Introduction on agriculture in Tunisia (Thabet et al., 1994)

Physical Setting

Located in North Africa, Tunisia is bound on the north and east by the Mediterranean sea, on the south by Libya and on the west by Algeria. Of all North African countries, Tunisia has the longest Mediterranean coast line, giving it a temperate climate and sizeable fishery reserves in addition to the tourist attractions which annually draw a minimum of one million foreign visitors.

Tunisia covers about 63,170 square miles, corresponding to approximately 16 million hectares. This total area consists of 30% arable land, 27% pasture and forests, and approximately 43% agriculturally unusable land. This means that only about half the country contributes to agricultural production.

With regard to climate, rainfall is a major, though variable, factor, ranging from an average of less than 100 millimeters a year in the south, to over 1000 millimeters a year in the extreme north of the country. In the northern part, the topography is more and more sloped, leaving relatively little cultivable land in areas of relatively high rainfall. The southern part is suggesting a negative correlation between the availability of arable land and that of water. This implies that agricultural activity is undertaken, for the most part, under limited and highly variable rainfall. Temperature and wind are no less variable, frequently causing serious damage to agricultural crops. The country is generally divided into the following naturally homogeneous subregions: northeastern, northwestern, east central, west central and southern regions.

Major Crops

The division of cultivated agricultural land by major crops shows the importance of grains (wheat and barley) and tree crops which together cover around 87% of the total: grains with 43% and tree crops with 44% (mainly olives). The remaining 17% is allocated between forage crops (7%), vegetable crops (3%), legumes (2.5%) and others (0.5%).

The share of grains in total cultivated agricultural land varies significantly in relation to the amount of rainfall recorded during the months of October and November. On average, however, grains cover 1.7 million hectares of land. The regional distribution of agricultural land cultivated under grains is as follows: 47% in the northern part of the country and 53% in the central and southern parts. The area under grain cultivation is relatively stable in the northern part while it varies greatly in the central and southern parts. This variability is strongly correlated with that of rainfall.

Tree crops on average cover 1.9 million hectares (44% of total cultivated agricultural land), out of which 1.4 million hectares are under olive cultivation. Other plantations are mainly almond, pistachio, date, apple and pear trees, and grapevines. Eighty-seven percent of the total area under tree crop cultivation is located in the center and south, and is cultivated under essentially rain-fed conditions.

Soils (Mtimet, 2001)

Since Tunisia is at the same time a Mediterranean and a Saharan country the soils show all the signs of this climatic, morphological and geological diversity. According to the French system of soil classification the soils of Tunisia are classified as podzols, vertisoils, red Mediterranean soils, calcic-magnesic soils (dominant soils), brown and isohumic soils, saline and hydromorphic soils and also poorly evolved soils. Their distribution is based on the catena system.

Major Agricultural Zones (Tunisian Ministry of Environment, 2015)

Bioclimatic diversity, geological and morphological diversity combined with diversified soils (natural vegetation, rainfed crops and irrigated crops) is at the origin for the existence of a mosaic of genetically different soils. These soils are faced with convergent natural factors (soft rock, steep slopes, sharp showers, sparse vegetation cover) that are responsible for the state of degradation. This is essentially the water and wind erosion, and salinization. Three main regions are distinguished by the nature of their soils and modes of exploitation of their land.

Northern Tunisia

It is divided into two parts, the North West and North East:

• The first, which is an agro-forest-pastoral region, is distinguished by hydromorphic soils and acid brown characteristic of the Kroumirie Mogods mountain and Calcimagnesian (Rendzinas and brown limy types of soil) covering the slopes of the glacis and the Tell vertisols; and associated with slightly evolved soils with more or less hydromorphic alluvial aspect , forming floodplains (The Upper Valley of Medjerda.).
• The second part is the north east, land has very diverse Rendzines: red soil, brown soil, forming a mosaic of soils occupying the glacis and the slightly developed and quite healthy slopes and soils on the plains. Holomorphic soils are confined to depressions and "Garâas". There are also the eroded raw mineral soils associated with brown soils covering the most prominent slopes (Jebel Abderrahmane).

Central Tunisia

It is an agro-pastoral region, dominated equally by:

• The heavy soils of the alluvial plains much of which is halomorphic (Lower steppe);
• Encrusted limy and skeletal soils of the great glacis of the High Steppe (alfa), 
• The deep and light soils of the Gammouda / Meknassy region where there was a rich course of the past that is almost completely converted to arboriculture (olive, almond, etc.)..

