Algeria - Pilot Forestry and Watershed Management Project

a - /Q/ 3/S '7-X+-A. . Doment of The World Bank FOR OFFICIAL USE ONLY MICROFICHE COPY Report No. 9390-AL Report No. 9390-AL Type: (SAR) KRAFFT, NI/ X32626 / H-4127/ EM2AG STAFF APPRAISAL REPORT DEMOCRATIC AND POPULAR REPUBLIC OF ALGERIA PILOT FORESTRY AND WATERSHED MANAGEMENT PROJECT MAY 18, 1992 Agriculture Operations Division Country Department I Middle East and North Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EOUIVALENTS (March IM92) Currency Unit - Algerian Dinar US$1.00 - DA 22.50 DA 1.00 - US$ 0.044 WEIGHTS AND MEASURES The metric system is used throughout this report FISCAL YEAR January 1 to December 31 GLOSSARY OF ABBREVIATIONS ANF National Forestry Agency ANN National Agency for Nature Conservation APC Popular Communal Assembly BNEDER National Office for Rural Development Studies BNEF National Office for Forest Studies CNP National Planning Council EDIMCO Enterprise for the Distribution of Construction Materials EMIFOR Local-Level Forest Enterprise ENAB National Enterprise for Wood Supply ENATB National Enterprise for Wood Processing and Furniture ENMGP National Enterprise for Joinery and Prefabrication EPA Autonomous Administrative Agency EPE Economic Public Enterprise EPIC Industrial and Commercial Public Agency GEF Global Environment Fund INRF National Institute for Forestry *esearch METAP Mediterranean Environmental Technical Assistance Program OAMV Office for the rPvelopment of the Mountainous Zones ONTF National Office or Forest Works ORDF Regional Forest Development Corporation PMT Project Management Team PTT Post and Telecommunications SNTA National Tobacco and Match Company SNTF National Railway Company UNDP United Nations Development Program  FOR OMCIuL USE ONLY STAFF APPRAISAL-REPORT DEMOCRATIC AND POPULAR REPUBLIC OF ALGERIA PILOT FORESTRY AND WATERSHED MAUNAGEMENT PROJECT TABLE OF CONTENTS Loan and Project Summary . . . . . . . . . . . . . . . . . . . i I. INTRODUJCTION .1... . . . . . . . . . . . . . . . . . . . . . . II. FORESTRY AND WATERSHED MANAGEMENT IN ALGERIA . . . . . . . . . . 1 A. Resources . . . . . . . . . . . . . . . . . . . . . . . . .1 B. Contribution to the Economy ... . . . . . . . . . . . . . 4 C. Forestry Institutions ... . . . . . . . . . . . . . . . . 5 D. Government Plans and Policies . . . . . . . . . . . . . . . 7 E. Forest and Environment Legislation . . . . . . . . . . . . . 8 F. Forest Operations .... . . . . . . . . . . . . . . . . . 9 G. Watershed Operations ... . . . . . . . . . . . . . . . . . 14 H. Nature Conservation . . . . . . . . . . . . . . . . . . . . 17 III. BANK INVOLVEMENT IN AGRICULTURE IN ALGERIA . . . . . . . . . . . 18 IV. THE PROJECT AREA. . . . . . . . . . . . . . . . . . . . . . . 20 V. THE PROJECT. . . . . . . . . . . . . . . . . . . . . . . . 21 A. Objectives, Rationale, and Main Features . . . . . . . . . . 21 B. Detailed Description .23 C. Project Costs . . . . . . . . . . . . . . . . . . . . . . 32 D. Financing .33 E. Procurement . . . . . . . . . . . . . . . . . . . . . . . 34 F. Disbursements .... . . . . . . . . . . . . . . . . . . . 35 c. Accounts and Audits ... . . . . . . . . . . . . . . . . . 37 VI. PROJECT IMPLEMENTATION .... . . . . ..... . . . . . . . . 37 A. Organization and Management ... . . . . . . . . . . . . . 37 B. Implementation .... . . . . . . . . . . . . . . . . . . . 38 C. Participation of Local Populations ... . . . . . . . . . . 38 D. Training and Consultancy Services . . . . . . . . . . . . . 39 E. Monitoring and Evaluation ... . . . . . . . . . . . . . . 40 F. Status of Project Preparation ... . . . . . . . . . . . . 41 This report Is based on the findings of an appraisal mission that vlsited Algeria in Nev ber-De _ber 1999 comprised of Messrs. N. Krafft (Mission Leader), J. Luis (Land Us. Speciallet), and M. I. Schizmr (Economiat) - Bank; and ossrs. D. Lauras (Forest Agronomist) and J. Mrian (Forest R-earchb Specialist) - Consultants. The toehical aspects of the project ware essentialy appraied in May 1990 by a mission coqWised of Masrt. B. Krafft (Mission Leader). J. Lewis (Land Use Specalist), C. olloway (aturae Consaervation) and l. I. SchirAer (Economist) - Dank; and Mecsrs. D. Leurans (Forest Agronoaist) and D. Lapaze (Wood Industries Specialist) - Consultants. A mission In March 1092 of H. Stier (Task Manager) and Cb. Lauara (Economist, PAD/CP) made the final adjustments to the project. This document has a restricted distribution and may be used by recipients only in the performance I of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Pae Table of Contents (cont'd). VII. BENEFITS A JUSTIFICATION .... . . ...... . . . . . . . . 41 A. General .... . . . . . . . . . . . . . . . . . . . . . . 41 B. Yields and Production .... . . . . . . . . . . . . . . . 42 C. Markets and Prices .... . . . . . . . . . . . . . . . . . 43 D. Benefits .... . . . . . . . . . . . . . . . . . . . . . 43 E. Economic Analysis and Sensitivity Analysis . . . . . . . . . 45 F. Project Risks .46 G. Environmental Impact Statement. 47 VIII. AGREEMENTS AND RECOMMENDATION .... . . . . . . . . . . . . . . 48 LIST OF ANN2XES ANNEX 1 - SUMMARY COST ESTIMATES Table 1: Project Cost Summary Table 2: Project Components by Year Table 3: Summary Accounts by Year Table 4: Estimated Schedule of Disbursement of Bank Loan ANNEX 2 - ASSURANCES RELATED TO THE IMPLEMENTATION OF THE PILOT WATERSHED MANAGEMENT COMPONENT ANNEX 3 - PROPOSED BANK SUPERVISION INPUT INTO KEY ACTIVITIES ANNEX 4 - SELECTED DOCUMENTS AVAILABLE IN PROJECT FILE ANNEX 5 - FINANCIAL AND ECONOMIC PRICES OF SELECTED OUTPUTS AND INPUTS CHART - Organization Chart MAP - Project Sites - IBRD 22755R  STAFF APPRAISAL REPORT DEMOCRATIC AND POPULAR REPUBLIC OF ANSERI PILOT FORESTRY AND WATERSHED HANAGENENT PROJECT Loan and Proiect Summary Borrover: Democratic and Popular Republic of Algeria Amount: US$25 million equivalent Terms: 17 years, including five years' grace, at the standard variable interest rate Executing Aeencies: National Forestry Agency (ANF) and National institute for Forestry Research (INRF) Project DescriDtion: The objective of the pilot project is to develop a long- term action program to better protect and manage Algeria's natural resources, specifically its forests, watersheds, and nature conservation areas. The project would st .engthen the institutions working in these areas and, through appropriate testing, would develop the policy and technical basis for replication of suitable project components as part of a national resource management program. The five-year project would comprise: (a) studies to: (i) identify how much of Algeria's wood needs could be met domestically, (ii) simplify and improve forestry planning and test ways to integrate forest activities with those of livestock owners, and (iii) provide a master plan for major watersheds; (b) pilot reforestation on 4,300 ha and pilot exploitation operations to introduce new, low- cost techniques; (c) watershed management operations to introduce new, low-cost approaches to increase farmers'. incomes and to reduce erosion on some 10,000 ha of fragile and eroded land; (d) nature conservation with emphasis on the strengthening of the applied research capacity in this area and on tra.ining and the supply of equipment; and (e) support activities including related applied forest research, institution strengthening, vehicle and equipment purchases, buildings, training and technical assistance.  eii - Benefits and Risks: The main benefit of the project would be the long-term improved and more efficient use of Algeria's natural resources. The project would develop a viable forest sector so that, in addition to environmental benefits, it would increase wood production, thereby saving foreign exchange. In the watersheds, employment and incomes of the poorest sections of the population would be increased, while erosion and sedimentation of dams would be reduced, thereby helping to alleviate Algeria's acute water shortage. Finally, the project would help ensure the preservation of the country's diverse flora and fauna. The main risks relate to the undertaking of a project that depends partly for its success on the presence of a decentralized and more market-based system, when Algeria's economic system is itself undergoing such profound reform. Fortunately, the Government has shown strong commitment to its overall program of reform. It is likely, therefore, that the economic environment in which the project would be implemented would be conducive to success. The secondary risks are project related, such as the quality of monitoring and evaluation, studies, and technical assistance and training. The project has been designed to address these issues.  - iii - Estimated Proiect Costs:A' Local Foreign Total ------- (US$ mill ion)-------------- A. Studies and Management Plans 0.7 2.2 2.9 B. Pilot Forest Operations 7.6 7.9 15.5 Reforestation (6.0) (5.9) (11.9) Exploitation (0.3) (0.6) (0.9) Applied Research (1.3) (1.4) (2.7) C. Pilot Watershed Development 5.0 3.5 8.5 D. Nature Conservation 0.1 0.3 0.4 E. Support Activities 1.6 2.0 3,6 Base Cost 15.0 15.9 30.9 Physical Contingencies 1.4 1.6 3.0 Price Contingencies 2.0 _15 3.5 TOTAL PROJECT COST 18.4 Financing Plan: Local Forei Total - --------------(US$ million)--------------- Government 11.6 - 11.6 Bank 6.0 19.0 25.0 Farmers 0.8 - 0.8 TOTAL PROJECT COST 18.4 37.4 Estimated Disbursements (USS million) FY93 FY94 FY95 FY96 FY97 FY98 FY99 Annual 1 2 3 5 5 5 4 Cumulative 1 3 6 11 16 21 25 Economic Rate of Return: 12X MaR No.: IBRD 22755R a/ Includes US$6.8 million of duties and taxes.  STAFF APPRAISAL REPORT DEMOCRATIC AND POPULAR REPUBLIC OF ALGERIA PILOT FORESTRY AND WATERSHED MANAGEMENT PROJECT I. INTRODUCTION 1.01 Algeria is faced with serious problems of erosion and land degradation that, in the long run, will undermine its productive potential and will exacerbate the problem of water scarcity. Rapid population growth over the past few decades, accompanied by encroachment and forest fires, has led to a steady decline in the area under forests, a subsequent increase in shrubs (maquis), and an expansion of agriculture in marginal areas, both in the mountains and the arid areas. Action must be taken to prevent a further decline in the area and quality of Algeria's forests -- which could make a much more significant contribution to the economy, saving scarce foreign exchange used for import of wood products. In the watersheds, the combination of cultivation and overgrazing has resulted in a loss of vegetative cover and consequent intense erosion. Not surprisingly, productivity in the upper catchment has started to decline, sedimentation is already significantly shortening the life of dams, and scarce water is wasted as runoff instead of being used to recharge the aquifer. 1.02 In the late 1980s, Government and the Bank started discussions on ways to address problems in the sector. The proposed project, which follows from these discussions, was identified by a joint Bank-FAO/CP mission and was prepared by FAO/CP in a report dated March 1990. The project was pre-appraised by the Bank in May 1990. Appraisal took place in November-December 1990, but a major reorganization in the sector caused delays; a Bank mission in February-March 1992 made the final adjustments to the pilot project. II. FORESTRY AND WATERSHED MANAGEMENT IN ALGERIA A. Resources 2.01 After Sudan, Algeria is the largest country in Africa and in the Arab-Moslem world. It covers some 2.4 million kmg, divided into three zones roughly parallel to the Mediterranean coast. Each zone is separated by mountain ranges: (a) a northern coastal strip, with relatively good rainfall, divided by the Tellian Mountains from: (b) a relatively mountainous, pre-desert steppe (these two zones cover 15Z of the country and incorporate nearly all the forest areas); and (c) south of the Sahara Atlas Mountains, the Sahara Desert covering slightly over 2 million kin, with an extremely harsh hyper-arid climate.  -2- 2.02 Forests: According to official estimates, gazetted forests cover some 3.7 mil?ion ha, 1.5X of the land area -- about the same as the amount of land cultivated in any year. Of the land under such forests, about half is considered to be relatively high-quality forest, while the remainder is low-grade forest or maquis. If the arid Sahara is excluded, forests cover about 15X of the northern part of the country. An additional 15-20 million ha are suitable for grazing, esparto scrub, and shrubs. 2.03 Table 2.1 shows the forest resource based on a 1984 inventory; this excludes parks and reserves (para. 2.06). Given the incidence of fire and encroachment, the area under forests is probably less than indicated in official statistics. The table shows the volume of wood that could tivioretically (biologically) be exploited each year on a renewable basis. rtuch of the timber is, however, in small and scattered stands and could not be exploited commercially. The amount that could be logged commercially is not known. Table 2.1: FOREST RESOURCES - AREA, VOLUME, AND POTENTIAL FOR EXPLOITATION Biological Average Actual Ha Standing Potential for Harvest as X of ('000) Volume Exploitation Biological (million m3) ('OOOm3/yr) ('OOOm3) Potential Productive Forest Aleppo Pine 880 30 850 145 17 Cork, Zeen Oak, and Cedar 300 22 250 22 8 Fast Growing Eucalyptus and Pine 75 3 170 48.A8 Subtotal 1,255 55 1,270 215 18 Protective Forest Holm Oak, Juniper, 220 and Thuya Miscellaneous Forest Miscellaneous 320 Maquis 1.875 Total 3,670 2.04 Rainfall generally increases from south to north and from west to east. The best growing conditions for trees are found in the northeast of tha country, where rainfall is the highest (at up to 1,200 mm). It is here that the cork, zeen oak, and cedar forests are found. As elsewhere in Algeria, much of this forest is degraded, and in some areas regeneration is severely hampered by the pressure from livestock. The fast-growing eucalyptus and pine forests are  -3. also found in the northeast of the country. The most important wood resource in the country is the Aleppo pine which constitute about half the "productive" forest area. Most of the stands are scattered throughout the central mountains, where rainfall averages 600-900 mm. In general, they are intermixed with holm oak, thuya, and juniper. In the semi-arid areas there are a variety of species suitable only for protective and environmental purposes. Some forest lands are so degraded that they are little more than scrub (known as maquis). 2.05 The three wilayate (provinces) of Skikda, Annaba, and El Tarf in the northeast of AlgetUa, bordering Tunisia and the Mediterranean coast, contain 97% of the productive pines, 91% of the cork oak, and 45% of the eucalyptus. 2.06 Reserves. Parks. and Wetlands: In addition to national forests, Algeria has put aside large areas for conservation purposes. There are four nature reserves, with a total area of 36,800 ha; seven national parks, covering 37,000 ha; four game reserves, with 50,000 ha; and two parks in the Sahara, covering 560,000 km't1. The fact that Algeria has set aside nearly a quarter of its land area for parks and reserves demonstrates the importance that the country attaches to the preservation of its natural resources. 2.07 Wetlands in Algeria are a varied group of fresh, brackish or saline, seasonal or permanent lakes and marshes. Many have been lost through drainage for agriculture, but slightly more than 1 million ha may still be considered of international interest. They occur largely, but by no means entirely, within the 250 km strip from the coast to the Saharan Atlas Mountains. They are of great ecological value because of their varied fauna and flora and, more particularly, because they provide a vital habitat for large numbers of migratory and overwintering wildfowl. The most important wetland in the country, which is included in the protected areas system, is the El Kala National Park (a wetland complex of 76,436 ha of saline and freshwater lakes and marshes). Also included as a protected area is the La Macta Nature Reserve, with 20,000 ha of sea-level marsh. 2.08 Watersheds: As noted above, because of population pressure, more and more marginal land is being brought into cultivation each year. At the same time, livestock owners are grazing their animals on land that, only a few decades ago, was either under forest or had gcod vegetative cover. As a result, some 12 million ha are threatened with erosion, much of which is in the ecologically fragile steppe area. It is estimated that the equivalent of 40,000 ha of Algeria's scarce agricultural land is lost each year because of erosion and desertification. Part of this erosion is reducing the storage capacity of dams through sedimentation. 2.09 Given that Algeria is facing a very serious water shortage and has extremely limited storage capacity for water, it is important that this storage capacity be preserved. It is also important that as much water as possible be stored as groundwater. The lack of vegetative cover, however, means that For the distinction between parks and reserves, see paras. 