multiple datums, projections, and coordinate systems;
· Clearly advertise data distribution fees;
· Strengthen procedures for maintenance of the database and the data catalogue; and
· Generate metadata files for all data catalogue listings.
Entity File Code Scale Date Comments
Infrastructure
Main Road ALL0008 250,000 Provincial datasets from OEPP
Railway ****************** (ALL0008 refers to CD-ROM)
ElectricTransmission Line ******************
Pipeline ******************
Dam ******************
Port ******************
Airport ******************
Soil Class
Soil Unit ALL0008 250,000 Provincial datasets from OEPP
Vegetation Cover
Land Cover Unit THA 0028 1,500,000 1991 Forest-Nonforest of Thailand
THA 0033 1,000,000 85-86 Land Cover map of Thailand (85-86)
THA 0034 1,000,000 92-93 Land Cover map of Thailand (92-93)
TH -WL 100 250,000 1997 Wetlands in the Mekong Corridor, Thailand
ALL0008 250,000 Provincial datasets from OEPP
Air Quality Measurements
Air Quality Observation ******************
Demography
Demographic Unit ******************
Climate Zonation
Agro-climatic Zone ******************
Administrative Bdys
Administrative Unit ALL0008 250,000 Provincial datasets from OEPP
TH-AM001 -017 50,000 1969 Administrative maps of NE Thailand from MRC
Topography
Elevation MRC DTM's ?
ALL0008 250,000 Provincial datasets from OEPP
Water Boundary / Body ALL0008 250,000 Provincial datasets from OEPP
Land Use
Land U se Unit ALL0008 250,000 Provincial datasets from OEPP
Geology
Geological Unit TH-GL008 - 100 250,000 Geological maps of NE Thailand from MRC
Major Harvesting
Activities
Agriculture ******************
Forestry ******************
Mining Location ******************
Fisheries ******************
Water Quality Meas. ******************
Soil Analysis Samples ******************
Appendix 1: Thailand Available Datasets
Entity File Code Scale Date Comments
Infrastructure
Main Road NEA 3 250,000 1997 Topography map of vietnam
Railway NEA 3 250,000 1997 Topography map of vietnam
ElectricTransmission ******************
Line
Pipeline ******************
Dam ******************
Port ******************
Airport ******************
Soil Class
Soil Unit VN-SL 001 250,000 1989 Soil Map of Mekong delta
VN-SL 002 250,000 1989 Soil Map of Mekong delta (raster)
NEA 30 250,000 1997 Soil Map of north -west area of Vietnam
NEA 31 250,000 1994 Soil Map of Mekong river delta
NEA 32 250,000 1990 Soil Map of Centro-highland Tay nguyen
NEA 33 250,000 1994 Soil Map of Centro-coastal area of Vietnam
NEA 34 250,000 1994 Soil Map of South-East area of Vietnam
NEA 25 1,000,000 1996 Soil Map of Vietnam - FAO/UNESCO classifi.
NEA 26 250,000 1994 Soil Map of Vietnam - former Soviet Union class.
NEA 27 250,000 1997 Soil Map,NW area of Vietnam - FAO/UNESCO
NEA 29 250,000 1994 Soil Map of Red river delta
Vegetative Cover
Land Cover Unit VIE 0005 1,000,000 85-86 Land Cover map of Vietnam 85-86
VIE 0006 1,000,000 92-93 Land Cover map of Vietnam 92-93
NEA 3 250,000 1997 Topography map of Vietnam
NEA 5 500,000 1997 Database of forestry cover of Vietnam (1943)
NEA 6 500,000 1997 Database of forestry cover of Vietnam (1983)
NEA 7 500,000 1997 Database of forestry cover of Vietnam (1995)
NEA 38 250,000 1995 Ecological Map of Red river delta
NEA 40 250,000 1987 Ecological Map of Mekong river delta of Vietnam
Air Quality Meas.
Air Quality Observation VN-MT 001 500,000 1993 Meteo-monitoring Stations ?
