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: Soil: Use crop rotations Mediterranean Farmland

Key messages

Organic matter (9 studies): Five replicated, controlled studies (two randomized) from Italy, Portugal, and Spain found less organic matter in soils with crop rotations, compared to continuous crops, in some comparisons. One replicated, controlled study from Syria found more organic matter in soils with crop rotations, compared to continuous crops, in some comparisons. Three replicated, controlled studies from Spain found similar amounts of organic matter in soils with or without crop rotations.

Nutrients (5 studies)

  • Nitrogen (5 studies): One replicated, randomized, controlled study from Australia found more nitrogen in soils with crop rotations, compared to continuous crops, in one of four comparisons. One replicated, controlled study from Italy found less nitrogen in soils with crop rotations, compared to continuous crops, in some comparisons. One replicated, randomized, controlled study from Spain found inconsistent differences in nitrogen in soils with or without crop rotations. Two replicated, randomized, controlled studies from Portugal and Spain found similar amounts of nitrogen in soils with or without crop rotations.
  • Phosphorus (2 studies): Two replicated, randomized, controlled studies from Portugal and Spain found less phosphorus in soils with crop rotations, compared to continuous crops, in some comparisons.
  • pH (2 studies): Two replicated, randomized, controlled studies from Portugal and Spain found similar pH levels in soils with or without crop rotations.

Soil organisms (3 studies)

  • Microbial biomass (2 studies): One replicated, controlled study from Italy found more microbial biomass in soils with crop rotations, compared to continuous crops, in some comparisons. One replicated, randomized, controlled study from Spain found less microbial biomass in soils with crop rotations, compared to continuous crops, in some comparisons.
  • Bacteria and fungi (1 study): One replicated, randomized, controlled study from Portugal found more fungi, but similar amounts of bacteria, in soils with crop rotations, compared to continuous crops.

Soil erosion and aggregation (4 studies): One replicated, controlled study from Syria found higher water-stability in soils with crop rotations, compared to continuous crops. One replicated, randomized, controlled study from Spain found lower water-stability in soils with crop rotations, compared to continuous crops, in some comparisons. One replicated, randomized, controlled study from Spain found inconsistent differences in water-stability in soils with or without crop rotations. One replicated, randomized, controlled study from Spain found no differences in water-stability.

Greenhouse gases (4 studies): One replicated, controlled study from Italy found higher carbon dioxide emissions from soils with crop rotations, compared to continuous crops, in some comparisons. One replicated, randomized, controlled study from Spain found similar carbon dioxide emissions from soils with or without crop rotations. One replicated, randomized, controlled study from Australia found lower nitrous oxide and methane emissions from soils with crop rotations, compared to continuous crops, but another one found no differences in nitrous oxide emissions.

Implementation options (2 studies): Two studies from Syria and the USA found similar amounts of nitrogen in soils with two-year or four-year rotations. One of these studies also found similar amounts of organic matter.

Supporting evidence from individual studies

1 

A replicated, randomized, controlled study in 1983–1996 in a rainfed wheat field in the Henares river valley, Spain, found lower stability in soils with crop rotations, compared to soils without crop rotations, in some comparisons. Organic matter: Similar amounts of organic carbon were found in soils with or without crop rotations (42.3 vs 40.4 Mg C/ha). Soil erosion and aggregation: Lower soil stability was found in soils with crop rotations, compared to soils without crop rotations, in three of four comparisons (in 1–2 mm pre-wetted soil aggregates: 76% vs 79% water stable; in 1–2 mm air-dried soil aggregates: 5.5% vs 6.1%; in 4.38 mm air–dried soil aggregates: 2% vs 7%). Methods: Wheat was grown continuously or in rotation with vetch (12 plots each, 20 x 30 m plots). Fertilizer and post–emergence herbicide were used on all plots. Soil samples were collected in June 1996 (plots with rotations) or July 1996 (plots without rotations), with four samples/subplot. Organic carbon was measured at 0–40 cm depth. Aggregate stability was measured at 0–30 cm depth.

