Balancing Development and Conservation in Pre-Saharian Tunisia

Balancing Development and Conservation in Pre-Saharan Tunisia

BY GEORGES NOVIKOFF AND MOHAMED SKOURI

In the immediate hinterland of the Sahara desert, increasing human and animal populations, combined with increasingly intensive agriculture, have led to desertification of once-fertile grazing lands. The authors suggest ways to prevent the further spread of the desert.

disc harrowing cultivation in Tunisia. The introduction of disc harrowing during the last 25 years has accelerated soil erosion. Tunisia is located at the northeastern edge of Africa, bordered by Algeria, Libya, and the Mediterranean Sea. Its Mediterranean climate brings high summer temperatures, with daily variations increasing with dis- tance from the sea. Precipitation occurs in autumn, winter and spring. The Atlas Mountains, which extend b through North Africa, divide Tunisia into two quite different climatic zones. In the north, precipitation varies from 1500 mm to 350 mm annually. In the central and southern regions which occupy 80 percent of Tunisia’s 15.5 million hectares, precipitation varies from 350 mm to less than 100 mm. Central and southern Tunisia were formerly grazing areas, traditionally ex- ploited by nomadic pastoralists. Summer droughts limit Tunisia’s forage resources, and the forage utilization and animal production were at one time regulated ac- cording to climatic fluctuations. Over- grazing was always a problem, particu- larly in the south, but it was corrected by flock reductions due to droughts and by movement of flocks from place to place in search of green pastures. However, since the beginning of this century the population has become less nomadic. The land cultivated for cereal and tree crops (olives, almonds) has pro- gressively increased, resulting in reduced grazing land. Similarly as the human population rose, so too have the sheep and goat populations,, exerting further pressure. The introduction of disc ploughing during the last 25 years has accelerated erosion of thin soils. The combina- tion of human and livestock population

AMBIO, 1981 Figure 1. Principal study areas of the different projects in pre-Saharan Tunisia.

I. DE DJERBA

Location of major project sites 07 i Garaet Zougrata

03:US/IBP Desert Biome

Native vegetation Coastal Area and Saline Soils

Cultivated saline soils

Fruit trees o 25.000 olive trees Steppe with rocky hills . 25.000 individuals of different fruit-trees f Palms Steppe with loamy soils Xerothermic regime (amount of days biologically dry) Alfa (Stipa tenacissima) steppe with light soils Mediterranean zone A A A From 250to300 0 Low shrubs (garrigue) A A A From200to250 Introduced or modified vegetation A A A From 150 to 200 Crops (From the international map of vegetation Tunis-Sfax Crop without fallow A. Vernet and H. E. Gaussen with M.Gounot, N. Le Houerou, Discontinuous cropping system A. Schoenenberger, A. Thiault, M.Novikoff) Oasis I Vegetable gardens