2.62-2.63.  - 4 - rainfall runoff rates are generally high and that water is wasted instead of recharging the aquifer. Unless rigorous actions are started soon to improve natural resource management in the watershed areas, Algeria's productive potential will be seriously undermined. This will be reflected partly through declines in agricultural productivity and partly through water shortages; both would have major economic and social consequences. B. Contribution to the Economy 2.10 The most important economic contribution of forests to eate has probably been ecological. Certainly, the quantifiable benefits have not been commensurate with the level of Government expenditures in the sector. Although the annual harvest of wood products more than tripled, from about 45,000 m3 in 1978 to some 200,000 m3 over the past two years, this exploitation represents only a small p.. 3rtion of the wood used; the balance is imported. Imports reached a peak of 1,350,000 m3 in 1983 but are currently running at around 1 million m3 because of a shortage of foreign exchange. Algeria imports some US$200-300 million worth of wood products a year, roughly 2.5X of merchandise imports. Total demand for wood products is estimated at around 1.5 million m3, and local production currently meets only 10% of demard. 2.11 As shown in table 2.1, only about 181 of the wood that could be exploited on a sustainable basis is harvested. If it is assumed that 50X of the Aleppo pine, cork and zeen oaks, and cedar and 100X of the fast-growing species can be exploited commercially, then Algeria could today produce over 700,000 m3 of wood annually on a sustainable basis -- nearly half its annual needs. This would result in an annual foreign exchange savings of over US$100 million. Moreover, if Algeria were able to plant the 400,000 ha that have been identified as suitable for replanting to fast-growing species (species with a growth rate of not less than 5 m3/ha/yr), over 2 million m3 of wood could be cut annually, which is considerably above present demand. 2.12 Much of the forest gives the impression of having been abandoned. Apart from the construction of fire breaks, forest roads, and water points -- largely conservation measures (although the water points also serve for livestock, which, if not properly managed, can reduce natural regeneration) -- there is little evidence that the forests are being actively managed. Most forests are slowly deteriorating, and there is relatively little replanting, controlled regeneration, thinning, or exploitation. As a result, livestock owners see the forests as a free resource to be exploited at will; indeed, forests contribute a significant proportion of national forage production (para. 2.47). One of the objectives of the project is to show livestock owners that forests can and should be better managed on a sustainable and economic basis to increase production (including forage production) and employment in the area.  -5 C. Forestry Institutlons General 2.13 The Ministry of Agriculture has overall responsibility for the development of the forestry sector, including watershed management. Recently the Government, in an effort to revitalize the sector, undertook a major reorganization of the institutions involved in the development and management of the country's forests and watersheds. The responsibility for the forestry sector was transferred from the Ministry of Hydraulics to the Ministry of Agriculture, and the different certral directorates responsible for various aspects of the forestry sector, and their local counterparts, were united under the National Forestry Agency (ANF). The National Forestry Agency (ANF) 2.14 The National Forestry Agency was created under an executive decree dated April 21, 1990, as modified and amended by the executive decree of February 23, 1991/. The ANF is an Autonomous Adtinistrative Agency (EPA), with a separate legal identity and financial autonomy, under the supervision of the Ministry of Agriculture. 2.15 The ANF is headed by a Director General, assisted by a Deputy Director General, and is charged with the administration and management of the forestry sector and the execution of Government forest policies. The ANF has a Supervisory Board that is charged with studying and making proposals to the Ministry of Agriculture of any questions in relation to ANF's organization and activities. The ANF board is chaired by the Minister of Agriculture, and members are representatives of the Ministries of Agriculture, Defense, Equipmsnt, and Interior, the High Commission for Research, the National Planning Commission, and the High Commission for the Development of the Steppe. In thls way inter. sectoral coordination is ensured. ANF Board members are appointed for three years. The Board may also invite any person whose participation is judged helpful for the issues under discussion. 2.16 The ANF, at its headquarters in Algiers, is divided into five central directorates responsible for: (a) administration; (b) planning and promotion of productive activities; (c) forest management and development; (d) forest protection; and (e) soil conservation. The ANF is represented in the field through six regional inspections, one for each of the major ecological regions of Alger!a (West, Western Steppe, Center, Eastern Steppe, East, Northeast). Each regional inspection is generally subdivided into three Forest Conservancies, totaling 16 (see the Organization Chart in the annexes). The Conservancies, in turn, are subdivided into a number of circumscriptions. 1/ Decrees No. 90-114 and 91-59, respectively.  - 6 - Regional Forest Development Corporations (ORDFs) 2.17 The new forest organization includes six Regional Forest Development Corporations (ORDFs)l', which overlap with the territories of the regional inspections of ANF. The ORDFs are autonomous Industrial and Commercial Public Agencies (EPICs) under the Ministry of Agriculture and are each headed by a Director General. The ORDFs' Supervisory Board is chaired by the Director General of the ANF and has as members representatives of the Wali (local government official) and of the Ministry of Equipment in the areas where they operate. The ORDFs have been established to carry out forest studies and forest operations at the local level in coordination with the ANF. In practice ANF commissions the ORDFs for such activities. The organization is still evolving, and it is expected that the more commercial activities of the ORDF would be gradually privatized by the creation of subsidiary companies run as Autonomous Public Corporations (EPEs). 2.18 The new organization is not fully in place, and the division of certain tasks between ANF and the ORDFs has yet to be clarified. Three risks can be identified with the new structure: (a) some redundancy of tasks, personnel, and equipment between ANF and the ORDFs if their respective mandates at the local level are not more clearly defined; (b) limited incentives for the development of a private sector as long as the ORDFs remain the main bidders for works; and (c) possible difficulties for multi-sectoral coordination and the participation of the local population in forestry and watershed management, since there is little formal coordination at the wilaya level. However, the structure is sufficiently well-defined to implement the pilot project, which will itself provide an opportunity to test the organizational arrangements. These can be further refined over time. Other Existing Specialized rorestry Institutions 2.19 The Forest Department commissions the National Office for Forest Studies (BNEF) and the National Office for Rural DeveloRment Studies (BNEDER) to undertake on its behalf studies to prepare detailed plans for the improvement of watersheds or the management of forests. Both agencies are Economic Public Enterprises (EPEs). Personnel in these well-staffed enterprises need further training to improve the quality of management plans. 2.20 The National Institute for Forest Research (INRF) was created in 1981 as an Autoromous Administrative Agency (EPA), under the auspices of the Ministry of Agriculture. It is organized into five technical departments (forest ecology, silviculture, reforestation and soil conservatior., forest protection, and information) as well as a department of administration. Its professional staff has increased from about 10 researchers in 1981 to 80 now. In addition, it employs about 50 research assistants, 40 administrative support staff, and 185 laborers. INRF has a main complex of offices and laboratories on the I/ Decree no. 90-116, dated April 21, 1990 and Decrees no. 80-312 to 80-318, dated October 13, 1990.  