Use of Olive-Mill Wastewater for the Containment of Phelipanche ramosa in Open Field of Processing Tomato Crop

Phelipanche ramosa (L.) Pomel is a root holoparasitic weed that infested many crops, particularly the processing tomato. In semi-arid regions of the world, this problem is increasing, both in intensity and in acreages. In the last few years, researches for the control of this parasite has increased its effort to use suitable methods, as well as the utilization of natural compounds. In this regard, four experimental trials were carried out over three growing seasons (2014, 2016 and 2017), at Foggia (Apulia Region, Southern Italy), in the open field, on processing tomato (cv. “Dres”) crop, infested by P. ramosa, where the olive-mill wastewater (OMW) was incorporated into the soil at the dose of 80 m3 ha-1 40 days prior to tomato seedling transplant in comparison with the untreated control. The numbers of P. ramosa emerged shoots (branched plants) were significantly lower in the OMW treatment than in the untreated control. The tomato marketable yield showed a significantly higher value in the OMW treatment than the untreated control. No significant differences for the fruit qualitative characteristics were observed.


I. INTRODUCTION
Apulia Region, a coastal area of southeastern Italy, is the most important agricultural area for processing tomato (Lycopersicon esculentum Mill.) crop, with the annual 17920 hectares, 90% of which are in Foggia province [1]. In this area, in the last years, the tomato crop has been increasingly infested by the obligate root holoparasite plants of Phelipanche ramosa (L.) Pomel (syn. Orobanche ramosa L.), that is also continuously expanding into new areas. This parasitic species attaches to tomato roots early in the growing season, at 14-28 days after transplanting (DAT), depending on temperature and wet soil condition. The P. ramosa branched plant, composed of shoot numbers between 2 and 30, emerge from the soil between 35 and 56 DAT [2], [3]. They cause great yield reduction, ranging from 5 to 100%, depending on the timing and severity of the infestation [4], [5]. The heavy infestations is due to the high production of seed number per plant (up to 500,000) [6] that have very small dimensions (about 0.2-0.3 mm) and that in the absence of a host can remain viable in the soil for extended periods (up to 20 years) [7]. The parasitized tomato plants initially show some stunted growth, and subsequently a decrease in the quantity and quality of the yield, because of the reduced ability to use nutrients and water removed from the parasitic plant.
Various methods have been tried to maintain the P. ramosa populations below the threshold levels for damage, including preventive measures, physical, chemical, agronomic, biological, biotechnological and integrated methods [8]- [10].
In agricultural systems, characterized by low external input, herbicides are largely or entirely avoided, and weeds are suppressed largely through physical and ecological strategies [11]. Therefore, many researches to control the parasitic weed paid attention towards suitable methods, such as the use of organic compounds [10], [12]- [14]. Soil fertility plays a key role in P. ramosa management. Indeed, it is long been recognized that P. ramosa infestation tends to be associated with less a fertile soil condition [15]. The phenomenon is due to increase in the biosynthesis of strigolactones and in the release of parasite germination stimulants from plant roots that suffer from nutritional deficiencies [16]. On the contrary, the availability of nutrients in the soil usually promotes less development of the radical system, decreasing the likelihood of root recognition by parasitic seeds [17]. In addition, fertile soil improves the nutrient status of the host plants, allowing them to maintain a high osmotic potential and enhancing the competitive ability for sap flux versus parasitic demand [18]. Improved soil fertility, by adding organic matter, was found to be a potential control of Pheliphanche species [19]. Organic compounds are widely used in cropping systems to increase soil organic matter, structural stability, water holding and cation exchange capacities, and as a source of nutrients [20]. In addition, organic soil amendments can also provide a rich source of plant disease suppressive microorganisms and then large populations and high diversity with biological control potential [21]. Recently, the use of olive-mill wastewater (OMW), as suitable method to contain the P. ramosa, has been proposed [22].
In this regard in Italy, the Apulia is the most important region of olive oil production (200,000 t year -1 ), which extraction process from olive fruit (i.e. without preliminary treatments) produces a large quantity of OMW. This wastewater, derived mainly from the centrifuge extraction method, can be applied to the soil for agronomical use at the maximum amount of 80 m 3 ha -1 [23].
The OMW availability is restricted to November-March period, when the fields are mainly cultivated with cereal crops or are in pre-sown fallow before the spring crops. Therefore, for the spring-summer crops such as tomato, applied OMW must be made 35-40 days before planting, without contraindications [24], [25].
The OMW has chemical properties (i.e. organic carbon, potassium and phosphorus contents) that could increase soil fertility [26]. In addition, its content of phenolic compounds can act in the soil surface layer on the germination of the weed seed [27], [28] or suppress plant diseases and phytotoxicity [29], [30].
Considering very few information on the use of OMW for P. ramosa control, this study is mainly focused on wastewater application into the soil, before transplanting tomato crop in the field, to assess: I) whether this method is efficient about to P. ramosa containment; II) the effects on the quantitative and qualitative parameters of the tomato yield.
Applied into the soil 40 days before the tomato seedling transplantation, at 80 m 3 ha -1 . This is the maximum amount permitted to be used in the fields on capable soil (in agreement with Italian Law No 574, 1996).
The experimental sites were in a typical semi-arid zone, characterized by a Mediterranean climate, with a mild winter and dry-warm summer. According to the long-term climatic data, the annual rainfall and the mean temperature are 526 mm and 10.9 °C, respectively [31].
In table II, meteorological data (mean monthly maximum, minimum temperature, and total rainfall) are reported. They were collected during the experimental years at the nearest meteorological station, few kilometers from the experimental site and supplied by the "Consorzio per la Bonifica della Capitanata".
In all the experimental trials, the tomato crop (cv. "Dres") was transplanted in open fields in double rows 200 cm apart, with 40 cm spacing between the paired rows and 30 cm spacing in each row, resulting in a theoretical plant density of 3.3 plants m -2 . A randomized block design with three replicates (i.e. plots) for the OMW treatment and control was adopted. Each plot was 20 m 2 (2 m × 10 m) separated by two double rows (4 m distance) of tomato seedling that were not included in these experimental treatments. Drip irrigation method was used with the drip lines placed between each pair of plant rows. The water volume at each irrigation varied from 100 m 3 ha -1 to 300 m 3 ha -1 , depending on the crop growth stage, with a watering interval of about 3-4 days. The agricultural management practices applied to tomato crop during the experimental trials were those commonly adopted by local farmers, such as for fertilizing and for weed and pest control.

