Main authors:	Mart Ros, Gerard Velthof, Oene Oenema, Meindert Commelin, Susanne Klages, Linda Tendler, Jenny Rowbottom, Isobel Wright, Donnacha Doody, Luke Farrow, Birgitte Hansen, Morten Graversgaard, Irene Asta, Andrej Jamsek, Katarina Kresnik, Matjaz Glavan, Jean-François Vernoux, Nicolas Surdyk, Christophoros Christophoridis, Kate Smith, Irina Calciu, Sonja Schimmelpfennig, Hyojin Kim, Piet Groenendijk.
FAIRWAYiS Editor:	Jane Brandt
Source document:	»Ros, M. et al. 2020. Identification of most promising measures and practices: 2. Reduction nitrate transport from agricultural land to groundwater and surface waters by management practices. FAIRWAY Project Deliverable 4.3, 72 pp

 

Contents table

1. Review of existing meta-analyses

2. Meta-analysis

3. Case studies

1. Review of existing meta-analyses

Results from reviews and meta-analyses that assessed the effects of different measures on NO₃ losses and soil NO₃ concentrations were summarized and are organized below. The amount of literature available greatly differed for the various practices: several reviews on cover crops and biochar have been published, but for other measures (e.g. adaptations in soil drainage or irrigation management) hardly any (quantitative) reviews were found. Most reviews did not include a cost-benefit analysis, but in two cases (both for nitrification inhibitors) they were reported.

Overall, assessment of these literature reviews showed that most included measures were effective to some extent at reducing the risk of NO₃ losses to water bodies. There is overwhelming evidence that the use of non-legume cover crops is an efficient practice, with reductions in NO₃ leaching from 35% to 98%. The effect does however diminish when legumes are used. Besides cover crops, the use of (nitrification) inhibitors is also effective. In particular for dicyandiamide (DCD), a lot of studies reported a reduction in NO3 leaching. For biochar, the effect differs from none at all, to considerable reductions in NO₃ leaching. The success of biochar applications seem to depend on soil and environmental conditions, as well as the nature of the biochar used. For changes in tillage systems (switching from conventional to no-till), the reviewed study did not show a significant reduction in NO₃ losses. Rather, for losses through leaching a significant increase was even reported. Switching to organic farming often includes no-till practices and did seem to reduce NO₃ losses. However, there was considerable variation in the results, and when losses were expressed per unit of produce, losses were often increased due to lower yields by organically managed farms.

Table 2.3 gives a summary on the studies and effects of the various measures in the reviews. Below, reviews on the various measures will be discussed per measure.

Table 2.3: Summary of the measures and main effects found in the literature search.

Study	Measure	Response variable	Observ.	Overall effect1	Comments
Basche et al. 2014	Cover crops	NO3 leaching	11	-98%
Borchard et al. 2019	Biochar	NO3 leaching	688	N.S.	Long-term studies show greater effect. No effect in grasslands.
Cai and Akiyama 2017	Inhibitors	NO3 leaching	45	-46%	DCD application
Daryanto et al. 2017	No-till	NO3 leaching	180	+13%	NO3 concentrations in leaching samples were similar.
Daryanto et al. 2017	No-till	NO3 runoff	61	N.S.	NO3 concentrations in runoff increased significantly.
Liu et al. 2018	Biochar	N leaching	156	-26%
Liu et al. 2018	Biochar	Soil NO3 concentration	350	-12%
Mayer et al. 2007	Buffers	N leaching/runoff	89	-68%
Mondelaers et al. 2009	Organic farming	NO3 leaching	?	-32%	Considerable variation
Nguyen et al. 2017	Biochar	Soil NO3 concentration	862		-11%
Qiao et al. 2015	Inhibitors	NO3 leaching	102	-47%	Cost-benefit analysis: DCD - $162.70 ha-1 y-1
Quemada et al. 2013	Fertilizer management	NO3 leaching	106	-40%
Quemada et al. 2013	Cover crops	NO3 leaching	59	-35%	Includes both legume and non-legume crops
Quemada et al. 2013	Improved irrigation	NO3 leaching	82	-58%	Average of several practices
Thapa et al. 2018	Cover crops	NO3 leaching	216	-56%	Non-legume crops only
Tonitto et al. 2006	Cover crops	NO3 leaching	?	-70%	Non-legume crops only
Tonitto et al. 2006	Cover crops	NO3 leaching	?	-40%	Legume crops only
Tuomisto et al. 2012	Organic farming	N leaching	48	-31%	But an increase in N leaching per unit of produce.
Valkama et al. 2015	Cover crops	N leaching	27	-50%	Non-legume crops only
Valkama et al. 2015	Cover crops	Soil N concentration	29	-35%	Non-legume crops only
Yang et al. 2016	Inhibitors	NO3 leaching	298	-55%	Both DCD and DMPP Cost-benefit analysis: DCD - $109.49 ha-1 y-1 DMPP - $15.67 ha-1 y-1