Demography
Demographic Unit NEA 4 100,000 1997 Demography map of Vietnam
Climate Zonation
Agro-climatic Zone NEA 50 1,000,000 1997 Climate map of Vietnam (precip and temp)
Appendix 2: Vietnam Available Datasets
Administrative Bdys
Administrative Unit VN-AM 001 250,000 1993 Administrative Boundary in Mekong delta
(raster)
NEA 1 250,000 1997 Administrative map of Vietnam
NEA 2 1,000,000 1997 Administrative map of Vietnam
NEA 3 250,000 1997 Topography map of Vietnam
Management Areas NEA 8 100,000 1997 Database of Protected Areas, NP of Vietnam
Topography
Elevation MRC DTM's ?
VIE 0004 5 minute Derived from ETOPO5
Water Boundary / Body VN-IN 001 250,000 1986 Inundation Map of Mekong delta (raster)
LM DN 005 250,000 1956 Drainage map - Ca Mau
LM-DN 006 250,000 1956 Drainage Map - Vinh Loi
LM-DN 007 250,000 1967 Drainage Map - Cai Nuoc
LM-DN 026 250,000 1954 Drainage Map - Hoai Nhon
LM DN 004 250,000 1955 Drainage Map - Saigon
Land Use
Land Use Unit VN-LU 001 250,000 1993 Landuse map of Mekong delta (raster)
NEA 20 1,000,000 1992 Landuse Map of Vietnam
NEA 21 250,000 1992 Landuse Map of Vietnam
NEA 22 750,000 1992 Landuse Map of Vietnam
NEA 23 1,000,000 1995 Land Map Unit (Land evaluation map) of
Vietnam
NEA 28 250,000 1994 Land Map Unit (Land evaluation map) of
Vietnam
Geology
Geological Unit NEA 9 500,000 1980 Geology map of Vietnam
NEA 10 200,000 1995 Geology map of Vietnam
NEA 11 200,000 1995 Geology map of Northern part of Vietnam
NEA 12 500,000 1980 Hydrogeology map of Vietnam
NEA 24 1,000,000 Geomorphology map of Vietnam
Major Harvesting Act.
Agriculture ******************
Forestry ******************
Mining Location ******************
Fisheries ******************
Water Quality Meas. ******************
Soil Analysis ******************
Appendix 2: Vietnam Available Datasets (cont.)
Entity File Code Scale Date Comments
Infrastructure
Main Road CAM0018 1,000,000 1988 Roads, Railroads map of Cambodia
Railway CAM0018 1,000,000 1988 Roads, Railroads map of Cambodia
ElectricTransmission Line ******************
Pipeline ******************
Dam ******************
Port ******************
Airport ******************
Soil Class
Soil Unit CA-SO 001 500,000 1986 Soil Map of Cambodia
Vegetation Cover
Land Cover Unit CAM 0041 1,000,000 1971 Vegetation Map of Cambodia
CA-VE 001 1,000,000 1971 Vegetation Map of Cambodia (same as
above?)
CA-LU 1010 250,000 1993 Landuse land cover map of Cambodia
CAM 0044 1,000,000 85-86 Land Cover m ap of Cambodia (85-86)
CAM 0045 1,000,000 90-91 Land Cover map of Cambodia (90-91)
Air Quality Meas.
Air Quality Observation ******************
Demography
Demographic Unit ******************
Climate Zonation
Agro-climatic Zone CAM 0017 2,000,000 1968 Climatic Zones map of Cambodia
Administrative Bdys.
Administrative Unit CA-AM 100 500,000 Cambodia Provincial Boundaries
CA-AM 101 500,000 Districts Boundaries
Management Areas ******************
Topography
Elevation MRC DTM's ?