B. Sampling and Measurements
In each tomato growth cycle, at different day after transplant (DAT), the branched shoots of P. ramosa (hereafter referred to as ʹshootsʹ) emerged from the soil were counted on sampling area of 4 m 2 for each plot. Given the high uniform of the density of tomato plants per m2 in all experiments, the number of emerged shoots per unit of surface area can be considered an adequate method to estimate the parasite infection. The tomato fruit were harvested in the same sampling area at full-maturity stage. At harvesting, the marketable yield was determined. Moreover, for a sample of 10 marketable tomato fruit from each plot, the following qualitative parameters were measured: mean weight (g), dry-matter content (% fruit fresh matter); soluble solids content (°Brix); pH and titratable acidity (g citric acid per 100 mL fresh fruit juice) [32]. Fruit color were recorded in the CIELab system, using a Minolta CR-400 colorimeter (Konica Minolta Sensing, Inc., Japan). In the CIELab colour space, L* indicates the brightness, a* ranges from green (−a*) to red (+a*), and b* varies from blue (−b*) to yellow (+b*) [33], [34].
All the data were subjected to analysis of variance (ANOVA), using the JMP software (SAS Institute Inc. Cary, NC, USA) and the means were compared by Tukey's test at 5% probability. In Table III, the particular description of each experimental trial is reported. In particular, as for the monitoring dates of P. ramosa emerged shoots in the fields, they were generally related to the following tomato phonological phases: initial fruit set, complete fruit set, initial fruit ripening and harvesting time. In table IV, the main physic-chemical characteristics of the soils of the experimental fields are reported.