Southern Tunisia

It is a pastoral region distinguished by the presence of numerous oases around the water points. It is characterized by:

• Slightly eroded hills, except for the Matmatas chain where anthropogenic soils are created behind dams called "Jessours" established through the many ravines whereby water and soil are collected and used intensively in arboriculture (olive trees);
• The coastal plains (the Jeffara) where we find encrusted glacis upstream and sierozems and slightly developed soils downstream. Depressed parts are occupied by gypseous and halomorphic soils.
• The large depressions or "Chott" are occupied by very salty and barren soil. Around them we find wind ridges (Dhraa), formed by deep sandy soils and occupied by the best oasis (Tozeur, Degache);
• The desert area, formed by Erg (succession of sand dunes) and Dhar where the soil is completely bare and stony (Reg).

Climate (Kayouli, 2000)

The Tunisian climate is Mediterranean, characterised by hot dry summers and cool moist winters that limit the growing period; precipitation is very irregular and the rainfall varies considerably from the North to South. Tunisia is divided into four large geographical units: Northern, Eastern, Central and Southern regions.

According to Emberger (1960) there are five bioclimatic zones in going from the most arid to the most humid based on rainfall. Rainfall is not the only bioclimatic determinant; temperature and especially winter temperature is also important. This is not only governed by altitude but by the degree of continentality; inland stations have relatively hotter summers and colder winter than areas which benefit from the buffering effects of the sea. Bioclimatically, therefore, the country is also divided into areas of warm, cool and cold winters.

 

The relief (Tunisian Ministry of Environment, 2015)

Tunisia North-West is located in the extension of the Atlas mountain range which starts in southern Morocco and is divided into two rows facing east-west: the Tell Atlas following the Mediterranean coast and the Saharan Atlas fading arriving in Cap Bon and the Gulf of Hammamet to the sea.

Between the north coast and the valley of the Medjerda the Northern Tell is in the form of three lesser high rows, reaching the eastern coastline between Cape Blanc and Ghar El Melh, which are: the mountains of Khroumirie (rising to 1000 m), Nefza Mountains (rising to 600 m) and Mogods (rising to 500 m).

South of these mountains, the valley of the Medjerda is served by several wadis (Mellègue, Tessa, Beja, Zarga), then follows the hilly area of the Monts de Téboursouk between the city of El Kef and the Gulf of Tunis: the Higher Tell.Further south, the Tunisian Ridge stretches from west to east, from the Mountains of Tebessa to the Algerian border, to Cap Bon on the eastern coastline. It consists of mountain ranges alternating with steep plateaus and depressions: Djebel Chaambi (1544 m), Jebel Semmam (1314 m), Djebel Serj (1347 m), Jebel Zaghouan (1295 m), Djebel Sidi Abderrahmane in the Cap Bon (637 m).

Further south the Ridge, the Saharan Atlas is reduced to a few units scattered throughout the mountainous high steppes: Djebel Mghilla (1378 m), Jebel Selloum (1373 m).The region of high plains to the west and low plains to the east is crossed from west to east by a few isolated mountains: Djebel. Majoura (874 m), Djebel. Bouhedma (790 m), Djebel. Orbata (1165 m), Djebel. Asker (608 m).

South of Gafsa major depression Chotts mark the beginning of the horizontal part of the Sahara. South of the Chott to the ends of Dahar, the Grand Erg Oriental is extended. The Dahar Mountains, the plains of Jeffara and El Ouara complete the landscape of southern Tunisia near the eastern Mediterranean with the island of Djerba.

Modelling yield potential, water limited yield potential and irrigated yield potential

For each crop in each country, its harvest area of recent years is classified into different climate zones based on three climatic indices: aridity, seasonality, and length of growing season (van Wart et al, 2013). Crop yields within the same climate zone are assumed to be comparable and similarly attainable for a given soil type. Yield simulations were performed at key locations where observed weather data are available (called reference weather stations or RWS) and crop harvest area is significant (You et al., 2009).  RWS were selected iteratively based on harvested area within a 100 km buffer zone (but using 50 km buffer zone for the smaller country of Jordan) until 50% of harvested area was within the buffer zone of all selected stations. This initial RWS selection was reviewed by country experts and stations were added or subtracted based on their expertise. Dominant soil textures and crop management information (sowing date, plant population, cultivar maturity) at district levels were collected in the area surrounding each RWS (Table 1). ISRIC-WISE soil data were used to determine dominant soil textures where observed data or expert opinion on dominant soils were lacking. Weather data were collected from national weather databases and then run through quality control measures as described in van Wart et al. (2015) and in the GYGA protocols.