¹ Negative numbers indicate a reduction in NO3 losses/concentrations ² N.S. Not significant

Cover crops

Basche et al. (2014) did a meta-analysis that focuses on the effect of cover crops on nitrous oxide (N₂O) emissions, but also includes a few studies with data on NO₃ leaching (3 studies, 11 data points; Figure 2.2). They only included studies in which the cover crop was not harvested. The authors argue that, while cover crops may not reduce (and sometimes even increase) N₂O emissions from agricultural fields, the N that is prevented from leaching as NO₃ represents a reduction of subsequent N₂O emissions from leachate once it has been transported outside field boundaries. Therefore, reducing NO₃ leaching to ground and surface waters may also benefit N₂O emissions from agricultural sources.

Figure 2.2

Thapa et al. (2018)also did a meta-analysis on the effect of cover crops on NO₃ leaching (28 studies, 238 observations). They reported a 56% reduction of NO₃ leaching by non-legume cover crops. Mixtures of legumes and non-legumes showed a response similar to the non-legume crops and both were more effective than legumes on their own. These results were obtained when one study was omitted from the results due to high variation. Results were affected by planting date, shoot biomass, and precipitation, but the lack of statistical information in the used studies prevented a deep analysis of contributing factors.

Tonitto et al. (2006) conducted a study on cash crop yields and N retention in systems with and without cover crops. They found that non-legume cover crops decreased NO₃ leaching by 70%, but there was no difference in cash crop yields between systems with and without a cover crop. Legume-based systems reduced NO₃ leaching by 40% on average. There was an overall 10% yield penalty when using legumes, rather than mineral fertilizer to provide N to the cash crops, but no negative effect was observed when legumes provided more than 110 kg N ha^-1. There were no differences in soil N status of conventional and green manure systems after harvest, suggesting that NO₃ leaching losses were mainly reduced by avoiding bare fallow throughout the cropping rotation.

Valkama et al. (2015) studied the effects of catch crops on nitrogen (either NO₃ or total N) leaching and yield of spring cereals in the Nordic countries (Denmark, Sweden, Finland, and Norway). In their meta-analysis, non-leguminous catch crops reduced N leaching by 50% (27 observations; Figure 2.3a) and soil NO₃/inorganic N by 35% in fall (29 observations; Figure 2.3b). For the effect on soil N, there were differences among species used: annual ryegrass was more effective (60% reduction) than perennial ryegrass and Westerwolds ryegrass (25% reduction). Legumes, on the other hand, did not reduce soil N. Studies with non-legume catch crops also reported a slight (3%) yield reduction, whereas the ones with legume or mixed (legume + non-legume) catch crops reported increases for yield and crop N content (6%).

Figure 2.3

Biochar

Borchard et al. (2019) investigated the effect of biochar additions on N₂O emissions, soil NO₃ concentrations, and NO₃ leaching. Their main findings show that, overall, soil NO₃ concentrations remained unaffected and the use of biochar did not significantly reduce NO₃ leaching (13% reduction, not significant). However, in studies that lasted longer than 30 days (shorter studies showed an increase in NO₃ losses) the effect was significant (26-32% reduction). Biochar decreased both N₂O and NO₃ losses in annual arable crops and horticulture, but no effect was found for grasslands or perennial crops. Besides this, addition of large additional N (> 150 kg/ha) as (mineral) fertilizer diminished the effect of biochar on NO₃ leaching. Although biochar addition may suppress soil N losses as N₂O emissions and NO₃ leaching, there is a higher risk of NH₃ volatilization when applying biochar.

