Silage Quality, Aerobic Stability, Gas Production and in vitro Digestibility of Pumpkin Waste Silages

This experiment was carried out to determine the silage quality of pumpkin waste with wheat straw (S) and alfalfa hay (A). The treatment groups as follows: 1) 100% pumpkin waste (PR), 2) 80%PR and 20% A (PW80+A20), 3) 80%PW and 20%S (PW80+S20), 4) 60%PW and 40%A (PW60+A40), 5) 60%PW and 40%S (PW60+S40), 6) 80%PW+10%A+10%S (PW80+A10+S10) and 7) 60%PW + 20%A + 20%S (PW40+A20+S20). The lowest dry matter content was observed in PW group (39.0 g/kg). The crude protein ratio in the alfalfa groups were higher than those of S and PW groups. The crude cellulose was lower and NDF higher in the PW group than those of other groups and ADF was highest in the 40%S group. Water soluble carbohydrate ratio in PW and PW60+A20+S20 groups were higher than those of other groups. In the PW group, the lactic acid ratio was higher than PW80+S10+A10 group. In the PW80+S20 group the highest gas production was observed at 48 th , 72 nd and 96 th hours. The energy values of PW group were higher than those of other groups. The aerobic stability of pumpkin silage was improved with hay addition especially alfalfa addition in 400 g/kg concentration.


Introduction
Pumpkin (Cucurbita pepo) have been producing to get seed, seed oil and snack. Pumpkin fruits contain a high carotenoid, digestible ber and carbohydrates, sugar, and pectin, also they are rich in vitamin and minerals [1,2]. It is reported that pumpkin fruits have about 9 to 12% dry matter, 11 to 16% crude protein, 14 to 19% crude cellulose [3,4] and 58% total digestible nutrient content [4]. After separation of seed from pumpkin fruit a large amount of waste material emerging (fruit meat and small seeds constitute 95% of the whole fresh fruit, determined by own). Fresh pumpkin wastes cannot be stored a long time due to the high-water content (87.5 to 95%) [3,4] and microbial deteriorations. Its sugar and pectin content, which most of them can be fermented by silage bacteria [5]. When it considers as a feed source it can bene t for ruminants a nutritive feed material especially in winter, which livestock may feed with poor quality forages (such as cereal straws) which have lack carotenoids and vitamin A. There were some efforts on suitability for silage and feed value of pumpkin waste [6-8], however get high quality silage from various vegetables, dry matter content of materials should be kept about 28 to 35% [9]. Thus, pumpkins are described di cult ensiling material due to high moisture content [10]. In practice, to get optimal dry matter, increase dry matter content and to prevent loss of nutrients with leakage; silage material should be admixed cereal hay or another hays. The aim of this study is to evaluate the nutritional, fermentation value and quality of pumpkin waste silage, ensiled with wheat straw and alfalfa hay and their combinations at various ratios.

Ensiling Process
The pumpkin wastes were obtained from a pumpkin seed harvest machine (Özkanlar Tarım, Kayseri, Turkey). The machine chopped of pumpkin fruits about 3-8 cm length and width. Before to ensiling process, moisture content of pumpkin fruits was measured (two days ago in the same eld pumpkin samples) and dry matter (DM) content was found between 4 to 7% (in experimental samples was 4.7%, Table 1). Pumpkin wastes were mixed with straw and alfalfa hay in different levels in order to increase dry matter. Pumpkin wastes were ensiled alone, with straw and alfalfa (20 and 40% and their combinations as 10 and 20%). The samples were ensiled in 5 liters of jars and remained for two months for fermentation at room temperature (between 20 to 24°C). were determined according to methods speci ed in AOAC [11]. Neutral detergent ber (NDF), acid detergent ber (ADF) and acid detergent lignin (ADL) were determined according to Van Soest and Robertson [12]. Water soluble carbohydrate (WSC) contents were determined by the phenol sulphuric acid method according to Dubois et al. [13] and aerobic stability were determined according to Kung et al. [14]. Lactic acid (LA) was determined by Lepper's methods [10]. The acetic, propionic and butyric acid analysis were done in a gas chromatograph with a capillary column (30 m × 0.25 mm × 0.25 µm, Restek), in a Shimadzu GC-2010+ (Kyoto, Japan).

Aerobic Stability Measurement
Silage samples (150 g) were incubated in glass jars at 24°C in an incubator (Nuve EN300, China). A Pt-100 probe (TMC6-HD) connected to a data logger measured the temperature in the centre of the material hourly. Aerobic stability was de ned as the time needed to increase the temperature 2°C above the temperature of a reference silage inoculated to prevent aerobic spoilage completely [14].

