Contents - Previous - Next

Suggested improvements

[Plate 22]

Given their physical isolation in the highlands, minifundio farmers mainly need more on-the-spot technical assistance. Various forms of State intervention need to be increased: the organization of practical training in mountain farming, instruction in the use of fertilizers, selected seed, and of course conservation techniques, etc. Relations between minifundio farmers, technicians and agricultural scientists must be increased without delay, while combining research and development activities.

The programme of international co-operation between the National Agricultural Directorate (DNA) of the Ecuadorian Ministry of Agriculture and ORSTOM was designed in this context. It is a relatively pioneering project for the country, and indeed for the whole Andean region (De Noni and Viennot 1987, 1989), which has set up research stations on the farmers' land which are jointly managed with the farmers themselves. The stations have large runoff plots of 1000 m² (50 × 20 m) for study of the effects of erosion on land under crops when improved with some simple conservation structures. The stations were set up in 1986 at the same time as the previously mentioned 100 m² plots (see Table 46) in order to allow comparison of the effects of erosion under traditional farming methods (100 m² plots) and under improved methods (1000 m² plots). While the plots were being laid out, a socio-agricultural field survey was carried out to determine the various farming systems used in the research area, with particular emphasis on identifying conservation methods. In the absence of traditional practices, it was decided to test the effectiveness of simple semi-pervious contour structures in combating runoff energy, with the structures gradually developing into pseudo-terraces (Roose 1971; 1986; 1987a, b). An effort was made to keep as close to farmers as possible, using materials commonly used in the region, generally to fence off plots. The three main types of structure were thus as follows: low walls made of clods of earth or blocks of hardened volcanic ash (cangahua), or, more simply, grass strips, either grazed or cropped (quinoa or lupin). The methods tested over the period 1986-88 gave the results seen in Table 47.

These data demonstrate that simple conservation systems within the reach of the local farming community - contour ridging combined with grass strips or low earth walls - can noticeably reduce erosion. On all the improved plots, whatever the station, earth loss is minimal and erosion tolerable - usually less than 8 t/ha/yr and often close to 1 t/ha/yr. Yields are also better; for example, at Mojanda the potato harvest was 4.3 t/ha on the control plot and 7.6 t/ha on the improved plot.

At Riobamba, for the period from 20 September to 12 November 1987 (the sowing date), three erosive downpours resulted in a soil loss of 33.8 t/ha on the traditional plot where seedbeds had been prepared, while erosion on the improved plot was only 1.1 t/ha for the same period and the same tillage. And at Tumbaco, rainfall on 19 October 1987 alone, in the middle of the fallow period and one month prior to sowing (on 18 November), caused a soil loss of 34 t/ha, while the improved plot lost only 140 kg over the same period.

Although encouraging, these preliminary data show that not all the problems have been overcome, and that before launching awareness and extension programmes it is essential to carry out observations under both experimental and on-site conditions. This remark is based on the example provided by the changing size and shape of the grass-clod walls on the Mojanda station. Initially they were about 30 cm high, made up of two layers of earth clods. Then, although erosion on the plot was insignificant (0.2 to 0.3 t/ha/yr), in the course of the cropping year the farmer moves considerable amounts of soil from the top to the bottom of the plot with the broad blade of his mattock (asadon). Digging or hoeing always starts at the foot of a wall, creating a hollow at the base, and the soil is then drawn towards the bottom of the plot until another wall comes in the way. These simultaneous processes of hollowing out in front of walls and filling up behind them meant that the walls had to be heightened several times, rising from 30 to 130 cm during the 20 months of observations, while the initially straight slope steadily developed into terraces. It is estimated that almost 40 tonnes of soil per 100 linear metres accumulates behind the walls in this way each year.

Chapter 14. The Mediterranean Montane Region of Algeria

AGRICULTURAL INTENSIFICATION... WITHOUT DEGRADATION

Eric Roose, Director of Soils Research, ORSTOM, Montpellier, France
M. Arabi, Engineer, INRF, Médéa, Algeria

The situation
Diagnosis: trial conditions
Hazards
Suggested improvements: influence of the farming system

The situation

Although the northern region of Algeria is by far the most productive, it is a very fragile area of young mountains, with soft rocks such as argillite, marl and schist alternating with hard rocks such as limestone and sandstone. The semi-arid Mediterranean climate brings fine, gentle, but saturating rain during the cool winter, and dangerous thundery rainstorms during the torrid summer months.

The soils (regesols, grey vertisols, brown calcareous soils, and red fersiallitic soils) often have a slaked surface, and are gravelly, poor in organic matter, and deficient in phosphorus and nitrogen.

Following various waves of colonization (Roman, Turkish and French) and recent population pressure (51 inhabitants per km² in the mountains), there are now frequent signs of overgrazing (six sheep per hectare) on completely bare mountains. Sheet and rill erosion, gullying and mass movement, displacement of wadi channels and degradation of riverbanks, destruction of roads and accelerated silting-up of reservoirs over the past 15 to 20 years are all signs of advanced and general degradation in this area.

In view of these serious erosion problems, between 1940 and 1970 foresters and rural engineering experts developed a strategy entailing capital investment in rural development (SPR), comprising:

• reforestation of high valleys and mountain tops;

• control of torrents and gullies;

• terracing of cultivated land: in 30 years Algerian terraces were built on more than 300000 hectares (at a cost of between US$ 1000 and 2000/ha).

