Conservation Evidence strives to be as useful to conservationists as possible. Please take our survey to help the team improve our resource.

Providing evidence to improve practice

Action: Crop production: Use no tillage in arable fields Mediterranean Farmland

Key messages

Read our guidance on Key messages before continuing

Crop yield (23 studies)

  • Crops (22 studies): Eight replicated, controlled studies (seven randomized) from Italy and Spain found higher crop yields in plots with no tillage, compared to conventional tillage, in some or all comparisons. Seven replicated, controlled studies (six randomized) from Italy, Lebanon, Spain, and the USA found lower crop yields in plots with no tillage, compared to conventional tillage, in some or all comparisons. Four replicated, randomized controlled studies from Italy and Spain found inconsistent differences in crop yields (sometimes higher with no tillage, sometimes lower). Three replicated, controlled studies (two randomized) from Italy, Portugal, and Spain found similar crop yields in plots with or without tillage.
  • Crop residues (5 studies): Two replicated, randomized, controlled studies from Lebanon and Spain found higher straw yields in plots with no tillage, compared to conventional tillage, in some comparisons. One replicated, randomized, controlled study from Spain found inconsistent straw yields (sometimes higher with no tillage, sometimes lower). Two replicated, controlled studies (one randomized) from Italy and Spain found similar straw yields in plots with or without tillage.

Crop quality (6 studies): One replicated, controlled study from Italy found less protein in wheat grains from plots with no tillage, compared to conventional tillage. One replicated, randomized, controlled study from Spain found heavier cereal grains in plots with no tillage, compared to conventional tillage. Two replicated, randomized, controlled studies from Spain found other differences in crop quality, but two replicated, controlled studies from Italy and the USA did not.

 

Supporting evidence from individual studies

1 

A replicated, controlled study in 1996–1998 in an irrigated tomato field in the San Joaquin Valley, California, USA, found lower tomato yields in plots with no tillage (and winter cover crops), compared to tillage (and winter fallows). Crop yield: Lower tomato yields were found in plots with no tillage, compared to tillage, in four of 16 comparisons (27–36 vs 39–42 tons/acre). Crop quality: Similar amounts of soluble solids were found in tomatoes in the treatment and control plots (data not reported). Methods: There were 12 plots (4.5 x 27.5 m plots) for each of four treatments (two grass-legume mixtures, or two legumes without grasses, as winter cover crops, sown in October 1996–1997, killed and retained as mulch, with no tillage, in March 1997–1998) and each of two controls (bare-soil fallows in winter, with or without herbicide, and conventional tillage in spring). Tomato seedlings were transplanted in April 1997–1998 and harvested in August 1997 and September 1998. The tomatoes were irrigated (two inches/week) and fertilized (0, 100, or 200 lb N/acre). It was not clear whether these results were a direct effect of cover crops or tillage.

 

2 

A replicated, randomized, controlled study in 1983–1996 in a rainfed wheat field in the Henares river valley, Spain, found similar crop yields in plots with no tillage or conventional tillage. Crop yield: Similar wheat yields were found in plots with no tillage or conventional tillage (2.7 vs 2.5 Mg/ha). Methods: No tillage or conventional tillage was used on four plots each. Each plot had two subplots (20 x 30 m, with or without crop rotation). A mouldboard plough (30 cm depth, in autumn) and a tine cultivator (10–15 cm depth, two passes, in spring) were used for conventional tillage. A seed drill and pre-emergence herbicide were used for no tillage. Fertilizer and post-emergence herbicide were used on all plots. Wheat was harvested at maturity (July 1996), and yield was measured in two strips/subplot (1.4 x 30 m strips).