outskirts of Algiers and 14 other field stations spread throughout Algeria. Research is organized into a number of programs covering forests, mountainous areas, the steppe, and the Sahara. Researchers are working on over 50 projects at present, which means that efforts are somewhat dispersed and that, consequently, the results of research take many years to realize. This problem is compounded by a shortage of recurrent budget, the bulk of which is spent on salaries. Moreover, some of the research issues that should be addressed are not. For example, there is insufficient research on the silviculture of indigenous wood species; development of techniques for regeneration of forests, including the need to tackle social and pastoral aspects; means to compensate livestock owners when forests have to be closed to grazing; production of forage trees; wood technology; the role of forest roads as major causes of erosion; mountain agriculture; finding answers to practical issues raised by foresters in the field; and improving means to diffuse the research results that do exist. In short, research needs to be better focused, prioritized, and linked to the social and pastoral aspects that have been largely neglected until now. D. Government Plans and Policies 2.21 Algeria is faced with rapid deforestation, decreasing productivity of forest rangelands, growing erosion damage, and increasing sedimentation of dams. Despite these pressing problems, the share of the public investment program allocated to the sector has remained roughly constant, at around 2X of the total (DA 1 billion out of a total of DA 50 billion in 1990). Part of the reason that Government has been unwilling to increase the share allocated to forestry has been the poor return to past investments (para. 2.10). Government's report on the execution of the last Plan notes that, despite the results achieved in the sector, the objectives anticipated were not realized. The report cites a lack of funds (DA 4.7 billion planned, against DA 7 billion requested by the Ministry of Agriculture), weak implementation capacity, weak organizational arrangements, inadequate definition of priorities, and inadequate effort. Given Government's tight budgetary situation, it is unlikely that the allocation of funds to the sector can be increased significantly. What is important is institutional arrangements to ensure that funds allocated to the sector are not diverted elsewhere and that better use is made of existing resources. A recent reorganization of the sector is intended to meet the first objective; the proposed project is intended to meet the second. 2.22 Government's objectives for the long-term development of the sector1' include the following: - the protection and improvement of watersheds, the improvement of mountainous areas, and the continued fight against desertification in the South (the aim is to help reduce erosion and siltation of dams and to increase rural incomes, thereby slowing down migration to urban areas); As stated in "Perspectives a moyen terme de developpement economique et social", Conseil National de Planification, December 1989.  - 8 - - the improved management of existing forest resources, including the extension of forests where possible (this entails increased protection and a more rational utilization of the wood and forage resources); and - the introduction of new technologies and the strengthening of research to modernize the sector, reduce costs, and increase the participation of farmers and livestock owners in the management of forests and watersheds. 2.23 Government has also stated its intention to increase the number of parks and reserves over the period 1990-99, adding ten new nature reserves (including four wetlands), seven new national parks, and seven new game reserves. The total area of these new conservation areas would exceed 400,000 ha. 2.24 Apart from the specific plans for parks and reserves, several of the long-term objectives noted above lack concrete proposals for implementation. For example, at present, domestic wood industries do not appear to be interested in purchasing local wood (para. 2.44). Utilization of forests therefore has to be linked to a better understanding of why forest industries are reluctant to use local wood despite the wood scarcity. Finally, Government plans to give more attention than it has in the past to forest areas as a potential source of faed for livestock. The proposed pilot project aims to address this and other issues and to develop concrete proposals for tackling them. 2.25 Government intends to continue its policy of providing low-cost natural gas to consumers for cooking so that, unlike in many countries, the demand to cut wood for fuel remains relatively low. E. Forest and Environment Legiulation 2.26 The basic law governing tL. Algerian forests is the General Forest Law (Law 84-12) of June 23, 1984, together with related implementing decrees and orders. The law states that all existing forests and other woody vegetation, as well as forestable areas, be they state or collective lands, belong to the national forest estate and are governed by the General Forest Law. Private forests are also to be managed according to this law. Forests are classified in three broad categories by purpose: productive forests, protective forests, and special-interest forests (parks, reserves, and forests of military or scientific interest). 2.27 The law establishes a legal framework for: (a) the protection of forests against agricultural encroachment, overgrazing, fires, pest and diseases, and construction; (b) the management of forests (management plans); (c) reforestation; (d) soil conservation and erosion control (public-interest areas for protection, rehabilitation, and development of degraded areas); and (e) forest police. However, comprehensive implementing decrees are yet to be issued. Although the legal framework still needs to be further developed, the main issue is lack of enforcement rather than the absence of legal provisions.  -9 - 2.28 Environment and nature conservation is governed by a General Law on the Protection of the Environment (Law 83-03) of February 5, 1983. Most implementing decrees are under active preparation or have been issued -- including a decree issued in February 1990, requiring assessments for all works or development schemes with potential negative environmental effects. Algeria has also ratified several major international conventions on flora and fauna protection, as well as on air and sea pollution. In short, Algeria has a generally adequate legislative framework for the conservation of nature. However, proper enforcement still remains an issue. F. Forest ORerations Management Plans and their Implementation 2.29 The 1984 General Forest Law states that forests are to be maintained according to management plans approved by the Minister of Agriculture and drawn up after consultation with local populations. The plans should detail how to manage, use, and protect the forest. Management plans are now prepared by the National Office for Forest Studies (BNEF). To date, detailed management plans are available for about 910,000 ha, or 50% of the total forest area excluding the maquis. Priority was given to the forests of Aleppo pine and cork and zeen oak. BNEF has prepared management plans for an average of 56,000 ha a year. It now intends to double this rate, to about 100,000 ha a year. Even at this accelerated pace, however, it would take more thapi 10 years to cover the total forest area in Algeria (excluding the maquis). 2.30 Management plans have been excessively detailed and complex and have not provided practical blueprints for action. They need to be simplified to become more readily operational and accessible to local-level forest staff and to implementation agencies. Because of their complexity, and also for financial and technical reasons, management plans have been poorly implemented. Management plans include three broad types of operations: infrastructure (construction of forest roads, culverts, etc.), logging (thinnings and fellings), and plantation or regeneration. In general, infrastructural objectives have been met, but not those of other operations. In particular, there has been almost no thinning of forests, an essential management technique to ensure the proper development of forests. Exploitation, and hence replanting has also been limited. Because of heavy grazing, virtually no regeneration of cedar and oak is taking place. As a result, many of the cedar and oak forests are overmature and are starting to decay. Forest research needs to put more emphasis on the development and dissemination of appropriate regeneration techniques. 