How do relocation time and length of storage after relocation affect fermentation and nutritive value of corn silage?

ABSTRACT This study aimed to determine the effects of relocation time (RT; Experiment 1) and storage time after relocation (ST; Experiment 2) on microbial population, fermentative characteristics, and chemical composition of corn silage. In experiment 1, corn silage was stored for 30 d, subjected to different RT (0–60 h), and stored again for 30 d. Thirty experimental silos were used in a completely randomized design, with three replicates per treatment. In experiment 2, after 150 d of ensiling, silage was removed from a bunker silo, exposed to air for 9 h, relocated to experimental silos, and stored for periods ranging from 0 to 128 d. Twenty-eight experimental silos were used in a completely randomized design, with four replicates per treatment. Relocation time had no effect on fungi counts and concentrations of lactic and propionic acids in corn silage but resulted in a significant increase in dry matter content. In experiment 2, dry matter recovery and concentration of non-fiber carbohydrates decreased in corn silage stored for more than 32 d after relocation. Exposure of corn silage to air during relocation for up to 60 h followed by 30 d of storage did not compromise the fermentation profile or nutritive value of the silage. Increased storage time of relocated corn silage (up to 128 d) consistently decreases its nutritional value. The storage period seems to have an increased impact on nutrient loss in relocated silage than the relocation period.


Introduction
Silage relocation is a common practice among farms across the world.It involves moving the silage to a new silo for different purposes.In tropical regions, farmers often face low forage availability because of incorrect feed planning or field crop losses caused by climatic occurrences, pests, diseases, and other factors, forcing them to buy silage from other farms (Michel et al., 2017).Selling silage is a common practice in some regions of Brazil because of the climatic variation, low availability of area, and adequate machinery and labor for the ensiling process (dos Anjos et al., 2018).
Silage relocation involves unloading from the original silo, transportation, re-compaction, and sealing in the new silo (Chen and Weinberg, 2014;Queiroz et al., 2021).These steps can take hours or days to complete, which exposes the silage to air during the process (Chen and Weinberg, 2014).This delay can result in a reduction in the nutritional value of the silage due to the oxidation of residual soluble sugars and degradation of lactic acid produced during fermentation, increasing the proportion of cellular content.Corn is the most common crop used for silage production in dairy farms in Brazil (Bernardes and do Rêgo, 2014) and is more prone to aerobic deterioration than other types of silage, especially in tropical conditions (Bernardes et al., 2009;Kung Jr. et al., 2018).However, research has shown that relocation lasting up to 48 h has no effect on fermentation, chemical composition, and dry matter recovery (DMR) of corn silage (Lima et al., 2016).
More studies are required to understand the changes in silage caused by prolonged exposure to air that can occur in different field situations, such as long-distance transportation, excessive rain, broken machinery, and labor unavailability.In addition, changes in the relocated silage over time can strongly affect silage quality, but this topic has not yet been studied.
We hypothesized that relocation and storage times could increase nutrient loss and modify the chemical composition of relocated corn silage.Therefore, we aimed to evaluate the effects of relocation time (RT; Experiment 1) and storage time after relocation (ST; Experiment 2) on the fermentation and nutritive value of the relocated corn silage.