 

3 

A replicated, randomized, controlled study in 1993–2001 in a rainfed cereal field in central Italy found lower crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Lower grain yields were found in plots with no tillage, compared to conventional tillage, in five of seven years (data not reported). Methods: Conventional tillage or no tillage was used on 48 plots each (21 x 11 m sub-sub-plots, in a split-split-plot experimental design), from 1994–2000. A mouldboard plough (30 cm depth, in spring) and a standard precision seed drill were used for conventional tillage. A direct seed drill was used for no tillage. Herbicide and fertilizer were used on all plots. Winter cover crops were grown on three of four plots, and cereal crop residues were retained over winter on one of four plots.

 

4 

A replicated, randomized, controlled study in 1996–1999 in three rainfed barley fields in the Ebro river valley, Spain (same study as (11-13,15)), found that tillage had inconsistent effects on crop yield. Crop yield: Higher grain yields were found in plots with no tillage, compared to conventional tillage, in two of nine comparisons (3,645–5,420 vs 1,557–4,229 kg/ha), but lower grain yields were found in two of nine comparisons (770–1,247 vs 1,672–1,888 kg/ha). Methods: No tillage or conventional tillage was used on 27 plots each (50 x 6 m plots). A mouldboard plough (25–30 cm depth) and a cultivator (15 cm depth, 1–2 passes) were used for conventional tillage, in August–September. Herbicide was used for no tillage. Two-thirds of the plots were fertilized (50–75 or 100–150 kg N/ha). Barley was sown, with a seed drill, in October–November.

 

5 

A replicated, randomized, controlled study in 1993–1997 in a rainfed barley field near Madrid, Spain, found lower crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Lower barley yields were found in plots with no tillage, compared to conventional tillage (1995–1997: 3,593 vs 4,312 kg/ha). Methods: No tillage or conventional tillage was used on four plots each (five 10 x 25 m subplots/plot, with barley monocultures or barley rotations). A mouldboard plough (30 cm depth) and a cultivator (10–15 cm depth, when needed for weed control) were used for conventional tillage. Pre-emergence herbicide was used for no tillage. The barley was fertilized (NPK: 200 kg/ha; ammonium nitrate: 200 kg/ha).

 

6 

A replicated, randomized, controlled study in 2003–2004 in an irrigated field in Davis, California, USA, found lower crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Lower chickpea yields were found in plots with no tillage, compared to conventional tillage, in one of two comparisons (with continuous cropping: 25 vs 193 g dry weight/m2). Methods: No tillage or conventional tillage was used on six plots each (67 x 4.7 m plots, three beds/plot). All plots had Cicer arietinum chickpeas (garbanzo beans) in rotation with other crops. Crop residues were incorporated to 20 cm depth, and the beds were shaped, on plots with conventional tillage (disk, lister, and ring roller). Crop residues were flail mown and spread on the plots with no tillage. All plots were fertilized in 2003, but not thereafter (112 kg P/ha phosphorous, 50 kg NPK/ha, and 67 kg N/ha). Cultivation was used to control weeds on plots with conventional tillage. Hand weeding was used on plots with no tillage. Herbicide was used on all plots. Some plots were irrigated. Chickpeas were harvested on 28 June 2004.

 

7 

A replicated, randomized, controlled study in 2002–2005 on three rainfed farms in the Ebro river valley, Spain, found lower crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Lower barley yields were found in plots with no tillage, compared to conventional tillage, in four of 10 comparisons (730–3,083 vs 1,314–3,514 kg/ha). Methods: No tillage or conventional tillage was used on ten plots each (Peñaflor: three plots each, 33 x 10 m plots; Agramunt: four plots each, 9 x 50 m plots; Selvanera: three plots each, 7 x 50 m plots). In Peñaflor, a mouldboard plough (30–40 cm depth) and a cultivator (10–15 cm depth) were used for conventional tillage. In Agramunt, a mouldboard plough (25–30 cm depth) and a cultivator (15 cm depth) were used for conventional tillage. In Selvanera, a subsoil plough (40 cm depth) and a cultivator (15 cm depth) were used for conventional tillage. Herbicide and a seed drill were used for no tillage.