The WOFOST crop simulation model (Diepen et al, 1989) was used to estimate the Yp, Yw and Ys (yield potential, water-limited yield potential and supplementally irrigated yield potential, respectively) of barley, potato and wheat for each water regime (i.e., rainfed, supplemental irrigation, or full irrigation) for major soil types in the area around the RWS. For wheat simulations, the WOFOST model does not contain a default spring wheat crop data file. Therefore, a default sprint wheat crop data file was created based on growth parameters as reported in Belhouchette et al., 2008 and as specified in the WOFOST default barley crop data file (Table 2). The temperature sum from emergence to anthesis (TSUM1), temperature sum from anthesis to physiological maturity (TSUM2) and the initial total crop dry weight (TDWI; indicative of sowing density) were adjusted so that simulated harvested index (HI) of Yp was close to reported average of 0.45, reported maximum leaf area index was within reported range of between 3-5, and crop matured around the average time farmers' crops were reportedly reaching physiological maturity.  For barley, WOFOST's default crop data file was used, and TSUM1, TSUM2 and TDWI were adjusted in the same way as for wheat described above. Lastly, a soil file was developed for each major soil texture identified at each site and used in the WOFOST model (Table 3). Current average amount and timing of supplemental irrigation was reported by country experts and applied to simulations, except for Jordan, where supplemental irrigation data were unavailable (see Table 4)

Table 1 Site management and soil parameters used in the WOFOST model, years simulated and reported rainfed and irrigated yields at each site for barley (A) and wheat (B)

(A) Site	Country	Sowing date	Growth duration (sowing to harvest)	Major soil texture around site	Years Simulated
Mednine	Tunesia	15-Nov	210	SANDY CLAY	90' - 12'
SidiBouzid	Tunesia	15-Nov	210	CLAY LOAM	90' - 12'
SidiBouzid	Tunesia	15-Nov	210	SANDY CLAY LOAM	90' - 12'
Monastir	Tunesia	15-Nov	210	CLAY LOAM	90' - 12'
Monastir	Tunesia	15-Nov	210	LOAM	90' - 12'
Beja	Tunesia	15-Nov	210	CLAY LOAM	90' - 12'
EbbaKsour	Tunesia	15-Nov	210	CLAY LOAM	90' - 12'
Mahdia_djem	Tunesia	15-Nov	210	CLAY LOAM	00' - 14'
Mahdia_djem	Tunesia	15-Nov	210	SANDY CLAY LOAM	00' - 14'
Mednine	Tunesia	1-Nov	210	SANDY CLAY LOAM	90' - 12'
Siliana	Tunesia	15-Nov	210	CLAY LOAM	90' - 12'
Nabeul	Tunesia	15-Nov	210	CLAY LOAM	00' - 14'

 

 

(B) Site	Country	Sowing date	Growth duration (sowing to harvest)	Major soil texture around site	Years Simulated
Kairouan	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
Kairouan	Tunesia	1-Dec	210	LOAM	90' - 12'
Nabeul	Tunesia	1-Dec	210	CLAY LOAM	00' - 14'
SidiBouzid	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
SidiBouzid	Tunesia	1-Dec	210	SANDY CLAY LOAM	90' - 12'
TunisCarthage	Tunesia	1-Dec	210	CLAY LOAM	00' - 14'
Monastir	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
Monastir	Tunesia	1-Dec	210	LOAM	90' - 12'
Beja	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
Beja	Tunesia	1-Dec	210	LOAM	90' - 12'
EbbaKsour	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
EbbaKsour	Tunesia	1-Dec	210	LOAM	90' - 12'
Mahdia_djem	Tunesia	1-Dec	210	SANDY CLAY LOAM	00' - 14'
Mednine	Tunesia	1-Dec	210	SANDY CLAY	90' - 12'
Siliana	Tunesia	1-Dec	210	CLAY LOAM	90' - 12'
Beja	Tunesia	15-Nov	230	CLAY LOAM	90' - 12'
EbbaKsour	Tunesia	15-Nov	230	CLAY LOAM	90' - 12'
Monastir	Tunesia	15-Nov	230	LOAM	90' - 12'
Kairouan	Tunesia	15-Nov	230	CLAY LOAM	90' - 12'
Nabeul	Tunesia	15-Nov	230	CLAY LOAM	00' - 14'
SidiBouzid	Tunesia	15-Nov	230	CLAY LOAM	90' - 12'
TunisCarthage	Tunesia	15-Nov	230	CLAY LOAM	00' - 14'

 