Liu et al. (2018) assessed the effect of biochar additions on the soil N cycle. Aside from N leaching losses, they summarized effects on gaseous losses and soil N pools. On average, biochar reduced N leaching by 26% (22% for NH₄ and 29% for NO₃) and soil NO₃ concentrations with 12%. NH₃ volatilization, on the other hand, was increased by 19% (this effect was larger in soils with a low buffering capacity). Wood-based biochars were the most effective, whereas manure-based biochars did not seem to have a significant effect. There was no effect of pyrolysis temperature on the effect size (with biochar/no biochar), but the effectiveness in reducing N leaching increased with higher biochar application rates.

Figure 2.4

Nguyen et al. (2017) reported a meta-analysis on the effect of biochar on soil inorganic N (56 studies, 1080 observations). They found that biochar applications reduced soil inorganic N (-11% for NH₄ and -10% for NO₃). Most of their studies were shorter than a year. They found plant-derived biochars and biochars pyrolyzed at lower temperatures (< 401 °C) to be more effective at reducing soil N concentrations than woody biochars. Higher biochar application rates were more effective, but application of urea alongside biochar decreased the biochar’s effect of lowering soil NO₃ concentrations and even increased them compared to the control. Biochar worked best to reduce soil NO₃ concentrations on neutral soils (Figure 2.4). Very acidic soils showed increased NO₃ levels when biochar was applied. Generally, time between application and observation had little effect on soil NO₃ concentrations. Climatic conditions may affect the effect of biochar on reducing nitrate leaching, but an assessment of climatic conditions was not included in the meta-analysis paper.

Inhibitors

Cai and Akiyama (2017) reviewed the effect of inhibitors and biochar on N₂O and NO₃ losses in urine patches on grasslands. Studies originated predominantly from temperate areas (UK and New Zealand). They reported a decrease of 46% in NO₃ losses when the nitrification inhibitor DCD was applied. When used in combination with the urease inhibitor n-butyl thiophosphoric triamide (NBPT) NO₃ losses were reduced by 42%. Effectiveness increased with higher doses of DCD. There was no significant difference between coated and liquid forms of DCD and study type or duration did not affect the results. Although there was no difference between the effects of DCD and those of DCD+NBPT, the authors state that if NH3 large losses are expected, a combination of DCD and NBPT would be the more logical option.

Qiao et al. (2015) collected 62 field studies of nitrogen enriched studies to summarize the effect of nitrification inhibitors on the nitrogen cycle. They found that the use of inhibitors decreased NO₃ leaching by 47%. Besides this, N₂O emissions were decreased by 44%, NO emissions by 24%, but NH₃ emissions were increased by 20%. They also conducted cost-benefit analysis and calculated that applications of nitrification inhibitors could increase the revenue of a maize farm by $162.70 ha^-1 y^-1 which would correspond to a 8.95% in financial gain (Table 2.4).

Yang et al. (2016) investigated the effect of dicyandiamide (DCD) and 3,4-dimethypyrazole phosphate (DMPP) on soil nitrogen transformations and plant productivity. They found that both nitrification inhibitors were equally effective at altering soil N transformations. Both inhibitors increased soil NH₄ content (DCD: 25.3% and DMPP: 41.1%) and decreased NO₃ content (DCD: 17.0% and DMPP: 20.7%). DCD and DMPP were equally effective at reducing NO₃ leaching (~55%, n=298), but NH₄ leaching was increased for DMPP, but decreased for DCD. Also, in neutral soils or when urea was applied, DMPP seemed more effective than DCD. Total N leached did not differ between the two inhibitors. For plant production, DCD was more effective than DMPP in increasing yields. These authors also conducted a cost-benefit analysis and concluded applying fertilizer N in combination with DCD had a benefit of $109.49 ha^-1 y^-1, whereas for DMPP this was only $15.67 ha^-1 y^-1 (Table 2.5). The authors do note that DCD has a higher toxicity to plants and human health than DMPP (although toxicity for both products is relatively low) and that this may change the cost-benefit analysis over a longer time.