 

8 

A replicated, randomized, controlled study in 2005–2007 in a rainfed field in the central Bekaa Valley, Lebanon, found lower seed yields, but higher straw yields, in plots with no tillage, compared to conventional tillage. Crop yield: Lower seed yields were found in plots with no tillage, compared to conventional tillage, in one of three crops (barley seeds: 3,550 vs 4,550 kg/ha), but higher straw yields were found in one of three crops (safflower straw: 9,950 vs 9,050). Methods: No tillage or conventional tillage was used on four plots each (14 x 6 m), in October. Conventional plots were ploughed (25–30 cm depth) and then shallowly disk-cultivated. Barley, chickpeas, and safflower were planted in November. Barley and safflower were fertilized (60–100 kg N/ha). Mature crops were collected in three quadrats/plot (0.25 m2 quadrats).

 

9 

A replicated, randomized, controlled study in 2003–2005 in an occasionally irrigated oat field in Portugal found similar crop yields in plots with or without tillage. Crop yield: Similar oat yields were found in plots with or without tillage (4.2 t/ha). Methods: Tillage or no tillage was used on four plots each (400 m2 plots). A disk plough was used for tillage (two passes, 15 cm depth). The plots were intercropped with oats and Lupinus albus lupins in 2003–2004 (residues were retained, and incorporated into the soil in the plots with tillage) and oats were grown in monoculture in 2004–2005. The plots were fertilized in 2003–2004 (60 kg P/ha; 100 kg N/ha), but not in 2004–2005.

 

10 

A replicated, randomized, controlled study in 1986–2008 in a rainfed wheat field in southern Spain found that tillage had inconsistent effects on crop yields. Crop yield: Higher wheat yields were found in plots with no tillage, compared to conventional tillage, in two of five comparisons (3,482–4,698 vs 2,463–3,926 kg/ha), but lower yields were found in one of five comparisons (4,571 vs 5,216 kg/ha). Methods: No tillage or conventional tillage was used on three plots each (five subplots/plot, 10 x 5 m subplots, with different wheat rotations). Mouldboard ploughing, disk harrowing, and/or vibrating tine cultivation was used for conventional tillage (depth not reported). Pre-emergence herbicide was used for no tillage. The wheat phase was fertilized with nitrogen in some sub-subplots (0–150 kg N/ha/year) and phosphorus in all plots (65 kg P/ha/year). Crop residues were retained. Wheat yields were measured in 2008.

 

11 

A replicated, randomized, controlled study in 1996–2009 in a rainfed barley field in the Ebro river valley, Spain (same study as (4,12,13,15)), found that tillage had inconsistent effects on the yield of barley straw. Crop yield: More barley straw was found in plots with no tillage, compared to conventional tillage, in one of six comparisons (511 vs 242 g/m2), but less barley straw was found in one of six comparisons (332 vs 541). In one of three years, the barley crop failed with conventional tillage, but not with no tillage. Methods: No tillage or conventional tillage was used on nine plots each (50 x 6 m plots). A mouldboard plough was used for conventional tillage (25–30 cm depth, 100% incorporation of crop residues). A seed drill and herbicide were used for no tillage. Two-thirds of the plots were fertilized (60 or 120 kg N/ha). Mature barley was harvested in June 2006–2009 (three samples/plot, 50 cm of one row/sample).

 

12 

A replicated, randomized, controlled study in 1996–2009 in a rainfed barley field in the Ebro river valley, Spain (same study as (4,11,13,15)), found higher crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Higher barley yields were found in plots with no tillage, compared to conventional tillage, in six of nine comparisons (1,350–2,500 vs 300–700 kg/ha). Methods: No tillage or conventional tillage was used on nine plots each (50 x 6 m plots). A mouldboard plough was used for conventional tillage (25–30 cm depth, 100% incorporation of crop residues), in October or November. A seed drill and herbicide were used for no tillage. Two-thirds of the plots were fertilized (60 or 120 kg N/ha). Mature barley was harvested in June 2006–2009.