Table 3 Soil textural parameters used in WOFOST model simulations

Soil texture	SMW*	SMFCF**	SM0***
Loamy sand	0.020	0.174	0.440
Sandy loam	0.069	0.252	0.460
Silt loam	0.090	0.279	0.510
Loam	0.093	0.278	0.500
Sandy clay loam	0.175	0.328	0.430
Silt clay loam	0.168	0.320	0.450
Clay loam	0.261	0.373	0.450
Clay	0.364	0.498	0.540
Silty clay	0.277	0.462	0.510

 

 

 

 

 

 

 

^*Soil moisture content at wilting point (cm³ cm^-3)

^**Soil moisture content at field capacity (cm³ cm^-3)

^***Soil moisture content at saturation (cm³ cm^-3)

Parameters changed from defaults	Wheat file parameter	Source
1). TSUM from sowing to emergence	100	Belhouchette et al., 2008
2). Max temperature for growth	35	Default WOFOST barley crop file
3). initial total crop dry weight [kg ha-1]	120	Default WOFOST barley crop file
4). leaf area index at emergence [ha ha-1]	0.274	Default WOFOST barley crop file
5). maximum relative increase in LAI [ha ha-1 d-1]	0.0075	Default WOFOST barley crop file
6). life span of leaves growing at 35 Celsius [d]	25	Default WOFOST barley crop file
7).extinction coefficient for diffuse visible light	0.48	Belhouchette et al., 2008
8). Correction factor transpiration rate	1.1	Belhouchette et al., 2008
9). Maximum rooting depth[cm]	150	Belhouchette et al., 2008

 

 

Table 2 Parameters changed from defaults in the default WOFOST crop parameter file and source of changed parameter value

 

Table 4 Supplemental irrigation applied to simulations of wheat in Morocco and Tunisia (mm)

Site	Sowing	Heading	Anthesis	Grain filling
	Tunisia
Beja			60
Nabeul			60
Siliana			60
Tunis-Carthage			60
EbbaKsour	90		90
Kairouan	90		90
MahdiaDjem	90		90
Mednine	90		90
Monastir	90		90
SidiBouzid	90		90

 

References

Belhouchette, H., Braudeau, E., Hachicha, M., Donatelli, M., Mohtar, R. H., and Wery, J. 2008. Integrating spatial soil organization data with a regional agricultural management simulation model: a case study in northern Tunisia. Transactions of the ASABE 51:1099-1109.

Diepen, C.A. van, Wolf, J., Keulen, H. van, and Rappoldt, C. 1989. WOFOST: a simulation model of crop production. Soil Use and Management. 5(16-24)

Kayouli, Chedly. 2000. Grassland and Pasture Crops Country Pasture/Forage Resrouce Profiles Tunisia. Accessed 15 September 2015 at http://www.fao.org/countryprofiles/index/en/?iso3=TUN

Mtimet A., 2001. Soils of Tunisia. In : Zdruli P. (ed.), Steduto P. (ed.), Lacirignola C. (ed.), Montanarella L.

(ed.). Soil resources of Southern and Eastern Mediterranean countries. Bari : CIHEAM, 2001. p. 243-262

(Options Méditerranéennes : Série B. Etudes et Recherches; n. 34)

Thabet B., Boughzala M., Ben Ammar B. 1994. Agriculture and food policy in Tunisia. In : Allaya M.

(comp.), Thabet B. (comp.), Allaya M. (collab.), Thabet B. (collab.). Food and agricultural policies in the

Middle East and North Africa: Egypt, Lebanon, Morocco, Sudan, Tunisia, Turkey. Montpellier : CIHEAM,

1994. p. 181-220 (Cahiers Options Méditerranéennes; n. 7)

Tunisian Ministry of the Environment. 2015. Country Environmental Profile. Accessed 15 September 2015 at http://www.environnement.gov.tn/index.php?id=98&L=1#.VfhFCxFVhBd

Van Wart, J., van Bussel, L.G., Wolf, J., Licker, R., Grassini, P., Nelson, A., Boogaard, H., Gerber, J., Mueller, N.D., Claessens, L., van Ittersum, M.K., and Cassman, K.G. 2013. Use of agro-climatic zones to upscale simulated crop yield potential. Field Crops Research. 143(44-55).

Van Wart, J., P. Grassini, H.S. Yang, L. Claessens, A. Jarvis, K.G. Cassman. 2015. Creating long-term weather data from the thin air for crop simulation modelling Agric. For. Meteorol http://dx.doi.org/10.1016/j.agrformet.2015.02.020 (In Press)

You, L.,Wood, S., Wood-Sichra, U. 2009. Generating plausible crop distribution maps for Sub-Saharan Africa using a spatially disaggregated data fusion and optimization approach. Agric. Syst., 99(126–140)