In vitro Gas Production, Energy Values and Organic Matter Digestibility
Organic matter digestibility (OMD) and metabolisable energy (ME) and net energy lactation (NEL) contents of silages in in vitro conditions were made according to the in vitro gas production technique reported by Menke and Steingass [15]. The tubes were incubated in a shaking water bath at 39°C and the amounts of gas formed by fermentation at 0 (start of incubation), 3, 6, 9, 12, 24, 48, 72 and 96 hours.
Cumulative gas production data were tted to the model of Ørskov and McDonald [16] by NEWAY computer package program.
Where, a, the gas production from the immediately soluble fraction (mL); b, the gas production from the insoluble fraction (mL); c, the gas production rate constant for the insoluble fraction (mL/h); a + b, potential gas production (mL); t, incubation time (h); y, gas produced at time "t". Organic matter digestibility (OMD), metabolizable energy (ME) [21] and net energy lactation (NEL) [15] contents of forages were estimated using equations given below:

Statistical Analysis
For statistical analysis of the data obtained from the study, one-way Anova was applied by using SPSS [17] program and Duncan multiple comparison test was used to determine the differences among the treatment groups. Probability was accepted as P < 0.05.

Results
For the pumpkin silages, it was observed that in the physical evaluation of the smell, color and structure, they had a pleasant and aromatic scent, a color changed natural yellow to light yellow and in the groups except pure pumpkin, the particles generally did not lose the structure. It was determined that the silage made with pure pumpkin showed a watery and gaseous structure in which the particles were scattered but there was no signi cant loss in the color. In this case a total of 7 points, "middle" degree has received. Nutrient composition of fresh pumpkin waste is given in Table 1. The dry matter of fresh pumpkin waste was very low (4.70%). The pH, chemical composition, water soluble carbohydrate, aerobic stability and eig score of pumpkin waste silage alone or with alfalfa and wheat straw are shown in Table 2. The pH value (4.54) in 40% alfalfa group was signi cantly higher than other groups and PW group's pH (3.90) was lower than that of other groups (P < 0.01). Dry matter content of silage samples was the lowest in ensiled alone pumpkin (7.27%) group and highest in PW60 + A40 and PW60 + S40 groups (37.87 and 36.13%) ( Table 2, P < 0.01). It was observed that the dry matter was low when it was added to the pumpkin waste with 20% concentration of alfalfa or straw hay. However, PW60 + A20 + S20 group pH value was similar PW60 + A40 and PW60 + S40 groups. The crude protein content of alfalfa groups was higher than those of other treatment groups and in the S groups was lower than those of other groups while the lowest protein content was found in the 200 and 400 g kg − 1 S groups. Pumpkin waste protein content was 9.61% and this ratio higher than those of S groups. There was no signi cant effect in terms of ether extract content of samples between the treatment groups. Crude cellulose (CC) percentage and its ratio in organic matter was highest in the 400 g kg − 1 S group. Ensiled alone pumpkin silage had lowest CC content (P < 0.01). The highest NDF content was observed in the pumpkin waste group and NDF content lowered by straw addition (P < 0.01), however, the ADF content of straw groups was higher than those of other groups (P < 0.05). Also, 20% and 40% straw addition to PW caused an increase in organic matter content compared to other groups (P < 0.01). Crude protein ratio in organic matter 20 and 40% alfalfa group were higher than those of 20 and 40% straw groups (P < 0.01). In this study, eig score value of silages was classi ed good (only PW) and the other groups have "very good" class. The eig score of PW60 + S40 group higher than of those other groups, whereas the eig score in the control group was signi cantly lower than those of the other groups (P < 0.01). Pumpkin waste silages lactic, acetic, butyric and propionic acid and ethanol concentrations are given in Table 3. The lactic acid concentration (45.01 g kg − 1 DM) of K60 + S40 group was lower than those of PW and PW80 + S10 + A10 groups. The ratio of acetic acid, butyric acid and propionic acid in the silage samples were not signi cantly differ among the groups. Ethanol ratio was highest in the PW group and this group's ethanol concentration signi cantly differ from PW80 + A20, PW80 + A10 + S10 and PW60 + A20 + S20 groups. Time-dependent gas production rate of silage samples is given in Table 4. In the group PW the gas production from quickly soluble fractions (a) (-10.22) was very high and in the PW60 + S20 + A20 group (4.21) lower (P < 0.05). According to the time dependent gas production, the maximum gas production (b) was in the PW80 + S20 (60.70) group and the lowest in the PW60 + A40 (48.91) group (P < 0.01). PW: pumpkin waste, A: alfalfa, S: wheat straw, a: the gas production from the quickly soluble fraction (mL); b: the gas production from the insoluble fraction (mL); c: the gas production rate constant for the insoluble fraction (mL h − 1 ); SEM, standard error of means; P: probability. RSD: residual standard error. a,b,c,d : The differences between means with different letters in the same column are signi cant.
The rate of gas production in the alone PW group was higher than those of the other groups in the 3, 6, 9, 12 and 24 h, however, PW80 + A20 The amount of gas released during the 96-hour fermentation period in the study material pumpkin waste silages is given in Table 5. The gas production rate of PW80 + S20 group was highest at 48th, 72nd and 96th hours (P < 0.01) and the gas production rate of PW60 + A40 group was lower at 72 and 96 hours (P < 0.01). The organic matter digestibility (OMD), metabolic energy (ME) and net energy lactation (NEL) values calculated using the in vitro gas production volumes of silages are given in Table 6. In the control group's the OMD, ME and NEL values were signi cantly higher than those of other groups (without PW80 + A20 group, P < 0.05). The aerobic stability of silages is given in Table 6. Aerobic stability was lowest in PW group silage (P < 0.05). The aerobic stability of straw and alfalfa groups was increased. It was observed that aerobic stability achieved the best results when straw, alfalfa or straw and clover were added with 40% of pumpkin silage (P < 0.05).