2.31 Two bilateral projects, financed by German and French assistance, aim to improve the management of forests in Algeria. The former, which is about to start, is an integrated rural development project on 45,000 ha of agricultural and forest land in the wilaya of Skikda. The main components include, inter alia, preparation and implementation of a forest management plan, experimental logging-and processing of cork and zeen oak, water resources development, and intensification of forage production. The latter project, which started recently, covers 34,000 ha in the Guelma wilaya. The main objective of the project is the conservation of the Barbary deer, a unique species. It also includes the updating of the management plan, silvopastoral development, and  - 10 - training. The proposed Bank project complements these two bilateral operations by focusing on areas and operations they do not cover. However, close contact will be necessary among these operations to ensure that lessons learned in one are incorporated in the others. Plantations 2.32. There are about 100 nurseries in Algeria producing seedlings for the forest service or for distribution to farmers in mountainous regions. The nurseries have sufficient capacity to meet demand in the foreseeable future. However, techniques used for the production of seedlings need to be updated. Present methods, which use a polythene bag filled with an overly dense substrate, result in the roots becoming entangled as they wrap around inside the bag. The main taproot becomes deformed or breaks when the seedling is removed. The seedling is thus bigger and less hardy than it should be, requiring both a larger hole for planting and more careful maintenance (including more irrigation); this technique also results in reduced growth and output. Unit costs of planting and maintenance could be reduced by introducing modern nursery techniques. This would also reduce seedling mortality and increase the productivity of plantations. 2.33 Records indicate that, in the early 1980s, the equivalent of some 50,000 ha of forest were planted a year; this has slowed to 30,000 ha a year more recently. The average success rate of these plantations is low, at 65X, giving an actual replanting of less than 20,000 ha a year -- almost exactly balancing losses by fire (an average of 19,000 ha a year). The low success rate of plantations substantially raises their costs, for several reasons. Many plantings are in semi-arid areas, but harsh climatological conditions could be overcome by the application of improved techniques. Frequently, decisions to plant are undertaken in the absence of soil surveys, so that inappropriate species are planted. More research needs to be undertaken to match soils and climatic conditions with the best-adapted species. 2.34 Land clearing and preparation is currently done by a bulldozer fitted with a blade. This overly disturbs the soil and removes the bulk of the organic matter contained in the topsoil. As noted above, poor seedling preparation means that taproots are often damaged or are planted horizontally. In short, techniques need to be modified and unit costs reduced. In 1988/89, EMIFORsl/ had a contract for replanting in the wilaya of Skikda at a negotiated price of DA 18,000 (US$1,440) per ha, or about one and half times the cost in neighboring countries. This cost is a function of the exchange rate and of the high financial prices for many inputs, but it also reflects high input use. For example, each worker typically plants some 50-60 seedlings a day, compared with 70-80 in neighboring countries and 100-200 in northern Mediterranean countries. The new techniques to be introduced under the pilot project will require a smaller planting hole, thereby saving time and increasing productivity. v/ Local-level Forest Enterprises, state enterprises carrying out forest operations in the Wilaya before reorganization.  - 11 - 2.35 Finally, local populations are often not adequately involved and consequently view the plantation as reducing the forage available to their animals. Consequently, they may deliberately attempt to destroy the plantations by fire, or they may not be willing to keep their animals out during the important establishment period. Programming of Fellings and Marketing of Forest Stands 2.36 Until recently, forest stands could in theory be felled only if a definite market had been found. Annual fellings were programmed and executed iteratively as follows. First, a tentative program was established by the forest service on the basis of recommendations in the management plans and estimates of the needs of processing industries. Second, the National Office for Forest Works (ONTF)1Y, or another logging operator selected by the forest service>, was responsible for finding a potential buyer and negotiating a contract with him. Third, the forest service would finalize the felling program for those stands on which there was a negotiated contract. Finally, after signature of the contracts, the local forest services would proceed with marking individual trees to be felled, before actual felling by ONTF. ONTF, or the selected contractor, did not have to pay for the standing timber, which was provided free. They were paid for logging the wood and, at the same time, were able to sell the wood to processing industries. In short, becaise of high unit costs, the only way that ONTF could make a "profit" on the sale of wood was by teceiving free timber and being paid the cost of logging the wood. 2.37 Many problems arose because of the poor implementation of this process. In general, it proved extremely difficult for ONTF to find buyers: exploitation and transport costs were high, the equipment utilized by the processing industries was often inadequate for local wood, and, because of the overvalued exchange rate, imports were cheaper. In some instances, where a contract had been negotiated, buyers would subsequently refuse to collect and buy the wood. Furthermore, some stands were logged even though this operation was not called for in the management plan and no effort had been made to find an outlet. 2.38 In future, the situation should improve because an implementing decree to the 1984 General Forest Law has been issued (Decree no 89-170, dated September 5, 1989), according to which forest stands are sold at stump by auction, or, under exceptional circumstances, by direct contracting. Exploitation 2.39 Because only small quantities of wood have been extracted from forests over the last two decades, none of the logging operators has much expertise in forest exploitation. They usually are poorly organized and operate with inappropriate equipment. In addition, in the absence of competitive I/ Before reorganization, now replaced by ORDFs. During recent years, ONTF logged about 90X of the total volume of wood felled, the private sector about 9.5X, and the EMIFORs about 0.51.  - 12 bidding, there was no incentive for keeping costs down. As a result, logging costs average about US$55/m3 for timber and US$25/m3 for panelwood, which is about five times more expensive than in neighboring countries in similar situations (converted at the official exchange rate)Z/. 2.40 Domestic wood supply consists of about 70X Aleppo pine, 20X eucalyptus, and 10 other species (essentially zeen oak, cedar, and Pinus pinaster). On average, domestic production is used for the following end-uses: about 50% for panelwood, 25% for timber, 20X for fuelwood, and 5% for miscellaneous uses. pomestic Consumption and Wood Processing Industries 2.41 The current consumption of wood products in Algeria amounts to about 1.1 million m3 a year. More than 90% is imported. Imports consist primarily of sawn wood (75% of total volume), with some lo, and plywood and veneer (20% and 5% of total volume, respectively). Wood is consumed by a very active construction industry and by wood processing industries. Public sector enterprises largely dominate the processing and distribution of wood and wood products in Algeria. 