کاشی سنتی رنجبران : Spatial Data in the SADC Region in support of Early Warning for Food Security

مقاله ها

صفحه اصلی
مقاله ها
Spatial Data in the SADC Region in support of Early Warning for Food Security

Spatial Data in the SADC Region in support of Early Warning for Food Security

1402/09/29

Spatial Data in the SADC Region in support of
Early Warning for Food Security and Natural Resources Management

C.A.J. van der Harten and K. Masamvu
SADC Regional Food Security Programme
Regional Remote Sensing Unit
Harare - Zimbabwe
P.O. Box 4046 Harare
Zimbabwe
cvanderharten@fanr-sadc.co.zw and kmasamvu@fanr-sadc.co.zw

Abstract - Agriculture is by far the most important activity in most countries of the 14-nation Southern African Development Community (SADC). This is being recognised by SADC which has adopted a Programme of Action covering co-operation in various sectors including that of Food Agriculture and Natural Resources. Zimbabwe is responsible for the co-ordination of this programme which includes the implementation of the SADC Regional Food Security Programme. Under this programme the use of remote sensing and GIS technology is an important part of the environmental monitoring activities which are being co-ordinated by the Regional Remote Sensing Unit. In order to support the analysis of remote sensing products the Unit started in 1994 the development of an unique spatial data base with information on national and sub-national boundaries elevation infrastructure hydrology major growing areas forests protected areas and cultural sites. This data base also includes a variety of satellite climate and agriculture information in raster format. While the main focus of the RRSU activities and data bases is on early warning for food security it has become clear that the activities data bases and outputs are also very beneficial for a wide range of environmental monitoring and spatial data activities. During the last years the RRSU has been increasingly recognised as the authority and one of the major spatial data sources in Southern Africa and consequently the RRSU is involved in a number of initiatives to further harmonise and improve spatial data infrastructures in the SADC region.

I.     INTRODUCTION

In order to enhance food security for all in the 14-nation• Southern African Development Community (SADC) SADC established a Food Security Programme which is being managed by the Food Agriculture and Natural Resources (FANR) Development Unit (DU) which is based in Harare - Zimbabwe. The central part of the programme is the Regional Early Warning Unit (REWU) which is being assisted by the Regional Remote Sensing Unit (RRSU). These two Units are the core of the Regional Early Warning System (REWS). In each of the SADC member states there is a National Early Warning System which is supported by a National Early Warning Unit.

This paper summarises the spatial data base activities of the RRSU in support of early warning for food security as well as the increasing involvement of the RRSU in many other environmental monitoring activities in Southern Africa.

II.     THE RRSU

The RRSU is a centre of technical expertise which can facilitate training programmes and technical support in the field of remote sensing and GIS in support of early warning for food security and natural resources management.   The RRSU started as a project in 1988 and after 10 years of financial assistance from the Governments of Japan and the Netherlands through the Food and Agriculture Organisation (FAO) the project was gradually integrated as a Unit in the organisation structure of the FANR DU. The RRSU is operated with funding from the SADC Member States and it receives additional financial and technical support through a bilateral agreement between the Government of the Netherlands and SADC.

This paper will mainly focus on the activities implemented by the RRSU to develop a spatial vector and raster data base in support of early warning for food security which is now also being recognised as one the major standard datas in the SADC region by many other environmental organisations in Southern Africa.

III.     OPERATIONAL ACTIVITIES

A.   Satellite data bulletins and training

The RRSU facilitates training programmes and technical support in the field of remote sensing and GIS in support of early warning for food security and natural resources management. On an operational basis the RRSU is using low resolution high temporal information from the Meteosat and NOAA satellites to produce information products on rainfall occurrence and vegetation development which is being distributed through the Regional and National Early Warning Units (NEWU’s) but also through its own publications reports and web-site. The Meteosat receiver is located at the Zimbabwe National Meteorological Department and is operated by the RRSU. The NOAA data is received through NASA-USAID/FEWS in Washington.

A variety of training courses and national and regional workshops are organised to create a core of trained experts in the SADC region. Country visits for technical backstopping are undertaken regularly to assist the national contact points in their own environment and solve problems on the spot.