136 AMBIO VOL 10 NO 2-3 increases, a shrinking land base, and strongly affected primary productivity, plant formation which is the most vulnera- agricultural expansion and mechanization and that an even distribution of rain ble to erosion and the most characteristic has led to a serious problem of degrada- throughout the year is much more favora- of the area. tion and even desertification of Tunisia’s ble to the development of vegetation than Two experimental stations, Dar ez Za- former grazing lands. a seasonal one. Furthermore droughts, oui and Henchir Essiaân, were also estab- In 1969 the Tunisia Ministry of occurring mostly in the late fall and early lished and meteorological recordings were Agriculture began a detailed study of pre- winter, are very detrimental to vegetation made during seven years. Annual precipi- Saharan environments, in an effort to re- growth. tation varied from 16.5 mm to 490 mm and store some ecological balance between In addition to confirming already two years out of seven, the rainfall was agriculture, production and basic resource known results, this research provides use- below 70 mm-the amount which defines utilization. Two major projects were un- ful quantitative data for precisely de- the beginning of a drought. dertaken, a Research and Development scribed environments. It also shows the Studies were carried out on plots with Project for Rangelands in Southern effect of rainfall and temperature on similar vegetation types, each plot being Central Tunisia (joint project UNDPlFAO physical and chemical features of the soil. used during a different season. Each year, /UNESCO), and a Desert Biome Project For example, it was shown that below a sheep and goats grazed in the same sea- under the International Biological Pro- relatively low threshold of precipitation sonal order on each plot. Results are gram (IBP). (200-250 mm per year), a sharp decrease shown in Table 1 for one year (1976-77). The aims of the projects are to improve in vegetation productivity occurs, indicat- The precipitation was 68 mm, only half the knowledge of the functioning of desert ing a very low soil trophic level. the yearly average (3). Some plant groups, fringe ecosystems and to identify and The results of the studies were used to such as the highly preferred small peren- formulate management methods that prepare range maps showing the major nial species, were consumed primarily would ensure rational multiple land use range units and other features, such as during the summer drought and less dur- for Tunisian rangelands. In addition the annual and seasonal primary productivity ing the other seasons. projects provide an approach to obtaining and optimal period of use, which are use- The groups of plants requiring only environmental information on desert trig- ful for rational land use planning. Such small amounts of precipitation (20-30 mm) gering mechanisms, and a natural re- plans were worked out for Oglat Merteba to grow in winter provided the most im- sources survey. (100 O00 ha), Zougrata (80 O00 ha) and Bir portant part of the diet of ewes during the With the creation of the UNESCO Man Lahmar (10000 ha), and several other critical lambing period. This was the case and the Biosphere Program, followed by smaller areas for the purpose of demon- for biannual plant species in the winter an expression of interest from Tunisia, stration and extension. Methods for in- plot. Another group of species, which both studies were incorporated into the creasing animal production were im- may be called the forage reserve compo- MAB research activity on grazing lands proved through various forage and animal nent of the range, is consumed in large and arid and semi-arid ecosystems. As husbandry experiments. For the Zougrata quantities only after the other plant such they represent one example of area, simulation models were worked to groups have been exhausted. In this study MAB’s efforts to apply ecological re- predict trends in productivity (2). this was the case of shrubs at the end of search to critical environmental and de- the winter grazing period. Annual species velopment problems. Range management studies were poorly represented in the animals’ Extensive studies were carried out on diet: due to low precipitation, few annuals ways of improving livestock management grew, and those that did were quite small. SUMMARY RESEARCH RESULTS and of stimulating the regeneration of the During a year when precipitation was AND DISCUSSION plant cover. For example, under the Des- closer to the average (such as the 154 mm Most of the research is concentrated in ert Biome Project a study was done of the between September 1977 and summer the coastal region of southern Tunisia arfej (Rhanterium suaveolens) steppe, the 1978), annual species represented 50 per- (Figure 1). The area includes the oases of Gabes and Zarzis and the towns of Mede- nin, Mareth, and Ben Gardan. Vegetation is a sparse and shrubby steppe. Annual rainfall ranges from 180-200 mm on the coast to 100-120 mm inland, typical for Table 1. Consumption of plant species groups in plots grazed during different seasons the pre-Saharan areas of Africa. Beginning of rainfall Research has been conducted on four main themes: primary productivity as a Summer Autumn Winter Spring basis for range management planning; Plant species groups Beginning End of Beginning End of Beginning range management studies of effects of of season season of season season of season grazing practices on vegetation and soil; desertification and ways to combat it; and Percent of diet by weight the preparation of management proposals taking account of ecological and socio- Litter Sheep 12.9 89.8 74.3 0.1 1.7 2.3 economic factors. Goats 2.6 60.6 42.9 O O O Primary productivity Annual Sheep 0.4 O 0.5 0.6 7.1 15.2 species Goats 1.1 , O 0.4 0.3 0.4 5.2 The studies on primary productivity at- Small Sheep 69.5 2.0 5.3 14.7 11.8 21.1 tempted to understand the functioning of perennial Goats 87.1 4.2 7.4 19.9 10.6 20.8 pre-desertic ecosystems at Oglat Merteba, species Zougrata, and Bir Lahmar, and identified Shrubs Sheep 4.6 7.8 12.2 7.8 67.9 20.9 the factors which affect the primary pro- Goats 4.5 34.9 45.4 17.3 84.0 42.9 ductivity of rangelands in these sites. Biannual Sheep 12.2 O 6.9 72.6 10.6 40.5 Data was gathered Òn productivity in species Goats 4.7 O 1.o 62.1 0.9 31.0 relation to different climatic, soil, and hu- Perennial Sheep 0.4 0.4 0.8 0.2 0.9 O man factors (1). It was found that the total grasses Goats 0.1 0.2 2.9 0.4 4.1 0.1 precipitation and its seasonal distribution