Table 2.4: Cost-benefit analysis for a maize farm applying nitrification inhibitors (NI) with fertilizer rate of 125 kg N/ha/yr. For change in N loss under NI, positive values indicate that NI increases N losses, and negative ones indicate N reduces N loss. For the monetary response, the positive numbers indicate the amount of the economic benefit, whereas the negative ones indicate the amount of the economic cost. From Qiao et al. (2015).

Table 2.5: Cost-benefit analysis of nitrification inhibitor (NI) application in a maize farm with fertilizer N rate of 125 kg N/ha/yr. For change in N loss under NI, positive values indicate that NI increases N losses, and negative ones indicate N reduces N loss. For the monetary response, the positive numbers indicate the amount of the economic benefit, whereas the negative ones indicate the amount of the economic cost. From Yang et al. (2016)

No tillage systems

Daryanto et al. (2017) compared no-till systems to conventional tillage systems in their NO¬3 losses through leaching and runoff processes for several field crops. They compared both NO₃ load and concentrations (Figure 2.5). No-till provided no overall reduction in either concentration or load than conventional tillage systems. No-till systems had higher NO₃ concentrations in runoff, but due to lower runoff volumes the load was similar. Leaching NO₃ losses were significantly higher in no-till systems. Soil drainage characteristics (texture, artificial drainage) are likely to play an important role in the effects of no-till on NO₃ losses. Fertilizer type (organic vs. inorganic vs. no fertilizer) had no effect on the NO₃ concentrations in runoff and leaching samples.

Figure 2.5

Vegetative buffers

Mayer et al. (2007) reviewed the effect of riparian buffers on nitrogen concentrations in streams and tried to link the effects of buffers to the buffer width (45 studies; 89 observations). Overall, buffers were very effective at removing N from streams (67.5%). Buffers were particularly effective at removing subsurface N. They found a wide variation in effectiveness and a small part could be explained by buffer width. Buffers > 50 m were more effective at removing N than were those <25 m. This was particularly true for horizontally transferred N removal, but not for vertical transferred N removal. No effect of buffer vegetation was observed, but buffers with herbaceous or herbaceous/forest vegetation became increasingly effective as they got wider.

Organic farming

Tuomisto et al. (2012) studied the effect of organic farming on the environment. Their main conclusion is that, while organic farming may have environmental benefits and may reduce N leaching per land unit, this is not necessarily true per unit of produce. This is a result from both the lower inputs and outputs of organic farming. Over 48 observations, they found a 31% reduction of N leaching in organic systems when expressed per land unit, but a 49% increase of leaching losses per unit of product. The lower leaching losses (and lower yields) were likely a result of reduced N inputs in organic farming systems. In addition to the effect on N leaching, Tuomisto et al. (2012) show a reduction in N2O emissions (per land unit) and an increase in soil organic matter content. Losses of ammonia and P were not significantly different between conventional and organic systems.

Mondelaers et al. (2009) did a meta-analysis on the differences in environmental impacts between organic and conventional farming. They assessed NO₃ losses among other parameters. Their analysis showed that NO₃ leaching was 32% lower in organic farming systems. However, the variability between studies was considerable.

Other

Quemada et al. (2013) conducted a meta-analysis (44 studies, 279 observations) on the effects of water and fertilizer management, cover crops, and fertilizer technology on NO₃ leaching and crop yield. They found that proper water application management can reduce NO₃ leaching by up to 80% without lowering crop yields. Improving fertilizer management reduced leaching by 40% (Figure 2.6), and the best results were obtained if fertilization occurred at the recommended rate. Cover crops reduced NO₃ leaching by 50% compared to fallow land, but only if the cover crops were not leguminous. Legume cover crops had no significant effect.

Figure 2.6

Wang et al. (2019) used meta-analysis (86 studies, 324 observations) to construct a model to describe the emission factor for NO₃ leaching from N fertilizer additions. They show that NO₃ leaching from N additions do not remain constant (as a set fraction of the added N), but increase with higher N additions according to a quadratic relationship. Their conclusion is that the emission factor for NO₃ leaching set by the IPCC (30% of N input) overestimates NO₃ leaching.