Experimental design and procedures (experiment 1)
Corn used as silage in this experiment was cultivated in a region located at 3°2'2" S latitude and 47°20'18" W longitude in the city of Paragominas, PA, Brazil.The climate of this region, according to Köppen classification, is tropical dry winter (Aw), characterized by a rainy tropical climate with an extensive dry season.This region has an annual average maximum and minimum temperature of 33 and 22 °C, respectively, and average relative humidity of 81%.The average total annual rainfall is less than 1.743 mm, with the rainy season recorded from January to May (Bastos et al., 2006).
The corn hybrid (PIONEER 30F90H ® ) was planted in January 2016 with a spacing of 60 cm between rows and harvested at 324 g/kg of dry matter (DM) using a self-propelled harvester (FX40, New Holland Agriculture, Italy).
A completely randomized design with ten treatments and three replicates, totaling 30 experimental units, was used.The treatments were: non-relocated (NR), and relocation times (RT) of 0 (silage was removed from the silo, homogenized, and immediately placed back in the silo), 6,12,18,24,30,36,48, and 60 h.After relocation, silos were stored in a covered barn with an ambient temperature of 28.1±0.167°C for another 30 d.The total storage time (ST; ST after ensiling + ST after relocation) was 60 d.
Thirty-five experimental silos (plastic buckets with a capacity of 15 L each) were filled with 9.11±0.0681kg of fresh forage to reach a density of 600 kg/m³ on a fresh weight basis.After filling, the silos were weighed and stored in a covered barn at ambient temperature.After 30 d of storage, the silage destined for relocation was removed from the experimental silos, mixed, and exposed as piles.After exposure, each pile was mixed again and returned to the original silo.Five extra buckets of silage were exposed to air to top up the buckets after the relocation process in case of excessive shrinkage or losses due to dehydration.During aerobic exposure of silages, the temperature of the piles was measured at two equidistant points with a thermometer inserted into the silage mass during relocation.The initial temperature (IT) was recorded at the beginning of the relocation process, and final temperature (FT) was recorded before silage was returned to silos.

Experimental design and procedures (experiment 2)
The corn used in this experiment was cultivated in the city of Igarapé Açu, PA, Brazil, located at 1°7'44"S latitude and 47°37'12" W longitude.The climate of this region, according to Köppen classification, is also tropical dry winter (Aw), characterized by a rainy tropical climate with an extensive dry season.The annual average temperature and total rainfall are 27.3 °C and 2.270 mm, respectively.The corn variety (AL Bandeirante) was planted at a spacing of 60 cm between rows, with a total plant density of 55 plants/ha.Corn was harvested at 350 g/kg DM using a tractor (3500,New Holland,Curitiba,Brazil) and an automatic harvester (JF C120, JF Máquinas, São Paulo, Brazil).
A completely randomized design with seven treatments and four replicates, totaling 28 experimental units, was used.Treatments were: NR silage, collected from the silo panel before relocation, and different ST after the relocation of 4, 8, 16, 32, 64, and 128 d.Silos were stored in a covered barn at ambient temperature, and total ST (ST after ensiling + ST after relocation) was 154, 158, 166, 182, 214, and 278 d for each respective ST.
Corn silage was prepared in a bunker silo that was opened after 150 d.At the time of first opening, any visibly spoiled silage at the top of the bunker was discarded, and a 20 cm layer was removed from the silo panel.Subsequently, the silo panel and top were exposed for 9 h to simulate a typical relocation process.Subsequently, the exposed silage was relocated to 24 experimental silos, and plastic buckets with a capacity of 15 L each were filled with 8.87±0.0542kg of fresh forage to reach a density of 600 kg/m³ on a fresh weight basis.

Experimental analysis
In both experiments, fresh silage (1.5 kg) was collected from each silo in the second opening of the experimental silos and transferred to a 9 L cleaned plastic bucket.In experiment 1, silages were exposed to air for 12 d in a room with a controlled temperature of 22.1±0.0281°C; silage temperature was recorded every 30 min with a thermometer (TESTO, Campinas, Brazil) inserted into the geometric center of the silage mass.In experiment 2, silages were exposed to air for 7 d in a room with a controlled temperature of 25.8±0.0780°C; silage temperature was recorded every 30 min using a data logger thermometer (Klimalogg, Incoterm, São Paulo, Brazil).In both experiments, the ambient temperature was recorded.Aerobic stability (AS) was defined as the time (hours) the silage temperature remained stable before increasing by more than 2 °C above the ambient temperature (O'Kiely, 1993).The following variables were determined: time to reach maximum temperature (HMaxT) in h, maximum (MaxT) and minimum (MinT) temperature in °C, and amplitude in °C, which was the difference between maximum and minimum temperatures.