AMBIO, 1981 137 cent of the total productivity, and their Table 2. Plant productivity (in kg) under different grazing schedules. size was taller than 10 cm. Annuals were Annual Biannual Small Rhsnterium Artemisia Other highly preferred in winter and spring, species species perennials suswedens campesfris shrubs Total ranking ahead of the biannual species. (shrub) (shrub) The forage reserve component was consumed mostly at the end of spring and Continuous year-long 191 A25 167.459 67.223 225.429 5.098 46.104 702.938* summer. grazing k 70.55 k 26.42 f 15.69 k 29.77 f 2.59 k 17.59 (Saad Boukli) It can be concluded, therefore, that to ensure optimal nutrition of the flock with Spring-deferred 409.643 51.418 76.693 117.769 29.236 178.237 862.996, minimum supplementary feeding, a pre- plot Saharan pasture should contain large Summer-deferred 341.932 45.738 66.097 137.783 5.964 129.949 727.462 amounts of each of the above plant plot 1 f 32.22 k 10.68 f 16.26 k 17.78 ? 2.79 k 17.85 groups. These species need to be de- Summer-deferred 275.426 19.724 73.999 133.304 98.249 88.286 688.988 veloped either through adequate grazing plot 2 k 39.98 k 8.98 k 16.02 k 35.20 k 30.08 f 41.81 treatments, or by reseeding the poorest ranges (4). * Difference between these totals was significant at the 90 percent confidence level. Studies of mixed flocks, half of sheep and half of goats, showed that goats in such a flock consumed much more of the tallest vegetation group (shrubs) than a Figure 2. Predicted trends of desertificationindexes with four different levels of exploitation. flock of goats alone. Sheep consume larger amounts of the annual and biannual species when in a mixed flock than in all- sheep flocks. From a practical standpoint, OGLAT MERTEBA STUDY AREA this suggests that the goat is not the nuis- f 20 O00hecibrsl ance animal that it is generally considered to be. Man himself, by associating goats , a - AREAS DESERT/f/ED. with sheep, has made grazing by goats Hn(La?d only gmzed I. Ha.(Land used for duo1 purpses I more destructive. Similar results have Level 4 been found in South America (5). Level 4 Leve1 3 The consumption of litter has serious Level 3 consequences for the protection of fragile soils against wind erosion (6). When rains Level 2 4 O00 I O00 are delayed, sheep and goats, finding no 3000 o 5 10 15 JO Years25 level 2 new forage resources, consume large 0- 0 5 10 15 20 25 amounts of litter. Excessive loss of litter !E Years. exposes the soil to trampling, loosening b - AVERAGE PR/MAUY RUNOFF CEFf/C//MIT FOR THE WOLE STUDY AREA. the soil particles which are then carried away by wind. The threshold amount of ‘/Kt1 (Average yew] %Xt 2 (moist years I litter was measured in the plots; it was 4Ot l6 f Level 4 found to vary with the vegetation com- position of the range. In terms of the Level 4 Level 3 quantity of litter consumption, it is in- Level 3 ::E, Level 2 teresting to note that sheep contribute Level 1 28 more to desertification than goats. Level 2 2L Level 1 Studies of grazing treatments showed o 5 10 15 20 x O 5 10 15 20 25 an increase of 22 percent in total pro- Yeors. Years. ductivity for spring deferred grazing, as c - CONSUMABLE PRIMARY Pht3DI/CTlV/TY OfTHE GRAZED LANDS. Level 1 compared to continuous year-long grazing Tons/ho. (yew with overage rainfalll.