Zhou and Butterbach-Bahl (2014) conducted a meta-analysis on the link between NO₃ leaching and crop yields in maize and wheat cropping systems. They showed that maize systems saw NO₃ losses that were about two times higher than those in wheat systems. Due to higher maize yields however, yield-scaled NO₃ losses were comparable between the two systems. They further conclude that NO₃ losses can be reduced by fertilizing close to the optimal N rate, as NO₃ leaching increased with N application rate.

2. Meta-analysis

In the meta-analysis we included 53 studies and 278 observations that compared a variety of measures to reduce NO₃ losses (Table 2.6). Because of a lack of studies and the absence of a solid, uniform type of pairwise comparisons (treatment group vs. control group), it was impossible to incorporate studies covering measures like implementation of balanced N fertilization, adaptations of N application timing or rate, restricted grazing, changes in crop rotations, and mulching (see Table 2.1). It was also necessary to combine studies with different indicators, and so the effect on N or NO₃ concentrations in soil and water is assessed jointly with reported results on NO₃ flux from soil (field) to water. This implies a significant generalization of the data and the results should thus be viewed with some caution.

Table 2.6: Comprehensive list of studies included in the meta-analysis, including the NO₃ indicator and the type of measure described in the studies.

Study	Indicator	Measure type	Study	Indicator	Measure type
Adams and Jan, 2006	NO3 flux	Cover crops	Martinez and Guiraud, 1990	NO3 concentration	Cover crops
Asing et al., 2008	NO3 concentration	Inhibitor	Mehdi and Madramootoo, 1999	NO3 concentration	Tillage
Askegaard et al., 2006	NO3 flux	Cover crops	Menneer et al., 2008	NO3 concentration	Inhibitor
Benham et al., 2007	NO3 flux	Tillage	Monaghan et al., 2009	NO3 concentration	Inhibitor
Besnard, 2004	NO3 flux + NO3 concentration	Cover crops	Nicholson et al., 2016	NO3 flux	Application method
Besnard and Kerveillant, 2006	NO3 flux	Cover crops	O’Connor et al., 2016	NO3 flux	Inhibitor
Bock et al., 2015	NO3 concentration	Biochar	Parkin et al., 2016	NO3 concentration	Cover crops
Bonaiti and Borin, 2010	NO3 flux + NO3 concentration	Controlled drainage	Pisani et al., 2017	NO3 concentration	Tillage
Bosch et al., 2015	NO3 flux + NO3 concentration	Tillage	Premrov et al., 2014	NO3 flux	Cover crops + Tillage
Dennis et al., 2010	NO3 flux	Inhibitor	Ritter et al., 1998	NO3 content	Cover crops + Tillage
Di and Cameron, 2012	NO3 concentration	Inhibitor	Saarnio et al., 2018	N content	Biochar
Drury et al., 2009	NO3 flux + NO3 concentration	Controlled drainage	Sanz-Cobena et al., 2012	NO3 concentration	Inhibitor
Dunn et al., 2011	NO3 concentration	Vegetative buffer	Schipper and Vojvodić-Vuković, 2000	NO3 concentration	Vegetative buffer
Eykelbosh et al., 2015	NO3 concentration	Biochar	Schmidt and Clark, 2012	NO3 concentration	Vegetative buffer
Francis et al., 1995	NO3 flux + N content	Cover crops	Shepherd, 2006	NO3 flux	Cover crops
García-González et al., 2018	N content	Cover crops	Shepherd et al., 2017	NO3 concentration	Inhibitor
Gordon et al., 2011	?	Tillage	Smith et al., 2002	N concentration	Inhibitor
Goss et al., 1993	NO3 flux	Tillage	Stolzenburg, 2010	NO3 flux + NO3 concentration	Cover crops
Guardia et al., 2018	N content	Inhibitor	Tauchnitz et al., 2018	NO3 concentration	Inhibitor
Hill et al., 2015	NO3 flux	Biochar + Inhibitor	Thorman et al., 2016	NO3 flux	Application method
Huang et al., 2015	NO3 flux	Tillage	Ventura et al., 2013	NO3 flux	Biochar
Jabro et al., 2016	NO3 concentration	Tillage	Vos et al., 1994	NO3 concentration	Cover crops
Johnson and Smith, 1996	NO3 flux	Tillage	Welten et al., 2014	NO3 concentration	Inhibitor
Jouni et al., 2018	?	Controlled drainage	Wesström and Messing, 2007	NO3 flux	Controlled drainage
Kaspar et al., 2012	NO3 flux	Cover crops	Yamulki and Misselbrook, 2016	NO3 flux	Application method
Krueger et al., 2011	NO3 concentration	Cover crops	Zaman and Blennerhassett, 2010	NO3 concentration	Inhibitor
Macdonald et al., 2005	NO3 flux	Cover crops