A. P. ramosa Emerged Shoots
In Fig. 1 are reported, for each treatment of the four trials, the number of emerged shoots per m2 at different DAT during the tomato crop cycles. Trial 1. The P. ramosa shoots were detected at 56, 78, 81 and 92 DAT. At 56 DAT a low infestation was observed in both OMW and control treatments. Later over time, the increase of infestation occurred particularly till 78 DAT, though with significant lower numbers of emerged shoots in the OMW treatment (18.6 m -2 ) compared to the control (28.6 m -2 ). Moreover, as reported also in Fig. 2, at the end of tomato cycle (92 DAT), the control showed a significant higher emerged shoots (33.6 m -2 ) compared to the OMW treatment (20.3 shoots m -2 ), with a reduction in the last of 39.6% was measured.
Trial 2. In this trial the infestation of P. ramosa was monitored at 74, 81, 93 and 103 DAT. AT 74 DAT, the numbers of emerged shoots in the untreated control was already significantly higher (4.4 shoots m -2 ) than in the OMW treatment (1.0 m -2 ). In the following days during the tomato cycle the greatest increase in the infestation was recorded up to 81 DAT in both treatments, although with a significant lower values in OMW (15.3 m -2 ) than the untreated ones (40.1 m -2 ). In the last part of the tomato cycle (from 81 DAT to 103 DAT) the shoot density was almost stable (Fig. 1).
Therefore, as reported also in Fig. 2, at the end of tomato cycle, in the OMW treatment a reduction of 62% of infestation was detected.
Trial 3. The P. ramosa shoots were detected at 56, 63, 75 and 116 DAT. At 56 DAT, a very low emerged shoots were observed, with no significant differences between the OMW treatment and the control. Afterward, through the tomato crop cycle, the emerged shoots slightly increased to 63 DAT and then increased later, with significant higher values in the control than the in OMW treatment. At the tomato harvest (116 DAT), as reported also in Fig. 2, 3.1 shoots m -2 were registered for the treated plots and 12.9 m-2 for the untreated ones, with the reduction in the last of 76.0 %. Trial 4. The P. ramosa emerged shoots were detected at 60, 70, 80 and 91 DAT. At first survey (60 DAT), shoot density was already significantly higher in the untreated control (3.2 m -2 ) than in the OMW treatment (1.1 m -2 ). Subsequently, during the tomato growing cycle, the emerged shoots increased. At harvest (91 DAT), as reported also in Fig. 2, the untreated control showed 18.3 shoots m -2 against a lower value of 12.0 shoots m -2 in the OMW treatment, with the shoots reduction in the last of 34.4% (Fig. 1). Moreover, the mean shoot values of all the trials, as average of each treatment at harvesting times, they were significantly lower in OMW (13.1 m -2 ) compared to the control (27.3 m -2 ).
Regarding the comparison among the four trials, as an average of treatments (Table V), significant differences of mean emerged shoot were noted.

B. Yield and Qualitative Traits of Tomato Fruit
Yield and qualitative traits of tomato fruit of each trial are reported in Table VI.
Trial 1. The marketable yield was clearly affected by the P. ramosa infestation, since the control gave the lowest value (55.5 t ha -1 ) significantly different from OMW treatment (67.7 t ha -1 ). The fruit quality parameters, such as mean fruit weight, dry matter, color index, pH and soluble solids content were no significantly different between OMW treatment and the untreated control.   Data are means ± Standard Errors of 6 sample. Means followed by the same letters in each column are not significantly different (P ≤ 0.05 Tukey's test).
Trial 2. The marketable yield, because of the different parasite infestation, was significantly higher in the OMW treated crop (81.4 t ha -1 ) compared with the untreated one (74.9 t ha -1 ). Also, in this trial the main qualitative fruit characteristics were not significantly different between the OMW treatment and the control.
Trial 3. The productivity of the tomato crop was enhanced by OMW treatment (96.8 t ha -1 ) in contrast to untreated control (69.4 t ha-1). Even in this experiment, for the qualitative fruit characteristics, there were no significant differences between treatments.
Trial 4. The positive effect on marketable yield was obtained in OMW treated crop (112.8 t ha -1 ) compared with untreated control (101.3 t ha -1 ). Also in this experiment, the main qualitative fruit characteristics were not significantly different between the OMW treatment and the control. As for the comparison among the trials (Tab. VII) the highest yield was obtained in the trial 4 (107 t ha -1 ), which differed significantly from the other ones (values ranged between 61.1 and 83.1 t ha -1 ). Regarding the main qualitative fruit characteristics, not significant differences were noted. Finally, neither of the trial factors considered (trial, treatment) showed any significant effects in their interactions (Tab. VII).
Moreover, the relationship between tomato marketable yield of the several trial and P ramosa infestation at harvesting time is such that the smaller numbers of emerged shoots for the OMW treatment resulted in increase marketable yield. In Fig. 3 is shown the negative correlation R2 = 0.299) between P. ramosa emerged shoots and marketable yield is reported.