2.42 Wood processing industries consist of large public enterprises, which absorb about 55% of total supply, and private cottage i'ndustries, which, together with the construction industry, absorb about 451 of total supply. The two largest public enterprises processing wood are the National Enterprise for Wood Processing and Furniture (ENATB), which makes panels and furniture exclusively of wood, either domestic or imported, and the National Enterprise for Joinery and Prefabrication (ENMGP), which makes Joinery and prefabricated houses from imported wood and metal. ENATB has 20 processing units, most of them grouped into four integrated complexes and spread throughout the country. It employs 7,000 people and accounts for 20% of total Algerian wood consumption (171 of total imports, essentially logs, and 831 of total domestic supplies). ENMGP comprises 23 processing units, employs 6,000 people, and accounts for 351 of total Algerian wood consumption and 381 of total imports (mostly sawn wood). Three other public enterprises -- the Post and Telecommunication (PTT), the National Railway Company (SNTF), and the National Tobacco and Match Company (SNTA) - - also consume small quantities of wood for telephone poles, sleepers and matches, respectively. The cottage industries are extremely small-scale, usually with part-time operations producing such items as custom furniture and wooden crates. 2.43 Algerian wood processing industries prefer imported wood in part because the qualities of local wood are poorly known for lack of research. Algerian Aleppo pine has the potential to become a good substitute for imported wood. This problem should be helped by a Belgium-financed technical assistance project. The proposed Bank pilot project complements this effort (para. 5.17), supporting INRF in the creation of a wood technology laboratory for research in wood utilization. I/ Converted at the shadow exchange rate, Algerian logging costs are on average three times higher than logging costs in neighboring countries.  - 13 - Markets and Marketing Channels 2.44 ImRort Channels. Until 1989, a state agency, the National Enterprise for Wood Supply (ENAB) had a virtual monopoly on wood imports for resale (ENATB, SNTA, and SNTF could import wood directly for their own consumption). Wood imported by ENAB was sold to the Enterprises for the Distribution of Construction Materials (EDIMCOs), which handled retail sales at the wilaya level. Wood processing industries other t1;ar. ENTAB, SNTA, and SNTF could purchase wood directly from ENAB or from the EDIMCOs. Since 1989, any wood processing industry can import wood directly, provided that it obtains an import license from the Office for External Trade and has foreign exchange. As a result, ENMGP, which requires large supplies, imports directly. Medium- and small-scale operators have started to associate in order to import collectively. However, ENAB and the EDIMCOs remain the major suppliers of importsd wood for small-scale industries and private individuals. 2.45 As noted, Algeria continues to experience chronic wood shortages, particularly for construction purposes. Internal prices remain substantially higher than import prices, both on the official and the parallel market. Most prices for wood and wood products are still administered, including sale prices from ENAB, the EDIMCOs, and ENATB. Because of the shortages, the EDIMCOs give priority to hospitals, schools, and other public sector entities before supplying small- and medium-scale industries. As a result, an active parallel market has developed in addition to the official ENAB-EDIMCO distribution channels. Prices on the parallel market can be two to three times the administered prices. In 1990, import prices of sawn wood averaged DA 1,200/m3; EDIMCO retail prices were DA 5,000/m3; and prices on the parallel market often reached DA 10,000/m3. 2.46 Domestic Supplv. The domestic supply of wood is marginal. In recent years domestic supply has averaged 130,000 m3/yr, excluding the exceptional 1987 year when domestic supply increased to about 210,000 m3. In 1990, Pinus haleRensis saw-logs sold for DA 640/m3, and eucalyptus panelwood for DA 403/m3, which translates into DA 740/m3 and DA 465/m3 respectively at mill gate after allowing for transport. Sawn wood from such local timber would cost about twice these prices, still significantly below the price of imported sawn wood, which illustrates the need to (a) better understand why the local market is reluctant to use local wood and (b) improve local wood technologies (para. 2.43). Pastoralism and Forest Range Management 2.47 Livestock, which has grown faster than any other sub-sector in agriculture over the past decade, contributes 40X of the gross value of agricultural production. The rapid growth of the livestock sector is a result of the liberalization of milk and meat prices in the early 1980s, when demand was increasing because of population and income growth. However, because forage crop cultivation is extremely limited in Algeria, this rapid growth relied mostly on overgrazing of open-access ranges such as the esparto grass and steppe areas, and forests. In 1985 it was estimated that open-access ranges accounted for about 50% of total forage supply, fallow land for 25X, and forage crops, agricultural by-products, and concentrated feed for another 251. It is estimated that  - 14 - livestock pressure on forest ranges increased fourfold between the 1950s and the 1980s and now exceeds carrying capacity in some locations. 2.48 Overgrazing plays a key role in the degradation of forests and of the natural environment of watersheds. In the latter, where livestock often accounts for a large part of agricultural incomes, overgrazing leads to the regression of the natural vegetative cover provided by forests and maquis, thereby contributing to soil erosion. Unless the benefits of forest management are adequately discussed with them, local populations sometimes oppose forest operations, such as fencing for regeneration or plantation purposes, because this reduces their access to grazing (overgrazing prevetts natural regeneration of cedar and oak stands). Because of the political sensitivity of curtailing grazing rights at the local level, the forest service has found it difficult to regulate forest ranges. In any event, given the vast areas to be controlled, "regulation" through enforcement is unlikely to be effective in the long run. Algeria has to start gaining experience with improving range and forest management by seeking the active cooperation of livestock owners and delegating control and exploitation to user groups. The proposed pilot project would contribute to improving this situation through the creation of compensatory forage resources in the plantation component of the project (para. 5.15), through the drawing up of an experimental management plan for multipurpose use of forests and maquis (para. 5.05), and by research on the means to better integrate livestock in production systems in mountain agriculture (para. 5.28). G. Watershed Operations Nature and ScoRe of the Problam 2.49 Some 12 million ha are threatened by erosion, and it is thought that approximately 40,000 ha of previously cultivated land are lost each year to erosion and degradation. This has se.ious implications for the sustainability of agricultural production in the nation's mountainous areas, for the social and economic life of the communities that live in the watersheds, and for the productive life of the dams that are so crucial to future water supplies for irrigation, domestic use, and industry. Algeria has some 48 major dams already constructed or under construction. An ambitious program of watershed management and improvement has been under way in 33 watersheds, but the program has not had the expected results. Erosion is continuing unabated. Costs of improvements are too high to replicate on a large scale, and local populations have not participated to the necessary extent. Priority is focused, first, on the 21 major dams that are currently in full use, then on the dams built but not yet full, and finally on the 12 new dams that are either under construction or at the design stage. The total catchment area of these dams is about 5.5 million ha. Sedimentation rates in these catchments vary considerably. In some, it is less than 1 m3/ha/year, while in others it is above 40 m3/ha/year. 2.50 Within the 5.5 million ha of priority catchment areas, there are approximately 320,000 families who depend, to varying extent, on agriculture for their living. Although productivity in the mountains is not as high as in other, more favored parts of the country, the extent of the mountainous areas means that any decline in the agricultural resource base will have significant social and  - 15 - economic impact, especially in terms of migration to the towns, increased urban unemployment, and an escalation in demand for essential resources and services in cities. The Process of Erosion 2.51 The process of erosion is both complex and varied. There are, however, two components that are particularly important and that need to be tackled in any effective soil and water conservation program. The first is run- off, and the second is erosion of the banks of ravines and riverbeds. 