Overall, the data was spread widely for the different measures assessed. Figure 2.7 shows the average and 95% confidence intervals for the effect size (ln(R)) of the different measures. The results from the meta-analysis show that implementation of a vegetative buffer, the use of cover crops, and application of (nitrification) inhibitors lead to a significant decrease in NO₃ losses (95% confidence interval not overlapping 0). For the other analyzed measures (tillage, controlled drainage, biochar, and changes in application method), no significant average effect was recorded in the compiled database. Moreover, although some of the measures had a significant effect on NO₃ losses, including ‘measure type’ as an explanatory variable in the meta-analysis model did not significantly improve it. This indicates that the variation of the effect explained by the different measures is limited.

Figure 2.7

For the measures for which enough studies and observations were available, we assessed the effect of the measure more closely. Figure 2.8 shows the effect of the individual cover crops on NO₃ losses (18 studies, 84 observations). Lupins, grass, barley, oat, mustard, and rye were particularly effective in reducing losses. For turnips and wheat the effect was not significant. In our database, we did not observe the large difference in effectiveness between legume and non-legume crops, as observed in other literature reviews. It should be noted, however, that the number of studies included in the current analysis was limited. Moreover, there was no significant effect of including ‘cover crop type’ as an explanatory variable in the model.

When examining changes in tillage practice (11 studies, 47 observations), we did see a significant improvement when the type of tillage was considered (p=0.0011, Fig. 2.9). Whereas studies that reported the effect of reduced tillage and no-till did not significantly affect NO₃ losses, there was one study (Gordon et al., 2011) that used row shaper and basin tillage that did not match with the tillage forms included in the other studies and was therefore kept separate. No-till and reduced tillage had no effect on NO₃ losses however, and this is in line with the results reported in previous meta-analyses.

For research on the use of nitrification inhibitors (14 studies, 67 observations) we were able to distinguish between DCD used alone, or in combination with a urease inhibitor. The analysis shows that by itself, DCD significantly reduced NO₃ losses, but the studies in which it was used in combination with a urease inhibitor showed no significant reduction (Figure 2.10). Including the differences between these two groups significantly improved the statistical model (p=0.0224). Overall, the effectivity of DCD as a measure is in line with the results found in previous meta-analyses.

Figure 2.8
Figure 2.9
Figure 2.10

3. Case studies

The results of the questionnaires sent out to the FAIRWAY case studies were collected and aggregated in a table (see »Annex 2). From 9 different case studies, 34 different measures were recorded. They were then aggregated by measure type and the average/overall scores for effectivity, cost, applicability, and adoptability were assessed from the individual records and comments.

In general, there was a wide variety of measures described (Table 2.7). Optimizing the rate and timing of fertilizer and manure applications were measures that were applicable throughout (almost) all of the case studies. With a highly rated effectivity, applicability and adoptability, as well as a relatively low cost these are measures that can be taken easily an may yield quick results. Nevertheless, storage space, weather conditions and labour demand may be limiting factors for implementation of these measures. Additionally, reducing application rates or balancing N fertilization may result in yield losses and potential to implement this may depend on the characteristics of the farm.

Table 2.7: Overview of the measure types applied and studied within the FAIRWAY case studies, with indications on effectivity, cost, applicability and adoptability.