(Table 2). Furthermore, there was better 7 O00 Level 2 productivity from desired species compo- nents: an increase from shrubs, which are 6000 Leve1 1 Level 2 consumable in summer, and an increase 5 WO 7 OW from annual plant species, which take Level 3 quick advantage of available rain. 4’0W Level 3 6 O00 Level 4 Level 4 During a year of rainfall (150 mm), the 3WO 5 O00 O 5 10 15 20 25 O 5 10 15 20 25 resultant evenly distributed vegetation Years. Years. supported the project’s flock throughout d- AVERAGE STORAGE Of WAER AVAILABLE TO VEGETATKN the summer without the help of supplementary food (Table 3). Despite the Rv-mm Level I heavy rainfall, neighboring farmers were forced to supplement their animals’ diet with barley, straw and residues from the KEY extraction of olive oil, due to unrotated grazing. Even without supplementary 95 Level 3 Leyel I - Exclusion frwn grazing feeding, the weight of animals studied was Level 2 - Ratioml regiowl land use. as much as 8.5 kg heavier than neighbor- Level 3 - Upkeep of the aclual system. ing flocks. 85 , ~ Levd 4 - lncrmse of fhe actual practices of lond use o1 5 10 15 20 25 Ywrs

138 AMBIO VOL 10 NO 2-3 Desertification and ways to combat it Table 3. Comparison of the weight of ewes in study area and in three neighboring flocks.

Large areas of Tunisia are subject to de- Average weight of ewes by age categories, from Supplementary feeding sertification (7), primarily through wind berkoussa (the youngest) to charef (the oldest) action on denuded soils. Most soils are berkoussa holiaâ rebaâ sedasse jamaâ charef sandy, lack stability, and consist primarily of coarse particles (up to 80 percent). Average weight 15.42 23.49 37.14 31.88 30.52 31.80 From the first of August, Overgrazing, disc plowing of fragile of three neigh- 200 g of barley as grain per soils, and harvesting of shrubs for building boringflocks (in kg) head per day and 300 g of straw, 200 g of residues of materials and fuel are basic causes of de- olive oil extraction sertification. However, the latter has de- Project's flock 20.65 31.50 32.46 36.13 34.55 40.30 None creased in importance as butane gas is average weight increasingly replacing wood fuel in (in kg) Tunisia. Difference + 5.22 + 8.01 - 4.68 + 4.24 + 4.02 C 8.50 In Oglat Merteba (8), studies over an between the area of 20000 ha, have shown that the project's and neighbors major cause of desertification was disc plowing. Trends in desertification and re- covery were simulated to the year 2000 under four intensities of land use (see Fig- ure 2). Remedial suggestions include im- proved range management practices, ir- rigation of pastures and supplementary feeding of flocks, and limiting small grain production to only the most favorable sites. Additional experiments on methods of combatting desertification were carried out at two sites (area No. 3 in Figure 1) where overgrazing has been a problem. Early detection of sand movement is very important. The movement can be either, toward erosion or accumulation. Meas- urements at Dar ez Zaoui showed that the occurrence of ripple marks, fine undulations in the soil surface at regular inter- vals, is an early indicator of desertification (9). Furthermore, the percentage of surface area covered by these undulations is a good measure of severity. Spring and summer deferred grazing treatments were applied to plots where 40 to 60 percent of the soil surface was covered by ripple marks. After five years, Goats grazing. Studies by the authors and others suggest that goats alone do not cause great damage to this surface was reduced to less than 5 vegetation unless they are pastured in mixed flocks with sheep. percent. The improvement was due to an increased amount of vegetational residues covering the soil surface, enough to prevent the movement of soil particles, and Figure 3. Comparative soil los- a-, Treatment 1 Oases manure addpd ses with different treatments 2;2z/ha straw hcorporated hsorl before plaugh;qanl$'~78 an increase in vegetation density suffi- sr!ib!JlQ maintained upon harvest. during 1978-79. @- Treatment 2 ! No straw added cient to decrease the windspeed at the soil no oases manure added level (10). 0- - - Treatment 3 Straw added (2,2 t/ha)m.Wl/78 Methods for reducing losses of culti- stubbleoases manureuprooted added upon harvest vated soils were compared on experimen- 0- - - Trmhnent 4 No straw added no oases manure added tal plots surrounded by concrete ditches sfubble uprooted so that eroded soil could be collected and All treatments were drsc ploughed weighed. The crop used was barley. Land cultivation included only disc plowing, even though an optimal experiment would have used a sweep which does not over- turn the soil (11). In treatment 1, oases manure prepared from the feces of sheep and goats was applied to the soil, together with straw at the rate of 2.2 tons per hectare. The soil was then plowed and planted. Stubble was retained after harvest. Treatment 2 simply retained stubble. In treatment 3, manure was plowed into the soil. Straw was added after planting, and stubble was uprooted after harvesting. Treatment 4 uprooted stubble with no amendments.