 

13 

A replicated, randomized, controlled study in 1996–2009 in a rainfed barley field in the Ebro river valley, Spain (same study as (4,11,12,15)), found higher barley yields in plots with no tillage, compared to conventional tillage. Crop yield: Higher barley yields were found in plots with no tillage, compared to conventional tillage (2,062 vs 1,155 kg/ha). Methods: No tillage or conventional tillage was used on nine plots each (50 x 6 m plots). A mouldboard plough was used for conventional tillage (25–30 cm depth, 100% incorporation of crop residues), in October or November. A seed drill and herbicide were used for no tillage. Two-thirds of the plots were fertilized (60 or 120 kg N/ha). Mature barley was harvested in June 2006–2009.

 

14 

A replicated, controlled study in 1991–2009 in a rainfed faba bean field in Sicily, Italy, found higher crop yields, but no differences in crop quality, in plots with no tillage, compared to conventional tillage. Crop yield: Higher faba bean yields were found in plots with no tillage, compared to conventional tillage (2.36 vs 1.80 Mg grain/ha), but no differences in straw yields were found (3.93 vs 3.86 Mg straw/ha). Crop quality: No differences were found in faba bean seed weight (99 vs 98 g/100 seeds), or the number of seeds/pod (2.7 vs 2.6), in plots with no tillage, compared to conventional tillage. Methods: No tillage or conventional tillage was used on two plots each (18.5 x 20 m plots). A mouldboard plough (30 cm depth; in summer) and a harrow (depth not reported; before sowing) were used for conventional tillage. Herbicide (before sowing) and a seed drill were used for no tillage. In all plots, a hoe was used to control weeds (depth not reported; 1–2 times/year). Faba beans were grown in rotation with durum wheat. During durum wheat growth, herbicide was used in all plots. All plots were fertilized (46 kg P2O5/ha). Faba beans were sown in December and harvested at maturity (month not reported). Yield and quality were measured in three samples/plot (four rows/sample, 3 m rows).

 

15 

A replicated, randomized, controlled study in 1996–2008 in a rainfed barley field in the Ebro river Valley, Spain (same study as (4,11-13)), found more barley straw in plots with no tillage, compared to conventional tillage, in two of three comparisons. Crop yield: More barley straw was found in plots with no tillage, compared to conventional tillage, in two of three comparisons (2,083–2,265 vs 1,571–1,748 kg/ha). Methods: There were nine plots (50 x 6 m) for each of two tillage treatments (no tillage: pre-emergence herbicide; conventional tillage: mouldboard plough, 25–30 cm depth). Plots were tilled in October or November. Two-thirds of the plots were fertilized (60 or 120 kg N/ha/year). Barley was harvested in June.

 

16 

A replicated, controlled study in 1991–2009 in a rainfed wheat field in Sicily, Italy, found less protein in wheat that was grown on plots with no tillage, compared to conventional tillage. Crop yield: Similar wheat yields were found in plots with no tillage or conventional tillage (4 Mg/ha). Crop quality: Less protein was found in wheat that was grown in plots with no tillage, compared to conventional tillage (135 vs 144 g/kg). Methods: No tillage or conventional tillage was used on six plots each (19 x 20 m subplots). Conventional tillage was mouldboard ploughing in summer (30 cm depth) and harrowing before planting (two passes). No tillage was direct drilling and pre-emergence herbicide. The plots had either faba bean-wheat, clover-wheat, or wheat-wheat rotations. Fertilizer and post-emergence herbicide were used on all plots. Yield was measured in three samples/plot/year (8.6 x 8.6 m), in 1996–2009.