CAM 0007 5 minute Derived from ETOPO5
Water Boundary / Body CAM0019 1,000,000 1988 Rivers, Lakes, Islands map of Cambodia
CAM 0021 2,000,000 1968 Drainage (Flooded area) map of Cambodia
LM-DN 001 250,000 1954 Drainage Map - Phnom Penh
LM-DN 002 250,000 1955 Drainage Map - Prey Veng
LM-DN 003 250,000 1954 Drainage Map - Long Xuyen
LM DN 004 250,000 1955 Drainage Map - Saigon
LM DN 018 250,000 1954 Drainage Map - Battambang
LM DN 019 250,000 1954 Drainage Map - Siemreap
LM DN 020 250,000 1954 Drainage Map - Stung Treng
LM DN 021 250,000 1954 Drainage Map - Veune Sai
LM DN 022 250,000 1954 Drainage Map - Chanthaburi
LM DN 023 250,000 1954 Drainage Map - Pursat
LM DN 024 250,000 1954 Drainage Map - Kratie
LM DN 025 250,000 1954 Drainage Map - Sre Khtum
CA-IN 001 500,000 1982 Inundation Map of Cambodia
CA-WB 100 250,000 1993 Drainage and open water bodies of Cambodia
Appendix 3: Cambodia Available Datasets
Land Use
Land Use Unit CAM 0038 2,000,000 88-89 Reconnaissance Landuse Map of Cambodia
CA-LU 100 250,000 1991 Landuse Map of Cambodia
CA-LU 001 250,000 Landuse map of Tonle Sap area
CA-LU 002 250,000 Landuse map of Sambor area
CA-LU 003 250,000 Landuse map of Stung Treng area
CA-LU 101 250,000 1993 Landuse land cover map of Cambodia
Geology
Geological Unit LM-GL 100 1,000,000 1988 Geology of Cambodia, Laos, Vietnam
Major Harvesting Act.
Agriculture ******************
Forestry ******************
Mining Location ******************
Fisheries ******************
Water Quality Meas. ******************
Soil Analysis Samples ******************
Appendix 3: Cambodia Available Datasets (cont.)
Entity File Code Scale Date Comments
Infrastructure
Main Road LAO 0015 1,000,000 1988 Roads, Railroads map of Laos
Railway LAO 0015 1,000,000 1988 Roads, Railroads map of Laos
ElectricTransmission Line ******************
Pipeline ******************
Dam ******************
Port ******************
Airport ******************
Soil Class
Soil Unit ******************
Vegetation Cover
Land Cover Unit LAO 0020 1,000,000 92-93 Land Cover map of Laos
Air Quality Measurements
Air Quality Observation ******************
Demography
Demographic Unit ******************
Climate Zonation
Agro-climatic Zone LAO 0014 2,000,000 1968 Climatic Zones map of Laos
Administrative Bdys.
Administrative Unit LAO 0017 2,000,000 Provincial Map of Laos
Management Areas ******************
Topography
Elevation MRC DTM's ?
LAO 0004 5 minute Derived from ETOPO5
Water Boundary / Body LAO 0016 1,000,000 1988 River, Lakes, Islands map of Laos
LM-DN 046 250,000 1955 Drainage Map - Si Mao
LM-DN 049 250,000 1955 Drainage Map - Luang Nam Tha
LM-DN 050 250,000 1954 Drainage Map - Lai Chau
LM-DN 051 250,000 1954 Drainage Map - Dien Bien Phu
LM-DN 052 250,000 1954 Drainage Map - Muong Ngoi
LM-DN 013 250,000 1954 Drainage Map - Ben Giang
LM-DN 017 250,000 1962 Drainage Map - Muong May
LM-DN 029 250,000 1955 Drainage Map - Xaignabouri
LM-DN 031 250,000 1955 Drainage Map - Muong Nan
LM-DN 034 250,000 1954 Drainage Map - Luang Prabang
LM-DN 035 250,000 1954 Drainage Map - Cua Rao
LM-DN 036 250,000 1962 Drainage Map - Vang Vieng
LM-DN 037 250,000 1962 Drainage Map - Khamkeut
LM-DN 038 250,000 1954 Drainage Map - Vinh
LM-DN 040 250,000 1955 Drainage Map - Thakek
LM-DN 041 Drainage Map - Ban Don
Land Use
Land Use Unit LAO 0019 1,000,000 88-872-39 Landuse map of Laos
Appendix 4: Lao Available Datasets
Geology
Geological Unit LM-GL 100 1,000,000 1988 Geology of Cambodia, Laos, Vietnam
Major Harvesting
Activities
Agriculture ******************
Forestry ******************
Mining Location ******************
Fisheries ******************
Water Quality Meas. ******************
Soil Analysis Samples ******************
Appendix 4: Lao Available Datasets (cont.)
United States National Map Accuracy Standards
1. Horizontal accuracy. For maps on publication scales larger than 1:20,000, not more
than 10 percent of the points tested shall be in error by more than 1/30 inch, measured
on the publication scale