 

2 

A replicated, randomized, controlled study in 1994–1998 in arable farmland in California, USA, found similar amounts of nitrogen in plots with four-year or two-year crop rotations. Implementation options: Similar amounts of nitrogen were found in plots with four-year or two-year crop rotations (29–42 vs 26–40 mg ammonium and nitate/kg soil, 0–15 cm depth). Methods: A four-year rotation (tomato, safflower, corn and wheat, beans) was used on 16 plots (four plots for each phase, each year), and a two-year rotation (tomato, wheat) was used on eight plots (four plots for each phase, each year). Each plot was 68 x 18 m. Fertilizer and pesticide were used on all plots. Soil samples were collected from tomato plots (every 2–3 weeks in the cropping season in 1994–1998, 0–15 cm depth).

 

3 

A replicated, controlled experiment in 1985–1995 in a rainfed durum wheat field near Aleppo, Syria, found more organic matter and greater stability in soils with crop rotations, compared to continuous wheat. Organic matter: More organic matter was found in plots with crop rotations, compared to continuous wheat, in five of six comparisons (11.4–13.8 vs 10.9 g/kg). Soil erosion and aggregation: More stable soils were found in plots with crop rotations, compared to continuous wheat (30–41% vs 22% of aggregates were water-stable). Methods: Durum wheat Triticum turgidum var. durum was grown continuously or in two-year rotations with lentils Lens culinaris, chickpeas Cicer arietinum, medic Medicago sativa, vetch Vicia faba, watermelon Citrullus vulgaris, or fallow (one plot for each crop phase, each year). Each plot was 36 x 150 m. Soils samples were collected each year, before planting (0–20 cm depth).

 

4 

A replicated, randomized, controlled study in 2004–2006 in an occasionally irrigated oat field in Portugal found less organic matter, phosphorus, and fungi in soils with a lupin-oat sequence, compared to an oat-oat sequence. Organic matter: Less organic carbon was found in soils with a lupin-oat sequence, in one of two comparisons (tilled plots: 6.2 vs 7.7 g/kg). Nutrients: Less phosphorus was found in soils with a lupin-oat sequence, in one of two comparisons (tilled plots: 51 vs 59 mg/kg). Similar amounts of nitrogen, and similar pH levels, were found in soils with both sequences (35–39 vs 41–44 g mineral N/kg, pH 5.2–5.4). Soil organisms: Fewer fungi were found in soils with a lupin-oat sequence (0.0023–0.0034 vs 0.0036–0.0045 colony forming units/g soil), but similar numbers of bacteria were found in both sequences (data not reported). Methods: Oats or white lupins Lupinus albus were grown in six plots each in 2003–2004 (year 1). Oats were grown in all plots in 2004–2005 (year 2). Each plot was 5 x 10 m. Half were tilled (15 cm depth), and half were not (crop residues were retained). All plots were fertilized with phosphorus (60 kg/ha), and oats were also fertilized with nitrogen (100 kg/ha). The seeds were sown in September and the oats were harvested in May. Soil samples were collected in year 2, in October, November, January, March, May, and July (0–15 cm depth). Bacteria and fungi were cultured from soil samples.

 

5 

A replicated, randomized, controlled study in 1993–2000 in arable farmland in Madrid, Spain, found less phosphorus in plots with fallow-barley rotations, compared to continuous barley. Organic matter: Similar amounts of organic carbon were found in soils with or without crop rotations (data not reported). Nutrients: Less phosphorus was found in soils with crop rotations, compared to continuous barley, in one of six comparisons (compared to fallow-barley at one depth: 16 vs 18 kg/ha). Similar amounts of nitrogen, and similar pH levels, were found in soils with or without rotations (0.9–1.8 Mg/ha; data not reported for pH). Methods: Barley was grown continuously (one plot), or in rotation with vetch Vicia sativa or fallow (one plot/phase), in each of three tillage treatments (conventional, reduced, or no tillage), in each of four blocks. Plots were 10 x 25 m. The barley phases were fertilized (8-24-8 NPK: 200 kg/ha; ammonium nitrate: 200 kg/ha). Before the experiment, barley was grown in these plots for over 10 years. Barley was harvested in June. Soil samples were collected after each harvest (0–90 cm depth).