Measure type	Country1	Target2	Effectivity3	Cost4	Applicability5	Adoptability6	Notes
Changes in cropping system or crop rotation	NL, SI	GW/SW/NUE	++	€	++	++	May improve soil health/quality, decrease chance of diseases.
Changes in fertilization timing	NL, DK, GR, RO, SI	GW/SW	+++	€	+++	+++	e.g. no manure spreading in the fall or splitting fertilizer applications. Expenses may increase if it demands more labor or requires additional manure storage space.
Changes in application method	DE, DK	GW	++	€	++	++	Effectivity may depend on the farm; may decrease other N losses such as greenhouse gases.
Changes in application dose (reduced input, balanced fertilization, or optimal fertilization)	NO, PT, DE, DK, GR, SI	GW/SW/NUE	++	€	+++	+++	May require soil testing. May be mandatory.
Cover crops	DK, GR, RO, SI	GW/SW	+++	€€	++	++	May increase soil OM content. Cost varies based on farm type. Less applicable/adoptable in Slovenia.
Reduced tillage	NO	SW	++	€€	+++	++	May prevent soil erosion.
Buffer strips (either between crops and waterways, or between rows of crops)	NL, FR, GR, RO, SI	GW/SW	++	€€	++	+	May contribute to landscape diversity, but decrease crop yields. Implementation costs differ per country.
Grassed waterways	NO	SW	+++	€€€€	+	+	May reduce erosion and contribute to landscape diversity. Reduces the amount of cropland
Farm-scale nutrient management tools	DE	NUE	*	€	+++	+++	Farmers may be obliged to use these tools.
Outreach and information events	DE	NUE	*	€	++	++	Effectivity depends on farm type and farmer knowledge.
Other	GR	GW/SW	?	?	?	?	Grassland and grazing management; improved fertilizer storage; no data available yet.

¹ Abbreviations for the various participating countries: NL Netherlands; SI Slovenia; DK Denmark; GR Greece; RO Romania; DE Germany; FR France; NO Norway; PT Portugal
² Target of the measure: GW groundwater; SW surface water; NUE nitrogen use efficiency
³ Effectivity is evaluated as Low (+, 5-10% load reduction), Moderate (++, 10-25% load reduction), High (+++, >25% load reduction), Variable (*), or Unknown (?).
⁴ Implication costs are evaluated as Low (€, < €10/ha), Moderate (€€, 10-50/ha), High (€€€, €50-100/ha), Very high (€€€€, > €100/ha), or Unknown (?).
⁵ Applicability is evaluated as No (+, on < 25% of the land), Partly (++, on 25-75% of the land), Yes (+++, on > 75% of the land), or Unknown (?).
⁶ Adoptability is evaluated as No (+, in < 25% of the cases), Partly (++, in 25-75% of the cases), Yes (+++, in > 75% of the cases), or Unknown (?).

 From the questionnaire results there was no clear distinction between the type of measures adopted in the different parts of the continent (»Annex 2). There were a few measures that were reported by just one or two case studies, but that does not directly imply that these measures are not used elsewhere. From the Greek case study, data on effectivity, cost, applicability, and adoptability was missing, as the case study had not been running for very long.

As reflected in the literature review and meta-analysis, the effectiveness of cover crops was rated as high. While it may not be the cheapest measure to implement, four out of eight case studies mentioned this measure. Buffer strips between crops and water ways (or between rows of crops) was also a frequently reported measure, but the effectivity was evaluated slightly lower and so was the adoptability. Compared to the literature review and the meta-analysis, there were several measures that were absent in the questionnaire results. Implementation of biochar and nitrification inhibitors was not reported by the experts. Measures on drainage or irrigation management were not reported either. The Norwegian case study reported a positive effect of reduced tillage, which in addition to decreasing nutrient transports to surface water, decreases erosion.

Another difference between the measures included in the literature review and meta-analysis on one hand, and the response from the case study questionnaires on the other is that the measures from the latter seemed to focus more on the farm-scale. Measures on outreach, information sharing, whole-farm assessments and large-scale N input reductions were reported. Although the effect of management decisions at this level is more difficult to quantify than field-ready measures such as cover crops, buffer zones, or inhibitors, they are relatively cheap to organize and may prove beneficial for reducing other N losses and increasing N use efficiency across the entire farm.

 

---

Notes:

For full references to papers quoted in this article see »References

Download the full report for Annex 2