Other RRSU activities include the development of better crop foreing methodologies improved rainfall estimates better – high resolution – remote sensing products and an Agricultural Potential Information System.

B.   Data base development

An important activity of the RRSU is the development of spatial data bases which are being distributed on CD. The RRSU data base include at present all the basic thematic information as well as the satellite image archive agricultural statistics soils digital elevation and climate information. In order to develop these information systems further the RRSU has strategic partnerships with a number of institutes in the SADC region but also in Europe and the USA. The RRSU spatial data base is recognised as a regional (and often a national) standard and because of this the RRSU is a recognised partner in a number of EIS related activities in the SADC region. At a regional level the RRSU collaborates with the South African National Spatial Information Framework (NSIF) on the development of a meta-data base which will have a regional outlook.

IV. SDI related activities

A.   Reason for SDI development

From the time of its establishment the RRSU has been working on the use of satellite information to monitor rainfall occurrence and vegetation development in support of early warning for food security. The satellite data covers the whole SADC region and the operational pixel size of the raster images is approximately 8 km. With the increased use of GIS technology and the availability of ever faster computers and more user-friendly GIS software programs there was a need to harmonise and standardise spatial datas not only the raster satellite images but also the vector data base.
In the early nineties most digital spatial data available in the SADC countries originated from small projects. Spatial data available from the Surveyor General Offices was often not in digital format or in an inaccessible digital format. As a result many Government offices small projects universities NGOs started to digitise their own spatial data bases.

Since one of the tasks of the RRSU is to introduce GIS technology in the SADC region the main problem it faced was the lack of a consistent spatial data base for the SADC region. For example national and sub-national administrative boundaries did hardly exist in digital format or were incomplete. For the data which did exist there was no cross-boundary compatibility. Other data on infrastructure basic land use hydrology did not exist or was scarce. A soil map had been prepared for a number of SADC countries but the digital format used made it impossible to use the data for further GIS analyses. The satellite images in raster format from the Meteosat satellite (for climate monitoring) and NOAA satellite (vegetation monitoring) were in a rare geographic projection the Hammer Aitoff projection which was hardly supported by any of the at that time more popular GIS software programs.

The task at hand for the RRSU was to start a number of activities to develop standards for the digital data bases and the objective was to develop a standard raster and vector data base for the SADC region which would allow easy use and analytical procedures in a GIS environment and facilitate regulars.

B.   Who were the principle organisers

The SADC RRSU took in 1994/1995 the lead in this development. The Office of Arid Land Studies of the University of Arizona ( Tucson AZ - USA) was contracted to assist the RRSU in developing the initial vector data. As a starting point several layers of the Digital Chart of the World (DCW) were used as well as the Africa Data Sampler (ADS) prepared by the World Resources Institute (WRI Washington - USA). The WRI provided the RRSU with a pre-release of the ADS in 1994 in order to facilitate a first review of the available data. At a later stage the University of Stellenbosch was contracted to review and correct the soil data base.

All satellite image raster data was converted by the RRSU into a 6-minute geographic projection. The new historical data base was distributed to all contact points in the SADC Member States. Backstopping missions and regional workshops were used to inform the contact points about the changes and the characteristics of the new data format.

Since 1994 the development has gone through several phases and has resulted in a uniform and standard satellite (Meteosat and NOAA) image data base; a standard and uniform thematic vector data base at the scale of 1:1million; and more recently in a regional climate data base in raster and tabular format.

All RRSU data is made available on CD-ROM the RRSU CD. At present Version 1.0 is available and will bed in March/April 2000. Included on the CD is also a program to view and analyse the vector and raster data. This program WinDisp has been developed with funding from the FAO USAID/FEWS SADC US Forest Service and US Geological Service. WinDisp is freeware software is regularlyd and is one of the most widely used programs in the early warning for food security community.

C.   What are the major components

All raster data is in the geographic projection at a resolution of 6-minutes (approximately 8 km.). All data is in 8-bit WinDisp format. Vector data is in the geographic projection at a 1:1million resolution. All data is in Arc/Info format but is being distributed in the Strategic Mapping ASCII BNA format.