2.52 The process of erosion starts when rainfall is not able to infiltrate into the soil. There can be many causes, but the most important are that the soil is saturated or that its surface has become impermeable, usually capped, so that the water cannot enter the soil. Both circumstances occur frequently in Algeria's mountainous areas. 2.53 Soils become saturated -- they reach field capacity -- usually toward the end of the winter rains, particularly during a good rainfall season. Moreover, many soils in the mountainous areas are susceptible to capping, which results mainly from a deterioration in the structure of soils left bare through overgrazing or cultivation. If organic matter is not replaced in fragile soils that are cultivated frequently, the soils start to crumble. This reduces their capacity to store water and increases their tendency to cap. Once this happens, heavy rains give rise to a great deal of runoff. Although a source of runoff, the soils do not generally contribute directly to sedimentation. Erosion rates from the catchment slopes are usually less than 5 tons/ha/year and the large amounts of water that move down the catchment are therefore relatively free of sediment. In the long term, however, it is inevitable that the topsoil will be washed downstream. 2.54 The situation changes significantly once the runoff water accumulates in ravines and riverbeds. The denuded hillsides offer little resistance to the flow of water, which quickly concentrates in gullies as fast-moving streams. This process is exacerbated by inappropriate design of forest roads and drains, which tend to concentrate the flow of water, thereby quickly forming new ravines and gullies. Once in the riverbeds, the water erodes the riverbanks, collecting enormous quantities of sediment. The erosion rate in badly gullied areas ranges from 100 to 200 tons/ha/year, while further downstream in the riverbeds it can range from 1,000 to 10,000 tons/ha/year. In short, the bulk of the sedimentation of dams comes not from farmers' fields in the upper catchment areas, but from ravines and riverbeds downstream. The solution consequently lies in slowing water runoff in the upper catchments, stopping the process that causes the structure of the soils to deteriorate, and stabilizing, to the extent possible, soils in the ravines and riverbeds. Present Approach to Erosion and Watershed Management 2.55 The problem of erosion is a problem of land use management and has, therefore, to be tackled on a multi-dimensional front. Soil conservation work has generally ignored the social, political, and economic factors that have contributed to the present state and that have led to the failure of previous  - 16 - attempts to redress the situation. The populations in the mountainous areas have not been involved either in the analysis of the cause of the problem or in the design of possible solutions. Soil erosion and land degradation in Algeria are frequently regarded as purely physical problems. Consequently, purely physical solutions have usually been applied. For the most part these have not worked. 2.56 When the local population has been involved, it has usually been as laborers paid to implement "Government" conservation works either on communal land or on farmers' fields. For example, the forest service has planted many trees and constructed large numbers of bench terraces on farmers' fields using paid labor - - the works have been provided free to the farmer. This approach has neither given farmers a sense of responsible ownership of the results of their work nor encouraged others to replicate the work elsewhere in the catchment. The farmers see the works as purely a Government activity done to benefit those downstream. Thus the "top-down", heavily subsidized watershed management programs have acted as a disincentive to the management and maintenance of existing programs and to the voluntary expansion of the work to other areas. There has been little done to encourage the responsible management of external initiatives by the local population; in many instances, conservation operations have been actively resented by the population because they have appeared to challenge local land rights. Furthermore, budget constraints have limited the impact of the work to those areas where direct Government intervention could be financed. 2.57 Not only has the approach tended to ignore the concerns of farmers in the upper catchments, but frequently the techniques used have been inappropriate. Conservation methods developed for other countries have been introduced with little adaptation to local conditions. For example, contour earth bunds, initially designed for use in North America, can, under certain circumstances in Algeria, contribute to increased rather than reduced erosion. Some 350,000 ha have been treated with this technique in Algeria, though it is now no longer widely used. Similarly, trees planted on farmers' fields have frequently not been of productive species (fruit, nuts, or forage), nor have they fulfilled their other objective of providing adequate vegetative cover to control erosion. Moreover, most techniques -- for example, gabion and stone-built weirs for control of gullying -- have been extremely costly and have been almost entirely financed by the Government, either directly or indirectly through subsidies to the farmers. 2.58 With some notable exceptions, little effort has been made to evaluate the impact and effectiveness of conservation systems or to develop new and cheaper techniques. As a result, the approaches and techniques that were introduced over 40 years ago, and which have either little impact or high costs, are still frequently being applied. INRF is aware of these shortcomings and has initiated a number of research programs to overcome them (for example, a joint INRF/ORSTOMI4 agreement to undertake applied research in selected areas of the country). This, and experience and progress of the German-aid-financed Integrated Watershed Management Project in Oued Mina, just to the west of the Bank pilot project watersheds, will be of considerable interest because it is Office for Scientific and Technical Research Overseas, France.  - 17 planned to implement a series of low-cost, participatory management techniques on exceptionally eroded areas. Need for an Alternative Approach 2.59 The generally narrow sectoral approach that has been used for watershed management to date has ignored the important linkages between the various elements of the production systems. There has been little contact between those working on forestry and soil conservation and those responsible for crop, livestock, and water development. As a result, solutions that involved alterations to the farming system or to the overall pattern of land use have infrequently been attempted. 2.60 The high costs, poor results, and eventual rejection by the population of past watershed management programs clearly indicate the need for a new and more acceptable approach and the applicatiorn of more effective and less costly techniques. Any additional investment that uses the existing strategies and techniques will incur continuing high costs and have very limited impact. 