Discussion
Silage quality can be predicted objectively as a tool for preliminary assessment [9]. Subjective evaluation may give a clue for silage quality without any chemical analysis results. In this experiment there was no substantial change of pumpkin waste particle size in the silage samples. DLG score [18] in silages made with corn varieties sum of the smell, structure and color scores may contribute to predict silage quality classi cation and it can be detected by the sensory organs. The additives used to reduce the amount of water in the pumpkin waste silages changed the physical appearance of the silage, which made it possible to get a high score in terms of this criterions. Konca et al. [19] found that silage quality class was satisfactory in pea silages in unusual feed materials in silage production, bad and satisfactory in triticale silages, good for artichoke silage, and satisfactory for corn silages.
The pH value ranges of this experiment can be evaluated for success a good silage quality. According to recommendation of good quality silage pH ranges should be 3.5 to 4.0 [9]. It has been reported that when pH is not low enough, there was not a su cient fermentation, so, it is presumably assumed silage is not harvested at an appropriate time, or insu cient carbohydrate provided for fermentation and therefore lactic acid bacteria are not developed [9,20]. Especially, when occurred insu cient fermentation, the pH of the silage does not t decrease su ciently, not enough lactic acid formation but undesirable butyric and propionic acids increase in the silo [9]. In the current experiment the pH value of silages was generally around 4.0 and it can be said to be su cient fermentation well done.
The dry matter of ensiled alone pumpkin waste was very low (47.0 g/kg). Dry mater of silages was increased by the alfalfa hay and straw supplemented groups. Scharrer et al., [6] reported that the dry matter of fresh pumpkin and silage pumpkin 88.1 and 80.1 g/kg respectively. Also, Hashemi and Razzagzadeh [8] added 286% straw and 0, 10 and 20 molasses and 5% urea to pumpkin waste and they found that 10% molasses addition increased DM content of silages (37.68%). Church [4] reported that the pumpkin fruit dry matter content was 90 g/kg and Mokhtarpour [3] found 12.5%. Both of these researcher results of DM values are higher than the values in the current study. However, it has been reported that the variety of pumpkin and harvest time may signi cantly affect the content of DM [8]. Silage DM is accepted as an important criterion in silage compaction and aneorobic fermentation and harvest time signal. The expected rate of DM in a good quality maize silage is accepted as 300 to 350 g/kg [15]. From this point of view, it can be said that the DM rates in the silage samples high and low values from these ranges are not suitable for good quality silages. On the other hand, in the silages obtained from seasonally grown plants and sometimes insu cient maturation, so, lower DM ratio than that of desired values and dry rough feeds such as hay or straw were mixed into the silage material to prevent seepage losses due to high moisture content. In this study, wheat straw and alfalfa hay were used to reduce the high humidity content of pumpkin wastes and it was observed that the content of DM reached to the desired (DM > 350 g/kg) levels especially in groups with 40% straw and alfalfa (Table 2). Kara et al. [21] reported that broccoli, parsley, lettuce, leek, cauli ower, cabbage and spinach contain low dry matter than that of pumpkin wastes and from these reports it has been determined that pumpkin waste is a silo material with better quality of crude nutrient content than most of vegetable wastes.
Fleig score calculate based on DM and pH values of silage, the silages samples were classi ed generally good (only PW) and the most of groups have "very good" class quality. Hashemi and Razzagzadeh [8] found that ensiling of PW with molasses and urea pH value ranges between 7.61 and 6.24 while Invanchuk [22] determined the pH value in the pumpkin silage between 4.4 and 4.5 and Mokhtarpour [3] reported that pH of pumpkin prior to ensile was 5.9 and after fermentation ranged from 3.87 to 4.54 and these pH values similar as the current experimental pH values. Konca et al. [19] reported that eig scores in the silages produced (such as peas, artichoke, triticale, and corn) in dairy cattle farms of Turkey were between 11.0 and 98.5and they have been classi ed between the "bad" and the "very good" quality. Kara et al. [21] found that the quality class of lettuce, leek, cauli ower and cabbage was "satisfactory" and celery, broccoli and parsley and spinach was not suitable for make silage. The highest eig score was 83.94 in leek, and the lowest in 15.66 broccoli. These values obtained are more satisfactory than the values obtained for the other materials which are ensiled as by-products. In the commercial farms, maize silage quality may be changed between 'middle' and 'famously' quality classes [23].