Microbial populations and fermentation profiles
After the final opening of silos and relocation, samples from both experiments were collected to analyze microbial counts (molds and yeasts) and fermentation profile.The aqueous extract (1:10) of the sample was prepared by adding peptone water (1 g per L of water) and a Stomacher homogenizer (MA 440/CF; Marconi, Piracicaba, São Paulo, Brazil) for 4 min in a sterile bag.The counts of molds and yeasts were determined by pour-plating 10-fold serial dilutions of the extract on potato dextrose agar (Sigma-Aldrich Brasil LTDA).After incubation at 26 °C for 5 d, the colonies that grew were counted separately for yeasts and molds, based on their morphological characteristics.
The pH was measured after each RT and 30 d after relocation when the silos were opened (Experiment 1), and after each ST (Experiment 2).The pH was also measured after 12 d (pH-12d, Experiment 1) or after 7 d (pH-7d, Experiment 2) of aerobic exposure.Approximately 25 g of fresh silage was mixed with 100 mL of distilled water (Bolsen et al., 1992), and the pH was measured with an electrode after 30 min (Tekna model T-1000).Dry matter recovery was estimated according to the equation described by Zanine et al. (2010).
To determine the concentration of organic acids (lactic, acetic, propionic, and butyric acids), we prepared an aqueous extract with 25 g of silage and 225 mL of deionized water and homogenized it in a Stomacher for 4 min (MA 440/CF; Marconi, Piracicaba, São Paulo, Brazil).The extract was then filtered, and only the liquid fraction was obtained, from which 2 mL of the sample was removed.Lactic acid was quantified by the colorimetric method according to Pryce (1969), and acetic, propionic, and butyric acids were determined using a gas chromatograph coupled to a mass spectrometer (GCMS QP 2010 plus; Shimadzu, Kyoto, Japan) and separated using a capillary column (Stabilwax; Restek, Bellefonte, PA, USA; 60 m, 0.25 mm, internal diameter 0.25 m).

Chemical composition
In both experiments, the chemical composition of silages was analyzed before and after relocation.To determine the chemical composition, we collected 200 g of silage from openings 1 and 2 of the silos and froze them at −20 °C for further analysis.Samples were dried in a forced-air oven at 55 °C for 72 h and subsequently processed in a mill with a 1 mm sieve (Thomas Wiley Mill Model 4; Thomas Scientific, Swedesboro, NJ, USA).

Statistical analyses
The pH and temperature data during relocation were subjected to descriptive analysis.Microbial count data were log 10 -transformed for analysis.Data from the two experiments were analyzed separately as completely randomized designs using the PROC GLM procedure in SAS (Statistical Analysis System, version 9.2).Data were subjected to analysis of variance (ANOVA) according to the following statistical model: in which y ij is the observed value j of a dependent variable after an RT of i, µ is the overall mean of the dependent variable, T i is the fixed effect of treatment (RT for Experiment 1 and ST for Experiment 2), and e ij is the random error associated with each observation.
Differences between control and treatments were evaluated using Dunnett's test (P<0.05).Orthogonal contrast analysis was used to determine whether RT and ST had linear or quadratic effects on the measured parameters.Differences were considered statistically significant at P<0.05.

Experiment 1
The DM concentration of fresh corn before ensiling was 325 g/kg.Corn contained 75.5 g/kg of CP, 18.6 g/kg of EE, 523 g/kg of NDF, and 343 g/kg of NFC (Table 1).The DM concentrations ranged between 295 and 320 g/kg.The pH after RT treatments ranged from 3.44 to 3.75.Silages with higher RT showed higher FT and MaxT values.Values of MinT showed no significant variation among silage piles.
An overall significant effect of RT (P>0.05) was not observed on the counts of molds or yeasts, the concentration of lactic and propionic acids, and the aerobic stability, amplitude, HmaxT, MaxT, and pH-12d values of the silage (Table 2).Concentration of acetic acid in silage with an RT of 12 h was lower than that in control.Silage pH after 60 h of RT was also higher than that of the control.The pH increased linearly as a function of RT.Aerobic stability had a quadratic relationship with RT.In contrast, increased RT resulted in a linear reduction in HMaxT.The DMR at RT of 12, 48, and 60 h was higher than that in the NR silage (P<0.01).We did not observe a significant effect of RT (P>0.05) on concentrations of OM, NFC, IVDMD, and WSC in silage (Table 3).The RT of 60 h resulted in a significant increase in DM (P = 0.05) and CP (P = 0.04) concentrations compared with that in the control silage.The concentration of EE in silage with an RT of 36 h also differed significantly from that of the control (P = 0.01).Silage exposed for 60 h during relocation showed a higher CP value than that of the control.The concentration of NDF in the silages with RT of 30 and 48 h was higher than that in the control (P<0.01).The RT of 6-60 h resulted in higher levels of NH 3 -N in the silage than that in the control (P<0.01).The EE concentration decreased linearly as a function of increasing RT.