AMBIO, 1981 139 Alfa grass (stipe tenacissima) on lower mountain slopes.

Results are shown in Figure 3. Adding The case of the Bir Lahmar area will be Animal ploughing has been replaced by oases manure with straw added before the only one presented in this paper. tractors, urbanization is taking place ra- or after plowing substantially reduced Although smaller in size, it includes the pidly (13), and jessours no longer are soil losses (treatments 1 and 3). Simply most basic characteristics of the others communally maintained. retaining the stubble from one year‘s due to its diversity of soils and the change Following flooding in 1969 the main jes- crop (treatment 2) did not produce a in land-use practices there. sours, which receive the first runoff and significant improvement over the control The Bir Lahmar irndu is an administra- which are responsible for 30 percent of (treatment 4). tive unit located at the junction of the agricultural production, were destroyed; Table 4 shows barley yields for the Matmatas Mountains and the Jeffara of it is predicted that their destruction will above four treatments plus the addition of Medenin (site No. 4 in Figure 1) where trigger the chain disappearance of the oth- straw alone. The latter increased yield by vegetation, soil, and socio-economic ers. Furthermore, the steppe was put into 1.9 times compared to the control. Adding studies were carried out (12). It includes cultivation, and farming and overgrazing manure and straw increased yield by 2.6 three diffeent land units. One is a of this fragile zone have brought about times (treatment 1) and 2.7 times (treat- mountainous area with traditional water desertification. ment 3). These experiments show that conservation structures, or “jessours”- simple techniques can be used to reduce little dams built with stones and earth and soil losses and increase crop yields at the planted with olive trees or fig trees. Be- Regional management models same time. fore 1954, this area was the center of A proposal for action was developed from agricultural activities. The second unit is the first phase of the project. It involved Drawing up of management proposals the pediment or lower mountain slopes. setting up eight main centers for demon- The studies were aimed towards land Previously used for grazing or Cultivated stration and popularization of range man- management and rational land use, for small grains, it has now become the agement in the arid and pre-desertic part whether they concerned soil surveys or core of urbanization. The third unit, or of Tunisia. Generally, the project con- phyto-ecological inventories or experi- steppe, located outside the pediment, is a sisted of managing small units of pasture- ments on evaluating the impact of the dif- rather fragile sandy area with high eroda- land (from a few hundred to several ferent environmental and social factors. bility, currently cultivated for small thousand hectares in size) by applying the Hence the results obtained were in- grains. most suitable forage production and ani- tegrated into two main types of action The population previously depended on mal husbandry techniques on the basis of proposals for development: pilot areas for subsistence farming, centered around a thorough knowledge of the natural and improving rangelands, forage and animal communally maintained jessours. How- human environment. Most of the pilot husbandry; and regional management ever, since 1954 traditional farming ac- areas have given conclusive results from models (12). tivities have disappeared. the technical point of view, but extending