 

17 

A replicated, randomized, controlled study in 2009–2011 in two irrigated pepper fields in central Italy found that tillage had inconsistent effects on crop yields. Crop yield: Higher pepper yields were found in plots with no tillage, compared to conventional tillage, in five of eight comparisons (14–38 vs 6–20 t/ha, fresh weight), but lower yields were found in one of eight comparisons (21 vs 26). Methods: A mouldboard plough (30 cm depth) was used on all plots in autumn, before winter cover crops were planted. Cover crops were mown or chopped in spring, before tillage. No tillage or conventional tillage was used on 12 plots each (6 x 12 m plots), in May 2010–2011. A mouldboard plough (30 cm depth) and a disk (two passes) were used for conventional tillage (which incorporated the cover crop residues into the soil). Cover crop residues were mulched and herbicide was used for no tillage. Pepper seedlings were transplanted into the plots in May, and fruits were harvested twice/year in August–October 2010–2011. All plots were fertilized before the cover crops, but not after. All plots were irrigated.

 

18 

A replicated, randomized, controlled study in 2009–2012 in two irrigated vegetable fields in central Italy found higher crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Higher crop yields were found in plots with no tillage, compared to conventional tillage, in two of six comparisons (in plots with hairy vetch as a winter cover crop: 23 vs 17 t/ha endive; 23 vs 15 t/ha savoy cabbage; fresh weights). Methods: No tillage or conventional tillage was used on nine plots each (6 x 4 m plots). Each plot had a winter cover crop (hairy vetch, oats, or oilseed rape). Cover crops were sown in September 2009–2010 and suppressed in May 2010–2011. A mouldboard plough and a disk harrow (two passes) were used for conventional tillage (incorporating the cover crop residues to 30 cm depth). The cover crop residues were gathered into strips of mulch (50 cm wide, along crop rows) in plots with no tillage. Pepper seedlings were transplanted into these plots in May 2010–2011 and were last harvested in October 2010 and September 2011. After the pepper harvest, endive and savoy cabbage seedlings were transplanted into these plots, and they were harvested in December 2010 and November 2011 (endive) or March 2011 and February 2012 (cabbage). No fertilizer was added while the crops were growing, but the plots were irrigated. It was not clear whether these results were a direct effect of tillage or mulch.

 

19 

A replicated, randomized, controlled study in 2008–2013 in a rainfed wheat-sunflower-pea field near Seville, Spain, found lower crop yields, and differences in crop quality, in plots with no tillage, compared to conventional tillage. Crop yield: Lower crop yields were found in plots with no tillage, compared to conventional tillage, in two of five comparisons (sunflower seeds in 2013: 105 vs 3,520 kg/ha; wheat grain in 2012: 2,940 vs 3,860 kg/ha). Crop quality: Less oil, less of nine nutrients, more saturated and monounsaturated fatty acid, and less polyunsaturated fatty acid were found in sunflower seeds in plots with no tillage, compared to conventional tillage (34% vs 48% oil; see publication for other results). Methods: No tillage or conventional tillage was used on three plots each (6 x 33.5 m plots). A mouldboard plough (25–30 cm depth), a chisel plough (25 cm depth, twice/year), and a disk harrow (12 cm depth) were used for conventional tillage. A seed drill and herbicide were used for no tillage. Wheat, sunflowers, and peas were grown in rotation. Wheat was fertilized, but sunflowers and peas were not. Sunflowers were sown in May 2013 (three months later than usual) and harvested in September. Yield and quality were measured in 16 sunflower heads/plot.

 

20 

A replicated, randomized, controlled study in 1987–2010 in rainfed cereal fields in the Ebro river valley, Spain, found higher crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Higher grain yields were found in plots with no tillage, compared to conventional tillage (4,449 vs 4,210 kg/ha). Crop quality: Heavier grains were found in plots with no tillage, compared to conventional tillage (35 vs 32 mg/grain). Methods: No tillage or conventional tillage was used on ten plot each (Peñalba: three plots each, 34 x 175 m plots, established in 2005; Agramunt: four plots each, 9 x 50 m plots, established in 1990; Selvanera: three plots each, 7 x 50 m plots, established in 1987). In Peñalba, a disk plough (20 cm depth) and a cultivator (10 cm depth) were used for conventional tillage. In Agramunt, a mouldboard plough (25 cm depth) and a cultivator (15 cm depth) were used for conventional tillage. In Selvanera, a subsoil plough (40 cm depth) and a chisel plough (15 cm depth) were used for conventional tillage. Herbicide was used for no tillage. Barley (Peñalba) or wheat (Agramunt and Selvanera) was planted in November 2009 with a seed drill (2–4 cm depth) and harvested in June–July 2010.