The crude protein content alone PW group was 96.1 g/kg and their results were similar to Enishi et al. [24] results who reported that pumpkin waste crude protein content was 94 g/kg. Alfalfa added groups' crude protein ratio were higher than those of other groups. Alfalfa hay contains a high crude protein ratio compared to straw (132 vs 40.2 g/kg, respectively (analyzed results). Bakshi et al. [25] noted that protein content of pumpkin was 151 g/kg and Mokhtarpour [3] and Church [4] noted that crude protein of pumpkin silage about 110 to 160 g/kg. Other researchers reported low CP content of pumpkin and corn grits mixture (2015). However crude protein content was higher than that of corn silages [19] and similar to NDF and ADF content and crude ash.
The wheat straw addition decreased crude ash content of pumpkin waste silages. Crude fat content was not in uenced by the treatments. Lozicki et al. [5] reported that pumpkin fruit ash was 86.4 g/kg. Hashemi and Razzagzadeh [8] found that dry matter content of pumpkin (71.4%) + straw (28.6%) silage was 8.38%; and found the highest in the group with 10% molasses added as 37.68%. Enishi et al. [24] reported that pumpkin waste NDF content was 94 and 251 g/kg respectively and higher digestibility in goat rumen compared to sugar beet pulp and carrot juice waste. Lozicki et al. [5] determined that NDF and ADF content of pumpkin alone was 217 and 180 g/kg, respectively.
It is known that short chain volatile fatty acids such as acetic, propionic and butyric acid prevent aerobic deterioration in silage by suppressing the growth of yeast and mold in silage. The lowest lactic acid concentration (45.01 g/kg DM) was obtained in PW60 + S40 group and highest lactic acid ratio (58.97 g/kg DM) observed in the PW80 + A10 + S10 group. The ratio of acetic acid, butyric acid and propionic acid in the silage samples were not signi cantly differ between the groups. Ethanol ratio was highest in PW80 + A20 group and lowest ethanol concentration was obtained in PW90 + A10 group. Ivanchuck [22] found 10.2 g of lactic acid, 1.3 g of acetic acid and 0.83 g of butyric acid per kilogram of pumpkin silage. Brune et al. [7] noted that when pumpkin ensiled with stubble alfalfa and sun ower up to 1/3 ratio evaluated "very good quality" in terms of acid content, digestibility and feeding value. Researcher claimed that pumpkin and corn silage was better quality than corn alone, higher crude protein digestibility with 2: 1 or 1: 1 ratio. Butyric acid was not detected in the silage of PW alone, PW60 + S40 and PW80 + S20 + A20 groups. Butyric acid is generally low in feeds made with good silos.
Improvement of rumen conditions is important in terms of cellulase hydrolysis. In terms of cellulasis, it can be desired pH values should be changed between 6.6 and 6.8, and if increase or decrease occurs in the rumen pH value it may affect cellulose digestibility. In particular, the pH value is less than 5.9, which suggests a decrease in cellulite activity [26]. In this experiment, the pH value was higher than 6.0 which create desirable range. As seen in the Table 2, while depend WSC content of samples in group C the gas production from quickly soluble fractions (a) (-10.22) was very high and in the PW60 + S20 + A20 group (4.21) lower gas production was observed (P < 0.05). According to the time dependent gas production, the maximum gas production was in the PW80 + S20 (60.70) group and the lowest gas production was in the PW60 + A40 (48.91) group. The rate of gas production in the control group was higher than those of the other groups. The amount of gas released during the 96-hour fermentation period in the study material pumpkin waste silages is given in Table 4. The gas production rate of PW80 + S20 group was the highest at 48th, 72nd and 96th hours (P < 0.01) and the gas production rate of PW60 + A40 group was low at 72 and 96 hours (P < 0.01). In the control group's the OMD, ME and NEL values were signi cantly higher than those of other groups (P < 0.05).