2.61 The scale and gravity of the situation in Algeria's watersheds requires a response, or set of responses, that is cheap and effective yet sufficiently attractive and simple to be implemented, for the most part, by the population themselves. Government cannot afford to pay directly for all catchment management work, and it is therefore essential that methods are found that will involve the population to a much greater extent in the design, implementation, maintenance, financing, and operation of watershed management schemes. GOA is aware of this problem, and increased emphasis is now being given to decentralizing the development process and to developing multi-sectoral, participatory approaches to watershed management. The pilot watershed management component of the proposed project will help test the practicability of implementing this new policy. H. Nature Conservation Parks and Reserves 2.62 The four nature reserves that were established under the 1983 environmental legislation are, in 1991, still strictly "nature reserves" by international definition. They have no human inhabitants, and entry is prohibited except for scientific study, which is in progress in all four areas. Two are fenced (Mergueb and Beni Salah), and one is on a mountaintop with only one entry road (Babors). As noted, there are seven national Barks that are managed by the Ministry of Agriculture and two enormous parks in the Sahara under the Ministry of Information and Culture. The Ministry of Information and Culture has now been disbanded, and the two parks are autonomous. They are expected to be transferred to the Ministry of Agriculture in the near future under the recently established National Agency for Nature Conservation (ANN). The main object of these parks is to conserve the natural environment, but all Algerian parks have comparatively high, permanent human and livestock populations and accord with the international definition of "protected landscape'. They all have  - 18 - management plans to guide their administrative, if not their scientific, management. 2.63 The four game reserves are completely fenced and totally protected while research on the species in the reserve is in progress, but they will eventually be opened to hunting. Two (Djebel Sahara and Moutas el Had) have management plans. They accord with the international definition of "managed nature reserve". 2.64 Although Algeria is relatively well-advanced in nature conservation, there are a number of constraints that need to be overcome. Algeria has 11 mammal species, seven bird species, and one reptile species that are threatened with extinction throughout their world range. Also, a considerable number of plant species and sub-species are under threat. The most serious threat to the flora and fauna comes from overgrazing; poaching has generally been overcome. In addition to measures to increase the .rea under parks and reserves, emphasis now needs to be placed on: implementation of existing legislation; gaining a better understanding of the scientific aspects of protected areas (the present management plans focus on administrative rather than scientific needs); and finding the means to take into account the competing demands of local populations (for example, of neighboring livestock owners) so that they cooperate in the protection of these areas. The project originally provided for the implementation of a management plan for the largest park under the Ministry of Agriculture, El Kala, from which lessons can be learned for replication elsewhere. The El Kala is now, however, being prepared as a separate project for Global Environment Fund (GEF) financing. III. BANK INVOLVEMENT IN AGRICULTURE IN ALGERIA 3.01 Bank Group lending for agriculture in Algeria started in 1975, and ten projects have since been approved, for a total of US$488 million. Performance under the first three projects (all of which are now closed) was mixed, reflecting institutional constraints in the sector (the projects were the Technical Assistance/Rural Development Project, Loan 1159-AL; the Meat Industry Project, Loan 1739-AL; and the Bas Cheliff Irrigation Project, Loan 1802-AL). 3.02 From 1980 until mid-1986, the Bank had limited involvement in Algerian agriculture, with the exception of the two ongoing projects. More recently, the Bank and Government have begun to develop a policy dialogue and a lending program to address some of the major issues facing the sector. The Bank's current involvement in the sector aims primarily to support the recently privatized state farms and to lay the base for longer-term growth. This would be achieved by: (a) improving tne availability of basic inputs; (b) assisting in the development of basic services and their adaptation to the new environment, particularly for irrigation, extension, and credit; and (c) fostering the establishment of a sound price and incentive system. Longer-term objectives will be met through an expansion of the area under irrigation, building up research services, and addressing issues of land use management, in particular erosion and degradation.  - 19 - 3.03 Three irrigation projects (the Cheliff Irrigation Project, Loan 2809-AL; the Irrigation Engineering Project, Loan 2978-AL; and the Mitidja Irrigation Project, Loan 3076-AL) are designed to rehabilitate and extend existing large-scale irrigation schemes in the northern part of the country, improve irrigation planning, support the development of an investment program for small-scale irrigation, and prepare a ten-year irrigation program for small- and medium-scale irrigation. The irrigation projects are progressing satisfactorily, albeit witk some delays because of shortages of certain construction materials. 3.04 Regarding support services, an Agricultural Credit Project (Loan 3009-AL) is designed to strengthen the agricultural credit bank so that it is in a position to lend to the newly created groups of private farmers. A pilot Research and Extension Project (Loan 3216-AL), approved in June 1990, aims to develop and extend appropriate technical advice for these farmers. Work in association with UNDP is aimed at expanding fertilizer supplies as part of a larger program of studies on input distribution. An emergency project, the Desert Locust Control Project (Loan 3017-AL), was approved in January 1989 in the context of the expected 1988 locust invasion. Since the plague did not materialize, $36 million of the $58 million loan amount has been canceled. The Sahara Region Development Project, approved in September 1991, will support the development of an environmentally sound and sustainable development program for the northern Sahara. In the future, attention will be given to the further liberalization of the sector, and an agricultural sector adjustment loan is being planned. Additional projects in irrigation and a follow-up agricultural credit project are under preparation. Lessons Learned 3.05 Lessons in Algeria. The main lessons learned from the implementation of Bank projects in Algeria are to: (a) avoid overly complex project design, (b) ensure the full commitment of Government at each stage of project preparation, (c) ensure technical assistance early on during project implementation (para. 6.09), and (d) take special measures to ensure satisfactory project accounting and auditing (para. 5.51). 3.06 Lessons in ForestrY. The Bank has had no previous forestry projects in Algeria but lessons learned from 12 years of experience with forestry and watershed management projects in neighboring Morocco and Tunisia include that (a) management of the forest resource requires an integrated approach, taking into account the various demands on it, e.g. domestic stock grazing (para. 6.05); (b) traditional forest management plans are too costly to prepare and are therefore not available for most of the forest, and are too complex to be properly executed -- what is needed is simple, practical management plans (para. 5.05); and (c) technical success in treating relatively small water catchment areas is not enough to guarantee its replicability and to ensure a real impact on the very large tracts requiring intervention in Algeria -- for this purpose methods used need to be cheap, requiring a minimum of budget funds and relying to a maximum extent on self-financing and the participation of local populations (paras. 5.20-5.27).  - 20 - IV. THE PROJECT AREA General 4.01 The proposed project is national in scope in that it aims to define a national strategy for Algeria's forests and watersheds and for strengthening institutions at the national and field levels. Although the pilot operations to be financed under the project are necessarily confined to specific areas, lessons learned -- particularly the processes learned -- would be replicable to other parts of the country. The Pilot Forest Areas 4.02 The areas proposed for the pilot reforestation and exploitation components and the two management plans are in the northeast of the country, not far from the major port of Annaba. As noted, the three wilayate of Skikda, Annaba, and El Tarf are considered to be the most favorable for industrial plantations. 4.03 The reforestation component would be in the wilaya of Skikda, bordering the Mediterranean. Rainfall, which is spread over 10 months, varies between 800 and 1000 mm. High humidity in the two dry months partly compensates for the lack of rainfall. The climate is temperate, with an average minimum temperature of 8