140 AMBIO VOL 10 NO 2-3 Table 4. Yields of grain barley. Each vertical column is the average of five statistical replicates (except 8. Oglat Merteba region, case study on desertifica- for straw in which there are only three replicates). tion, UN Conference on desertification, Nairobi, August 29-September 9, 1977. 9. Hierarchical classification of sand accumulation Yields of Treatments Manure added Manure and Straw added Control (includes forms, in Dune Stabilization a survey of literature bar!ey before plowing, straw added ’ after planting treatments2 and4) on dune formation and dune stabilization, grain straw added before plowing geographisches Institut Universität Wiirzbourg, after planting August 1977, Deutsche gesellschaft fiir Tech- (treatment 3) (triatment I) nische Zusammenschaft, pp. 93-94. 10. Classifying rangeland for conservation planning, In grams per square 85.8 82.0 61.33 31.60 Agricultural Handbook No. 235, (United States 5 10.22 ?r 10.80 Department of Agriculture, Soil Conservation meter 2 9.25 ? 10.48 Service, 47 p. 1962. In kilograms per 858 820 613.3 316 11. D W Fryear, Reducing wind erosion in the South- hectare 92.5 102 108 104 ern Great Plains. (Texas agricultural experiment rt rt rt k station, MP 929, Texas A and M University, 1969, USA.) 12. Recherche et développement des parcours du centre-sud tunisien, Tunisie, synthèse des travaux réalisés au cours du projet et recommandations the methodology to other areas has to be From the practical point of view, the (FAO, Rome, AG/DP/TUN/69/001). weighed against land appropriation and difficulties encountered in implementing 13. T Bouhaouach, Contribution Ci l’analyse socio- éconorniaue de la répion de Bir Lahmar (Institut other land-use priorities. this program on a larger scale do not result des régi& arides, -Ministère de l’Agr&ulture, From these pilot areas, proposals for from the local population contesting the Médenine, 1980). 14. This project is supported jointly by UNESCO- the management of much larger areas validity of the technical results, but are MAB andi the United,Nations,EnvironmentPro- were prepared, and specific recommenda- more related to constraints such as the gramme (UNEP). tions were made for the three need for grazing rotation. pedomorphological (mountain, glacis or Popularizing the grazing technique, pediment and plain) units: through persuasion and through incen- 111 moiintain areas: rehabilitation and tives is part of the strategy employed in maintenance of jessours, with fruit trees the hope of ultimately combating de- (fig and almond) planted in the most sertification. In the past, considerable los- I Georges Novikoff received his docto- favorable parts, and forage bushes ses of livestock were frequent; now they rate in natural sciences from the Uni- (cactus, acacia, Parkinsonia) in valley are rare. At the same time, there has been versity of Montpellier in 1964 for his bottoms, wadi beds and the bases of hills. adequate forage during the years of research on relationships between On glacis: cereal culture improvement by drought. halophytic vegetation and soils in better use of runoff water and by impro- The proposals formulated through these Tunisia. At present he is principal ving soil working techniques and fertiliza- research activities emphasize an in- technical adviser for the MAB 3 pro- tion, since there is little regeneration of tegrated and multidisciplinary approach to ject on arid lands in southern Tunisia the natural plant cover. basic resource management in order to (UNESCO-UNEPhstitute of Arid Lands of Medenin). Earlier he was the 011the plain: limitation of cropping in or- diversify production and make the best resident scientific coordinator of the der to preserve the most productive use of southern Tunisia’s potential. This Tunisian pre-Saharan project spon- rangelands and to encourage the regenera- approach also takes into account the re- sored by the US/IBP Desert Biome tion of naturally-occurring species by con- quirements