Examining Table 2, it will be seen that there is no statistically signi cant difference between the groups in terms of WSC values for pumpkin waste silages. However, the WSC values obtained in this study were found to be closer to the values determined in legumes [27] than the WSC values of corn silage reported by other researchers [28]. In many cases, materials with high WSC content have the advantage of providing suitable fermentation development [29]. If there is not enough WSC in the medium, silage quality will decrease. In order to get a better-quality silage, it must be contained at least 30 to 50 g/kg fermentable carbohydrate in DM, especially hexoses. From this point, it can be said that the WSC content of the pumpkin waste silages is su cient to get good quality silage.
The time-dependent gas production rates are given in Table 4. When the in vitro gas production kinetics are considered, the highest group was the PW60 + S40 (pH = 6.79) group and the lowest value was in the PW80 + S20 (pH = 6.64) group. However, it has been found that the pH change intervals are very narrow and vary between 6.79 and 6.64. Differences in nutrient content of feeds may affect in vitro gas production parameters, energy values and OMD in silages [30]. Similarly, in this study, it can be seen that between the gas production values of the groups varied in different silage groups.
The silages OMD values were lower than in another study PW ensiled with wheat bran and urea [31]. In the present study, organic matter values were higher in the PW80 + S20 and PW60 + S40 groups and NDF values were lower than those of other groups. At the same time, the PW80 + S20 group' gas production at 48th, 72nd and 96th hours signi cantly higher than the other groups. In this matter, differences of silage samples nutrient compositions may explain of difference of gas production values. On the other hand, protein concentration of samples may affect of gas production and high percentage of feeds negatively effect of gas production [30]. Similarly, in this study, the PW60 + A40 group could explain the fact that it has less gas production value than those of the other groups.
Digestibility of organic matter, metabolisable energy and net energy lactation values were higher in PW alone silage samples these values were higher than those of other groups (without PW80 + A20 group). It shows PW alone silage have a better digestibility, NDF ratio and higher gas production. Both alfalfa hay and wheat straw mixture decreased these values. Especially, as well-known straw hay lower digestibility and feed value. So, pumpkin waste can be used a valuable silage material in terms of higher digestibility and energy value.
Aerobic stability is an important factor that determines the silage quality and feed value [32]. Aerobic stability indicates the capacity of oxygen-exposed silage to resist microbial growth. Silage is exposed to oxygen during storage or animal feeding. A number of changes occur in the silage material that has been exposed to oxygen. These changes are determined by following the silage temperature. In the method used to determine the aerobic stability of the silage material, the temperature change caused by the silage material exposed to oxygen is measured [33]. In addition to these measurements, pH measurements are also evaluated together [32]. As shown in Table 7, when the pH values of spoiled silages were examined, it was observed that spoilage occurred in all groups. When the groups were examined, the aerobic stability of the groups of straw and alfalfa was affected positively. It was observed that aerobic stability was best when 40% straw, alfalfa or straw and alfalfa were added to pumpkin waste silage alone. However, when the results are examined, it can be said that alfalfa is more effective than wheat straw to increasing the aerobic stability.

Conclusions
Agricultural wastes and agricultural industry by products remain in the production areas and may cause environmental pollution. This study showed that the pumpkin waste materials may be ensiled with 40% alfalfa hay and/or wheat straw. However, more research is needed for the effects of animal performance as in vivo studies.

Declarations Con ict of interest
There is no con ict of interest for the publication of this article.

Ethical Approval
No need.

Consent to Participate
All authors agree to participate in the current work. Consent for Publication All authors agree to publish the ndings of the current research.

Author Contributions
Conceptualization by YK, SBB; methodology by MB, MK, SBB; investigation and validation by MK, MB; Writing original draft, review editing by YK, MB and SBB.