 

6 

A replicated, randomized, controlled study in 2003 in a rainfed barley field in the Ebro river valley, Spain, found less organic matter, but inconsistent differences in stability, in soils with barley-fallow rotations, compared to continuous barley. Organic matter: Less organic carbon was found in soils with barley-fallow rotations, compared to continuous barley, in one of three comparisons (0–20 cm depth: 2,306 vs 2,743 g C/m2). Soil erosion and aggregation: Fewer large aggregates were found in soils with barley-fallow rotations, compared to continuous barley, in one of nine comparisons (water-stable aggregates >2,000 µm, 0–5 cm depth: 0.09 vs 0.15 g aggregate/g soil), but more were found in one of nine comparisons (53–250 µm, 0–5 cm depth: 0.6 vs 0.5 µm). Methods: Barley was grown in rotation with fallows on three plots, but barley was grown continuously on three other plots. Plots were 33 x 10 m. Soil samples were collected with a flat spade (0–20 cm depth) in July 2003.

 

7 

A replicated, controlled study in 1994–2001 in a rainfed cereal field in the Duero valley, northern Spain, found similar amounts of organic matter in soils with or without crop rotations. Organic matter: Similar amounts of organic carbon were found in soils with or without crop rotations (36–42 Mg/ha). Methods: Cereals (wheat and barley) were grown continuously (one plot/year), in rotation with vetch Vicia sativa (two plots/year: one cereal, one vetch), or in rotation with fallow (two plots/year: one cereal, one fallow), in each of three tillage treatments (conventional, reduced, or no tillage), in each of four blocks. Each plot was 450 m2, and there were 60 plots in total (five plots/three treatments/four blocks). The cereals were fertilized (8-24-8 NPK: 400 kg/ha; ammonium sulphate: 300 kg/ha). Herbicide was used on all plots. Soil samples were collected in October 1994, 1997, and 2000 (three samples/plot, 0–30 cm depth), before tillage in November.

 

8 

A replicated, randomized, controlled study in 2005–2007 in a wheat field near Madrid, Spain, found less organic matter in soils with wheat-fallow rotations, compared to continuous wheat. Organic matter: Less organic carbon was found in soils with wheat-fallow rotations, compared to soils with continuous wheat, in one of four comparisons (November 2006, 0–7.5 cm depth: 7 vs 8 Mg/ha). Soil erosion and aggregation: No difference in stability was found in soils with or without rotations (25–55% of aggregates were water-stable). Methods: Crop rotation (wheat-fallow) or continuous cropping (wheat-wheat) was used on 12 plots each (10 x 25 m plots) in 2005–2007. All plots were fertilized. Soil samples were collected after the seedbeds were prepared (three samples/plot, 0–15 cm depth), in November 2006 and October 2007.

 

9 

A replicated, randomized, controlled study in 1996–2008 in rainfed farmland near Aleppo, Syria, found similar amounts of organic matter and nitrogen in soils with two-course or four-course crop rotations. Implementation options: Similar amounts of organic matter and nitrogen were found in soils with two-course or four-course crop rotations (organic matter: 10–18 g/kg soil; nitrogen: 0.76 g/kg soil). Methods: The crop rotations were vetch-barley (two-course) or vetch-barley-vetch-wheat (four-course). Each rotation was grown on twenty plots (25 x 25 m). Soil samples were collected in 2003 (0–30 cm depth) and 2008 (0–20 cm depth).

 

10 

A replicated, randomized, controlled study in 2009–2010 in a rainfed wheat field in the Wongan Hills, Western Australia, found that less nitrous oxide was emitted from, and more methane was absorbed by, soils with a lupin-wheat sequence, compared to a wheat-wheat sequence, over two years. Greenhouse gases: Less nitrous oxide was emitted from plots with a lupin-wheat sequence, compared to a wheat-wheat sequence, in one of two comparisons (without added lime: 100 vs 130 g N2O–N/ha, over two years). More methane was absorbed by plots with a lupin-wheat sequence, compared to a wheat-wheat sequence, in one of two comparisons (without added lime: 991 vs 601 g CH4-C/ha). Methods: Wheat or lupin Lupinus angustifolius was planted on six 150 m2 plots each, in June 2009. In June 2010, wheat was planted on all plots. Lime was added to half of the plots (3.5 t/ha). Different fertilizers were used on each crop (e.g., no nitrogen was used on lupin). No plots were tilled. Nitrous oxide and methane were measured with chambers (500 mm x 500 mm chambers, eight measurement/day/plot, for two years beginning in June 2009).