Vector data is available for the whole SADC region. Also raster data covers most of the SADC region with the exception of the DR Congo Mauritius and the Seychelles for which no data is available at present.

Raster information/data:

•    Meteosat Cold Cloud Duration (CCD) archive (September 1988 to present - 10-daily intervals).
•    NOAA Normalised Difference Vegetation Index (NDVI) GAC archive (1981 to present - 10-daily intervals).
•    NOAA Normalised Difference Vegetation Index (NDVI) LAC archive (1996 to present - monthly intervals) Note: processed at 2-minute resolution.
•    Digital Elevation Model (Source: USGS Eros Data Center) Note: available in standard raster format - but can made available at a higher resolution.
•    Rainfall PET Temperatures Windspeed Radiation and Relative Humidity (average of 1987 to 1995).
•    FAO Rainfall and PET (average of 30 years).
•    SADC Climate Data base and Data Layers: Rainfall and Temperatures (1961 to present - 10 daily intervals). Note: also in tabular format.
•    SADC Climate Data base and Data Layers: PET (1987 to 1995 - 10 daily intervals)

Vector information/data

•    National and sub-national administrative boundaries.
•    Elevation.
•    Infrastructure.
•    Hydrology (further developed by the FAO ALCOM Project based in Harare - Zimbabwe).
•    Major growing areas.
•    Forests.
•    Protected areas.
•    Cultural sites.
•    Soils (only for Angola Botswana Malawi Mozambique South Africa Tanzania Zambia and Zimbabwe)

D.   Organisational approach

Leadership and stakeholders

Leadership was taken by the RRSU. Major stakeholders in the development were the University of Arizona and the University of Stellenbosch. The NEWU’s and National Meteorological Departments (NMD’s) in the SADC countries played an important role in the evaluation of the datas and provide suggestions for corrections or better data. Other stakeholders were the international institutions who contributed data to the initiative i.e. World Resources Institute; USGS Eros Data Center; FAO; UNEP GRID; and the USAID Famine and Early Warning System (FEWS). At regional or national level data was provided by: the NEWU’s; NMD’s; National Remote Sensing Centres; Environmental Councils; and various Government Departments.

The responsibilities and role of the stakeholders in the organisational structure can be summarised as follows:

•    Initiative and leadership

Taken by:    the SADC RRSU.
Role(s):    (i) identify needs and formulate plan; (ii) implement development with strategic partners; (iii) assess availability of data; (iv) organise data collection; (v) ensure evaluation and quality control of the outputs; and (vi) distribute the output.

•    Development

Done by:    the SADC RRSU; the University of Arizona; and the University of Stellenbosch. Those who made data available should also be mentioned (see above).
Role(s):    (i) process data; (ii) create basic data layers; (iii) prepare documentation; and (iv) store data on transportable medium with user-interface to view and analyse the data.
Note:    An Harare based FAO project: ALCOM used the hydrology layers to develop a comprehensive hydrological data base which is fully compatible with the RRSU spatial data bases.

•    Review and Evaluation

Done by:    the SADC RRSU; the National Early Warning Units (NEWU) and National Meteorological Departments (NMD) in the SADC countries.
Role(s):    (i) make data available for evaluation; (ii) workshops/meetings to introduce the data bases; (iii) collect evaluation comments/reports; and (iv) ensure incorporation of corrections/additions.

•    Distribution and use

Done by:    the SADC RRSU; intended users: the National Early Warning Units (NEWU) and National Meteorological Departments (NMD) in the SADC countries. A range of government institutes; Ministries; national regional and international organisations; private trading and industrial sector; banking and finance groups; large-scale and small-scale farming organisations; and NGOs.
Role(s):    (i) make data available in a user-friendly format; (ii) create awareness through workshops/meetings and the Internet; (iii) be open for suggestions.

Relationship and unresolved issues

The RRSU used contractual agreements with the University of Arizona and the University of Stellenbosch. All other contacts were basically informal. Data was provided as part of a mutual agreement with the understanding that the RRSU would correct and datas and return it in the new format to the originator/provider of the data.