Experiment 2
Our results showed no significant effect of ST on yeast count or lactic acid (P>0.05)(Table 4).Counts of molds in silages with ST of 4, 8, 16, 32, and 64 d were lower than those in the control (P<0.01).Concentrations of propionic and butyric acids in silages with ST 0-128 d were less than 1 g/kg of DM (not shown in Table 4).The acetic acid concentration in silages with ST of 8, 32, 64, and 128 d was greater than that in the control (P<0.01).The pH measured at the opening of silos was higher in the control silage than that in silages with ST of 4, 8, and 128 d.Only the silage with 32 d of ST showed a higher pH value than that of the control silage.Aerobic stability in silages with ST of 8, 16, 32, 64, and 128 d was significantly greater than that in the control (P<0.01).In contrast, the amplitude in silages with ST of 64 and 128 d was lower than that in the control (P<0.01).The HmaxT of the silage with ST for 4 d was higher than that of the control (P<0.01),whereas the MaxT was lower than that of the control only in the silage with 128 d of ST (P<0.01).pH-7d differed from that of the control in silages with 4 and 128 d of ST (P<0.01).An increase in ST resulted in a linear reduction in amplitude, MaxT, pH-7d, and DMR of the silage, whereas mold counts, concentration of acetic acid, and values of aerobic stability and HmaxT increased linearly with ST.
We observed no significant effect of ST on EE, NH 3 -N, or WSC concentrations in silage (P>0.05;Table 5).
The DM concentration in silages with ST of 4,16,32,64,and 128 d was lower than that in the control (P<0.01).The OM concentration in the silages with ST of 32, 64, and 128 d was lower than that in the control (P<0.01),whereas for the same ST, the CP concentration was higher than that in the control (P<0.01).The NDF concentration in silages with ST of 64 and 128 d was higher than that in the control (P<0.01),whereas for the same ST, the NFC was lower than that in the control.The IVDMD was reduced when the relocated silage was stored for 64 d compared with that of the control.Increased ST resulted in linear reductions in the DM, OM, and NFC concentrations of the silage, whereas the CP and NDF values increased linearly as a function of ST.
How do relocation time and length of storage after relocation affect fermentation and nutritive value... Santos et al.