 

11 

A replicated, randomized, controlled study in 1992–2010 in a rainfed wheat field in southern Spain found inconsistent differences in nitrate between soils with or without crop rotations. Nutrients: Less nitrate was found in soils with crop rotations, compared to continuous wheat, in three of four rotations (55–117 vs 124 kg nitrate/ha), but more was found in one of four rotations (wheat-faba bean: 139 kg/ha). Methods: Wheat was grown continuously (one plot/year) or in two-year rotations with chickpeas, faba beans, sunflower, or fallows (each with two plots/year: one wheat, one alternate), in each of two tillage treatments (conventional tillage or no tillage), in each of three blocks. Each plot had four subplots (10 x 5 m), each with a different amount of fertilizer (0–150 kg N/ha). Soil samples were collected every three years (0–90 cm).

 

12 

A replicated, controlled study in 1999–2010 in a rainfed durum wheat field in Sicily, Italy, found less carbon and nitrogen, but more microbial biomass and higher greenhouse-gas emissions, in plots with wheat-bean rotations, compared to continuous wheat. Organic matter: Less organic carbon was found in soils with wheat-bean rotations, compared to plots with continuous wheat, in two of three comparisons (29–33 vs 33–36 Mg/ha; 19 vs 21 g/kg). Nutrients: Less nitrogen was found in soils with wheat-bean rotations, in two of three comparisons (1–1.1 vs 1.3–1.4 g/kg). Soil organisms: More microbial biomass (measured as carbon) was found in soils with wheat-bean rotations, in two of three comparisons (293–509 vs 208–330 mg C/kg). Greenhouse gases: More carbon was emitted from plots with wheat-bean rotations, compared to continuous wheat (carbon output, in one of three comparisons: 3.1 vs 2.4 Mg C/ha/year; soil respiration: 17–22 vs 14–19 mg C/kg/day). Methods: Durum wheat Triticum durum was grown continuously or in a two-year rotation with faba beans Vicia faba on four plots each (18.5 x 20 m plots). Fertilizer and herbicide were used on all plots (half were tilled, and half were not). Soil samples were collected after harvest, in June 2009 (three samples/plot, 0–15 cm depth). Carbon dioxide was measured on 36 days in April 2008–April 2009 (closed chambers, 12 measurements/plot, 9–11 am).

 

13 

A replicated, randomized, controlled study in 2010–2011 in a rainfed field in Western Australia found more nitrogen in plots with canola-wheat or wheat-wheat sequences. Nutrients: More nitrogen was found in plots with a canola-wheat sequence, compared to a wheat-wheat sequence, in one of four comparisons (after planting: 106 vs 93 kg total N/ha). Greenhouse gases: Similar nitrous oxide emissions were found in plots with different crop sequences (0.03–0.18 g/ha/hour). Methods: Wheat or canola was grown on three plots each, in 2010, and wheat was grown on all plots in 2011. Each plot was 1.4 x 40 m. Fertilizer (150 kg/ha/year) and herbicide were used on all plots. Soil samples were collected in September 2010–December 2011 (0–150 cm depth). Nitrous oxide was measured in closed chambers, five times in May–October 2011 (250 mm diameter, 325 mm height, two chambers/plot, one hour/plot).

 

14 

A replicated, randomized, controlled study in 1994–2013 in a rainfed wheat field near Madrid, Spain, found less organic matter and microbial biomass in plots with four-year rotations, compared to continuous wheat. Organic matter: Less organic carbon was found in soils with rotations, compared to continuous wheat, in two of 12 comparisons (5–7 vs 6–8 g C/kg soil). Soil organisms: Less microbial biomass (measured as carbon) was found in soils with rotations, compared to continuous wheat, in one of 12 comparisons (200 vs 260 mg C/kg soil). Greenhouse gases: Similar carbon dioxide emissions were found in plots with or without rotations (20–42 mg CO2-C/kg soil/day). Methods: Continuous wheat crops or four-year crop rotations (fallow-wheat-vetch Vicia sativa-barley) were used on 12 plots each (10 x 25 m subplots). The cereals were fertilized (NPK, 200 kg/ha, twice/year, in October and March). The crop residues were shredded and retained (but some of the plots were tilled). Soil samples were collected in October 2010, April 2011, November 2011, May 2012, October 2012 and April 2013 (50 mm diameter, 0–15 cm depth).

 

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.