An unresolved issue is the use of climate data. In the SADC region the distribution of climate data is the responsibility of the NMD’s. Since the NMDs are SADC institutes they have made data available to the RRSU in order to develop a regional tabular data base and create climate (raster) layers to be used for analytical purposes and research. At present the RRSU is not in a position to distribute these tabular datas or climate layers without prior approval from the NMD’s.

E.    Implementation approach

A.   Listing of events

Activities started in 1994. The DCW was used as a starting point and wasd through a merger with regional and national data bases. All data was georeferenced and stored in a uniform and common standard format. The events can be listed as follows:

September - December 1994:
•    Assess the need for uniform data standards for the SADC region.
•    Identify possible partners in the development.
•    Prepare contracts for the collaborator in the study.
January - October 1995:
•    Collect data for the SADC Vector Data base.
•    Develop the data base (implemented by the University of Arizona).
•    Review/evaluate work.
•    Store data base on transportable medium for evaluation.
May - November 1995:
•    Review and correction of the SADC Soils data base (implemented by the University of Stellenbosch).
July - August 1995:
•    Convert all raster satellite image data to the geographic projection and standard resolution.
September 1995:
•    Introduce new data standards for the raster data during Annual Regional Workshop.
January - December 1996:
•    Distribute data in the region (often in the format of country datas).
•    Encourage evaluation of the datas.
•    Collect evaluation result and document this.
June - December 1996:
•    Transfer of IDA analytical functions to the WinDisp software (financed by the RRSU and implemented by the University of Arizona - Eric Pfirman).
January - May 1997:
•    Implement changes to the SADC Vector Data.
•    Develop a user-friendly interface in WinDisp.
•    Develop file tree structure.
•    Develop/review file names.
•    Prepare master RRSU CD Version 0.9 (May 1997).
•    Duplicate of CD to each SADC Member State for evaluation and comments.
•    Announce RRSU CD by email and through the SADC FANR Website
June - July 1997:
•    Collect evaluation results and comments.
•    Prepare master RRSU CD Version 0.9a (June 1997)
•    Duplicate of CD and distribute.
•    Introduce RRSU CD during the RRSU GIS Workshop in Stellenbosch - South Africa.
January - March 1998:
•    RRSU CD Version 1.0 (March 1998)
April 1998 to present:
•    Establish GIS technical facility at the RRSU (October 1998).
•    Distribution of the CD on request.
•    Collect new datas and incorporate those in the data.
•    Start development of Meta-data base in collaboration with NSIF in South Africa.
March/April 2000:
•    Release of RRSU CD Version 2.0

B.    Data policies - ownership

Rather informal contacts have contributed to a process of awareness in which different organisations government institutes and other interested parties have made available data for this regional data initiative. The RRSU has taken every opportunity to advertise the need for a uniform regional data base.

All data produced and maintained by the RRSU is freely available. The RRSU has introduced a small fee – on cost recovery basis – for the distribution of the RRSU CD. The CD is provided at no charge to SADC institutions or government departments.

V.   Conclusions

Agriculture is by far the most important activity in most countries of SADC. This is being recognised by SADC which promotes regional cooperation and economic development and has adopted a Program of Action covering cooperation in various sectors including that of food agriculture and natural resources. Food security and natural resources management is one of the main pillars for economic development and social welfare in the region.

A solid harmonised and uniform regional spatial data base contributes to an improved information in support of managing scarce resources which are required to secure food security and human well-being in the region.

The spatial vector and raster data base developed and maintained by the RRSU is a good example of collaboration between different countries in terms of data exchange and sharing which has resulted in a uniform spatial data which is not onlyting a standard at regional level but in some cases also at national level.

REFERENCES

Harten C.A.J. van der and K. Masamvu (2000) Use of Remote Sensing and GIS in support of Early Warning for Food Security. In: Proceedings of the 28th International Symposium on Remote Sensing of Environment.
RRSU (2000) The RRSU Data base. RRSU Working Paper No. 6. SADC RRSU Harare - Zimbabwe
SADC “A Food Security Strategy Framework” (1997) SADC FANR DU Harare – Zimbabwe.

Harare – Zimbabwe
March 2000

درباره ما

روابط عمومی

آدرس