The main objective was a reduction in the siltation of dams, since suitable sites for reservoirs are limited.

However, by 1977 the failure of SPR in the rural environment was clear: the farmers rejected the terracing system which deprived them of 10 to 20% of their arable land and offered almost no improvement in soil productivity; wood production was still too low; and the siltation rate was still rising! Terracing projects were halted for economic reasons (the second oil crisis) (Heusch 1986). Forestry experts carried on with their work of reforestation and torrent control (RML), but apart from some land improvement projects (subsoiling on brown soil with calcareous crusts), little more was done to stabilize the land farmed by small farmers (Roose 1987a, b).

The first measurements of erosion on runoff plots (Kouidri, Arabi and Roose 1989) confirmed the view that sheet erosion on slopes accounted for only a minor part (0.2 to 1 t/ha/yr) of the sediment load in wadis (Heusch 1970, Demmak 1982). This may explain why the reduction in the siltation rate of reservoirs was still so high even when the slopes had been well reforested or terraced. The wadis are the focal point of the various phenomena, for gullying and wholesale crumbling of slopes sheared away by wandering wadis are the primary source of the sediment carried by rivers during the heaviest flood flows.

Nevertheless, sheet erosion from slopes can be very high - up to 80% of the heaviest rainstorms falling on slaking or sealing crusts or on soil compacted by overgrazing, roads, cattle trails, and fallows left as common grazing land. And it is this runoff from bare slopes that creates gullies, swells very dangerous peak flows, undermines riverbanks, and leads to heavy sedimentation in reservoirs.

In view of the present-day difficulties and slowing down of industry, the Algerian government is promoting a return to the land, with the intensification of mountain farming. However, it hopes that this can be done without hastening montane degradation and siltation of the reservoirs that are so indispensable for irrigation and the constantly expanding towns.

Before farmers can take an interest in maintaining their land and the quality of surface water, it would seem vital to address their immediate concerns - those of increasing their income and security while improving the management of water and nutrients on productive land. The first thing to be done is not to stabilize gullied land, but rather to analyse and improve production systems and the water and mineral balance of the best land. Restoring forests and treating gullied slopes is still the main concern of the forestry department.

This new approach has given rise since 1985 to a "land husbandry aid programme" concerned with research and training, and involving the participation of a dozen research scientists from INRF and ORSTOM. The programme covers three sub-programmes:

• two surveys of the effectiveness of SPR, first by the forestry department in order to discover the most effective interventions, and then by a multidisciplinary research team in order to make a scientific analysis of the reasons for successes and failures;

• the treatment of small catchment areas (20 to 300 ha) near Médéa, Mascara and Tlemcen;

• evaluation of the volume of runoff and erosion with the help of a rainfall simulator and a network of runoff plots and gullies.

Here only the main data from the INRF station at Ouzera (Arab) and Roose 1989) are reported, although similar data have been gathered near Tlemcen (Mazour 1992).

Diagnosis: trial conditions

Fifteen runoff plots (22.2 × 4.5 m) were prepared on farmers' fields around the Ouzera research station, 90 km south of Algiers.

The landscape consists of a series of calcareous uplands (900 to 1200 m in altitude), with steep slopes (12 to 40%) and deep valleys where wadis flow intermittently.

Soil types depend on lithology and topographical position (Pouget 1974, Aubert 1987), and are as follows:

• pale yellow lithosols on grey calcareous colluvial deposits, rich in CaCo₃, but poor in organic matter;

• darkish grey vertisols on marl, well-structured, 2% organic matter, saturated with calcium, pH 7 to 8, very resistant to rainsplash, but prone to gullying and mass movement;

• red fersiallitic soils on soft limestone, poor in organic matter, fragile and lacking stability on the surface;

• brown calcareous soils on colluvial deposits, 2 to 3 % organic matter, a shallow profile, but well-structured on the surface.

In this mountainous region (Blidéen Atlas), the forest cover decreased from 18 to 13% between 1982 and 1991, while vineyards and orchards increased from 2 to 7% and 8 to 14% respectively - a typical effect of population growth and the development of mountain farming. Tillage practices are confined to ploughing twice to control weeds, then a cover crop to dig in fertilizers (N33, P45, K90) and slightly reduce the size of clods.

The average annual rainfall over the past 40 years is 680 mm at Médéa, although between 1986 and 1990, rainfall at Ouzera station (7 km away) varied between 408 and 566 mm, and the rainfall aggressivity factor (Wischmeier's R_USA) was about 46.

The aim of these trials was to compare potential runoff and erosion risks on a bare tilled plot (international control) with those under four production systems (a vineyard, an orchard, an agropastoral system and a sylvopastoral system) on four typical soils in a semi-arid Mediterranean zone with mild winters.

The improvements to the regional control plot consist of rough but carefully oriented tillage, the use of herbicides and pesticides, improved seed, balanced fertilization, a fodder fallow of legumes, and intercropping in rotation under the orchard. The recorded parameters are rainfall (amount, intensity, erosivity), runoff (Kaar, the average annual coefficient, and max KR %, the maximum coefficient for a heavy rainstorm as a percentage of rainfall), erosion (fine suspended matter and coarse sediment), production of biomass and fruit, net income and the soil surface parameters (closed, open and covered surface, surface humidity).

Contents - Previous - Next