**4. Discussion**

The outperformance of squarrose clover as green manure with respect to dead mulch was evident in both years of experimentation, although the positive effect of using a cover crop over a bare soil was year dependent. In fact, plant biomass, N uptake, and yield were improved where clover was incorporated as green manure in the first year of trials, but this was not noticed in the second year. This could be attributed to a lower mineralization rate in that year compared to 2016 as shown with the soil nitrates results, and their asynchrony with plant needs. Despite the higher N supply to the soil by leguminous cover crops and their capacity to improve N recovery of tomato, they can be no more effective than other cover crop species or chemical fertilizers in retaining nitrates in the soil profile, mainly due to the high mobility of nitrate ions [28–30]. Plastic mulching increases N mineralization and accumulation in soil and was reported in a large number of studies to increase crop yields, and this was mainly due to the increase in soil temperature, by 2 to 6 ◦C, and soil moisture as we confirmed [31–33]. In the first year of the experiment, transplantation occurred in late June which may have inflicted a thermal stress on tomato seedlings during the early growth of the plant, thus hindering their performance.

The response of organic vegetables to no-tillage conditions has not been consistent in the literature and the success seems to depend on an adequate context-specific management. Some studies showed tomato growth and production unaffected by tillage and cover crop residues managemen<sup>t</sup> [34–36]. Other results from reduced tillage in bell pepper, onion, and zucchini production have ranged from statistically equal or even higher [37–39] to 20% and more than 90% reduction of no-till yields in these and other horticultural crops [40–42]. In our case, this could be attributed to both low soil nitrates availability and high weed competition during tomato growth. Both factors have been responsible for yields' decline in organic reduced tillage systems compared with ploughed systems in many previous experiments [43]. The slow mineralization of cover crop laid as dead mulch explained the low soil nitrates available for plants with respect to other residues managemen<sup>t</sup> affecting plant nutrition [12–14] and partly the depression in plant performance. Nevertheless, the low mineralization may increase the N use efficiency of vegetables as demonstrated with tomato and eggplants cultivated on legume dead mulches ranging from 39 to 60% when compared to conventionally tilled systems [35,44]. Placing cover crop residues on soil surface may enhance the synchronization between N mineralized and

eggplant N demand in legume cover crops, while in others (i.e., cereals) it appears to mitigate the shortage of soil inorganic N for the following vegetable [45].

One of the most important attributes of an e ffective mulch is biomass production with high quantity of the residues necessary for the control of an increased weed pressure, although the limit depends on the specific characteristics of the growing system [2,10,46]. Squarrose clover in our study did not exceed the 3.5 t ha−<sup>1</sup> with which low performance was a ffected by sowing and killing date along with fluctuations in weather conditions, i.e., lower precipitations in 2017 during cover crop growth (Figure 1). The dead mulch did not ensure weed control in the first season with originally high field weed infestation, and in the next season succeeded to reduce weed infestation (38% lower weed biomass in NT-CC systems compared to NT-NC) but was not enough to increase plant performance and to decrease the competition over soil nitrates. In systems that received additional weed mowing over the dead mulch, an increase in plant performance was noticed although in some instances it was not statistically significant. Mowing, however, is not an e ffective measure to control weeds over the dead mulch and it disturbs the mulch and its uniformity. For these reasons, multi-tactic weed managemen<sup>t</sup> should be considered in organic no-till as it is di fficult in some cases for cover crops to be the unique method for weed control. Mechanical weed control practices that can perform on high residue conditions, a complex crop rotation and the use of allelopathic cover crops or mixtures of cover crops, are tools to be exploited in order to reduce weed pressure. The feasibility of no-till depends on field conditions and, for this, the preparation of suitable conditions before the implementation can be crucial for its success. In case of high weed seed bank, for example, stale seedbed in coordination with some previously cited practices (to mitigate the e ffects of frequent tillage), can be performed if possible before shifting to no-till.

Although some studies showed reduced tillage associated with a risky increase in soil compaction [47,48], our trials showed a modest soil compaction on topsoil that could not have been attributed to stress or yield depression if considering the threshold of 2000–2500 KPa for root proliferation and plant growth inhibition [27].

Tillage and cover crop residues managemen<sup>t</sup> did not show pronounced e ffects on fruit basic quality where higher TSS, lower pH, and firmer fruits are preferred. This result is in accordance with other studies that showed these characteristics una ffected by tillage systems in tomato production [49,50]. However, an increase in the vitamin C content was obtained in the dead mulch system left without weed control. In previous studies, a high N concentration in the nutrient solution/fertilization was shown to favor plant leaf area development and to decrease light penetration into the canopy and the vitamin C content in fruits, what may have been found with plants from CT systems [51].