 

21 

A replicated, randomized, controlled study in 1996–2013 in two rainfed barley fields in northeast Spain found higher crop yields in plots with no tillage, compared to conventional tillage. Crop yield: Higher barley yields were found in plots with no tillage, compared to conventional tillage (1,554–5,692 vs 246–2,263 kg/ha). Methods: No tillage or conventional tillage was used on three plots each, in each of two fields (from 2010–2013 in the short-term field, and from 1996–2013 in the long-term field; plots size not clearly reported). A mouldboard plough (25 cm depth) and a cultivator (15 cm depth) were used for conventional tillage in the long-term field, and a chisel plough was used in the short-term field (depth not reported), in September–October. For no tillage, the residues were chopped and spread, and pre-emergence herbicide was used. Some plots were fertilized (0–150 kg N/ha). Crop yield was measured in 2011 (short-term field) and 2011–2013 (long-term field; the reported yield was the sum of three years).

 

22 

A replicated, randomized, controlled study in 1994–2009 in a pea field near Madrid, Spain, found higher pea yields and different crop qualities in plots with no tillage, compared to conventional tillage. Crop yield: Higher pea yields were found in plots with no tillage, compared to conventional tillage, in one of four comparisons (0.6 vs 0.3 t/ha), but no differences in straw yields were found (2.6–3.5 vs 2.3–3.3 t/ha). Crop quality: Smaller peas were found in plots with no tillage, compared to conventional tillage, in one of four comparisons (238 vs 267 g/1,000 peas). Longer pea pods were found in plots with no tillage, compared to conventional tillage, in one of four comparisons (7.0 vs 6.5 cm), but shorter pods, with fewer peas, were found in one of four comparisons (5.9 vs 6.6 cm pods, 5.4 vs 6.4 peas/pod). Methods: No tillage or conventional tillage was used on four plots each (each with three 10 x 25 m sub-plots, with different pea-cereal rotations), in October or November. A mouldboard plough was used for conventional tillage (30 cm depth). A seed drill and herbicide were used for no tillage. Peas were planted in November 2005–2008 and harvested in June 2006–2009. The peas were not fertilized.

 

23 

A replicated, randomized, controlled study in 2009–2011 in an irrigated eggplant field in central Italy found that tillage had inconsistent effects on crop yield. Crop yield: Higher eggplant yields were found in plots with no tillage, compared to conventional tillage, in two of four comparisons (18–38 vs 14–32 Mg/ha fresh weight), but lower yields were found in one of four comparisons (11 vs 18). Methods: A mouldboard plough (30 cm depth) was used on all plots in autumn, before the winter cover crops were planted. The cover crops were mown or chopped in spring, before tillage. No tillage or conventional tillage was used on 12 plots each (6 x 4 m plots). A mouldboard plough (30 cm depth) and a disk (two passes) were used for conventional tillage (which incorporated the cover crop residues into the soil). Cover crop residues were mulched and herbicide was used for no tillage. Eggplant seedlings were transplanted into the plots in May, and fruits were harvested four times/year in July–September 2010–2011. All plots were fertilized before the cover crops were grown, but not after. All plots were irrigated.

 

Referenced papers

Please cite as:

Shackelford, G. E., Kelsey, R., Robertson, R. J., Williams, D. R. & Dicks, L. V. (2017) Sustainable Agriculture in California and Mediterranean Climates: Evidence for the effects of selected interventions. Synopses of Conservation Evidence Series. University of Cambridge, Cambridge, UK.