IV. DISCUSSION
In the several trials carried out in this study, the OMW treatment showed a significant reduced number of emerged P. ramosa plant (with values ranged between 34 and 76%) respect to the untreated control, limiting the additional seed production of this parasite. Many studies [35] reported that the containment of infested areas and prevention of seed distribution should be major objectives of parasitic weed management strategies. This should be achieved through the prevention of seed distribution along with the direct control interventions against the parasite. The significant reduction of the P. ramosa emerged shoots in the OMW treated tomato crop, could be due to the organic and mineral compounds, as nitrogen, phosphorus and potassium contained in the OMW, that could improve the nutrient status of the tomato plants in addition to the effects of phenols present in the OMW that could produce a reduction of P. ramosa seed germination. This agrees with earlier studies which reported that mineral nutrients, organic matter and phenols introduced into the crop rhizosphere can promote severe physiological disorders for the germination of P. ramosa seeds, with reductions in the number and developing tubercles [16]- [29], [36], [37].
Some differences in the time of P. ramosa attack and shoots density obtained among years are likely linked to the different environmental conditions before transplant and during the crop cycle. Indeed, seeds of parasitic plant only germinate in response to germination stimulants (strigolactones) present in the exudates from host roots [38], [39] and to the release from dormancy following a period (conditioning) of wet conditions and optimal temperature of 20 °C, immediately before germination [40].
Moreover, the mean numbers of emerged P. ramosa shoots, although with differences among experiments and between treated and untreated plots, increased slower in the last period of the tomato cycles, because the high temperature values that occurred in this period could induces in a secondary dormancy of parasitic seeds.
The results show the significant higher tomato marketable yield in the OMW treated crop than that in untreated crop, where higher P. ramosa infestation was registered. The higher yield obtained in the trial 4 compared with the other trials was probably due to the earlier transplant data (April 10, 2017) of the tomato crop.
For the fruit characteristics, such as mean fruit weight, dry matter, color index, pH and soluble solids content there were no significant differences between OMW treatment and the control. These results disagree with a previous research performed on tomato [41], in which, however, the P. ramosa infestation was much higher (108-209 shoots per host plant) than that recorded in our trials (maximum value of 40 shoots per host plant) (data not shown).

V. CONCLUSION
In view of the importance of processing tomato as a major cash crop for farmers, and the heavy losses in the field mainly due to Phelipanche ramosa infestation in the Apulia region (southern Italy), it is very important to control this harmful weed. Thus, the search for sustainable methods to contain this parasite has become increasingly important.
The conclusion to be drawn from the present study is that the olive-mill wastewater (OMW) incorporated into the soil at the dose of 80 m3 ha -1 (amount permitted by Italian Law No 574, 1996), 40 days prior to tomato seedling transplant, appears to be effective for the reduction of the infestation of P. ramosa in this crop. Moreover, the above OMW spreading amount on suitable soils can have a generally positive effect on soil and crops [42]. However, considering that this technique does not provide a complete control of the parasite it is assumed that the positive effect can be improved by combining this treatment with other, especially with the aim of gradual and continuing reduction of the "seed bank" of the parasite in the soil. Therefore, many studies are still necessary in order to increase knowledge about the OMW specific use for P. ramosa control in processing tomato crops.