of the population, thus and Smithsonian Foreign Currency trolling grazing. Dunes that have already harmonizing the overall production objec- Program. He has been associated as formed can also be fixed by planting tives with the need to conserve natural a plant ecologist with a number of woody forage species (es Acacia cyano- resources. studies in land use planning at the phyllo, Retama raetiim). Investigations are still continuing in or- soil science section at the Ministry of Fruit crops grown with appropriate der to improve the recommendations in Agriculture of Tunisia and with various range management studies in techniques and in favorable areas situated the light of a dynamic population. In the Tunisia and Algeria. He is also Profes- close to villages in each of the three units final analysis it is hoped that the experi- sor of Range Management at the In- described above, are also advised. ence will provide guidelines for a land-use stitute of Arid Regions in Medenin. These proposals form part of an in- policy (14). His address: Institute of Arid Re- tegrated system of land management gions, Medenin, Tunisia. based on ecological principles and taking into account essential nutritional require- Mohamed Skouri received his docto- ments. The problem is to know whether References and Notes rate in natural sciences from the Uni- the local populations will accept the pro- 1. C Floret, R Pontanier, Relations sol-climat-vég- versity of Paris in 1966 for work on étation dans quelques formations végétales the regulation of food intake in ru- posed solutions and strategies. These spontanées du Sud tunisien-production végétale et points are being examined in a socio- bilan hydrique des sols. (Document technique No. minants in relation to the nutritive 1, Institut des régions arides, Ministère de economic study which is currently under- value of the diet. Subsequently, he l’Agriculture, Médenine, 1978). was Director of the Laboratory of way in Tunisia. 2. C Floret, E Le Floc’h, R Pontanier, F Romane, Modèle écologique régional en vue de la planifica- Animal Husbandry and Forage Pro- tion et de I’aménagement agro-pastoral des ré- duction at the National Institute of gions arides. (Document technique No. 2, Institut Agricultural Research of Tunisia (IN- des régions arides, Ministère de l’Agriculture, CONCLUSIONS Médenine, 1978). RAT), head of the Faculty of A considerable amount of data on 3. D and K Crocker Bedford, in Tunisianpresaharan Agricultural Sciences of the Uni- project, progress report No. 8 (USIIBP Desert versity of Tunis, and from 1970 to ecosystem dynamics of the pre-desert Biome, Utah State University, Logan, 1977), pp. 1975, head of the Research Division areas of south Tunisia has accumulated as 144-178. 4. Basic problems and basic techniques in range re- of the Ministry of Agriculture. He a result of the research undertaken in this search (National Academy of Sciences, National then served until 1978 as President zone. This data has been used to draw up Research Council, Washington, 1962). Director-General of the Board of Ani- 5. F di Castri, Division des Sciences écologiques, mal Production and Pasture Manage- action proposals for land management on Unesco, personal communication. a fairly large scale, and decision makers 6. N P Woodruff, F H Siddoway, Soil Science ment of Tunisia. He now works for have used these proposals as a basis for Society of America Proceedings, 29, pp 602-608 UNESCO as a specialist in arid land (1965). ecology. His address: Division ,of planning the development of these areas. 7. C Floret, E Le Floc’h, R Pontanier, Carte de Ecological Sciences, UNESCO, 75700 The implementation of the management sensibilité à la désertisation en Tunisie centrale et méridionale (Processus de dégradation en cours Paris, France. program of the area first studied, Oglat des sols et de la végétation) au 1/100.000 (Sols de Merteba, began about five years ago. Tunisie, Ministère de l’Agriculture, 8, pp. 4-68).

AMBIO, 1981 14 1