Experiment 1
The action of the hydrolytic enzymes produced by plants is effective when the material is ensiled.
Structural and non-structural carbohydrates serve as energy sources for microorganisms during fermentation (Rooke and Hatfield, 2003).This hydrolysis causes a reduction in the NDF values, which was reflected in the results of the silages in the first opening of the silos, 30 d after the first ensiling (Table 1) in the current study.
Aerobic deterioration of silage occurs when it is exposed to atmospheric oxygen, which triggers a series of reactions such as an increase in the population of yeasts, temperature, and pH (Pahlow et al., 2003).
Changes in temperature are the first response to aerobic deterioration.An increase in RT may result  in other changes, such as changes in populations of molds and other aerobic bacteria, which intensify the production of heat and increase the temperature of the silages, as observed in the present study.However, this change in microbial population was not observed after relocation, even in silages with longer RT (60 h).Therefore, we discuss whether such changes could influence the characteristics of the silage after silo opening.
In the present study, modifications that could compromise the fermentation profile and preservation of the silage did not occur, even though they were exposed to air for up to 60 h, as no differences were observed in their microbial populations and fermentation profiles relative to those in the control silage.This demonstrates that aerobic exposure before relocation did not affect the fermentation profile or silage conservation.The highest pH value (3.9) observed was still within the recommended pH range for corn silage (3.8-4.2;McDonald et al., 1991).Other studies have reported few changes in fermentation profile and microbial populations in maize and sorghum silage relocated for up to 48 h (Chen and Weinberg, 2014;Lima et al., 2016;dos Anjos et al., 2018).However, reductions in concentrations of lactic acid and NH 3 -N in the silage and increases in effluent losses were observed in sorghum silage relocated for up to 12 h (dos Anjos et al., 2018).The low acetic acid content at RT of 12 h is not biologically explainable.
Although no changes occurred in fermentation products in the silage, its aerobic stability increased up to 36 h of RT and decreased for longer RT, 48 and 60 h.This increase in aerobic stability probably occurred because of oxygen penetration into the silage mass during aerobic exposure at the time of relocation.Oxygen promotes the proliferation of deterioration-initiating microorganisms, such as yeasts, which metabolize soluble carbohydrates, starch (McDonald et al., 1991), and organic acids (Driehuis et al., 1999), producing carbon dioxide, water, and heat (Muck, 2010).With the consumption of fermentative products, the pH of the silage is raised to values above 4.5, favoring the growth of a wide variety of other aerobic microorganisms, which accelerates the deterioration process and causes even greater heating in the silage (Muck, 2010).Thus, the higher stability of the silage with RT longer than 24 h can be explained by the lower availability of substrates for yeasts under these conditions, because these would have been consumed during the relocation and subsequent storage processes, which could be observed in the chemical composition of the silage observed after relocation (Table 3).It is possible that RT longer than 48 h made the silages more prone to aerobic deterioration, considering that their pH was above 5.0 after 5 d of aerobic exposure, whereas silages with RT lower than 48 h had a pH below 4.5 on the same day.It suggests that longer RT increases aerobic deterioration because of the long period that the silage remains exposed to air during the process, which will affect microbial populations and other silage characteristics that accelerate the deterioration process.However, in the current study, the result could not be attributed to any parameter evaluated in the silage.
The increase in the DM content of the silage as a function of RT occurred because of dehydration resulting from the contact with air during the aerobic exposure of silage.Recovery values of DM above 100% demonstrated that prolonged aerobic exposure during relocation probably caused water loss in the silage and concentrated its DM.Chen and Weinberg (2014) also observed the drying of silage during aerobic exposure.Changes in EE, CP, and NDF concentrations in the silage were proportional to the changes in its DM content and were related to decrease in levels of other nutrients, such as soluble carbohydrates, which are included in the NFC.
Small changes in chemical composition did not affect the nutritive value of the silage, demonstrating that its final quality was not compromised by relocation, which corroborates the results of previous studies.Therefore, well-preserved silages, such as corn, are safe to relocate within 60 h without major changes in fermentation characteristics and chemical composition.

Experiment 2
The acetic acid concentration increased after 32 d of storage compared with that in the control, which resulted in a lower pH than that of the control at silo opening and increased aerobic stability (Table 4).
Changes in these components after long periods of storage indicate that some microbial activity persists in silage, even when the pH is low (Kung Jr., 2013).
In a study in which corn silage was stored for up to 360 d, Der Bedrosian et al. ( 2012) reported an increase in the concentration of acetic acid and 1,2-propanediol in the silage as the storage time increased.The main explanation for this increase is the metabolism of heterofermentative lactic acid species, such as Lactobacillus buchneri that can metabolize lactic acid to acetic acid and 1,2-propanediol under anaerobic conditions (Oude Elferink et al., 2001).
Other studies have reported that some strains of Lactobacillus plantarum, Lactobacillus bifermentans, and Streptococcus faecium under anaerobic conditions and at low WSC concentrations begin to produce formic acid and acetic acid using lactic acid as a substrate (Kandler et al., 1983;Lindgren et al., 1990;London, 1968;Murphy and Condon, 1984).
The antifungal action of acetic acid has been confirmed by the easy penetration of the cell membranes of yeasts when the pH is lower than the dissociation constant (pKa) of these acids.At that point, they take on their dissociated form, causing the release of H + ions inside the yeast cells (Walker, 1998).As a consequence of the entrance of H + ions, the pH of the yeast cells decreases, compromising their growth due to the energy required for the active transport carried out by the cell in an attempt to eliminate H + ions from its interior (McDonald et al., 1991).
The increase in aerobic stability and HMaxT of relocated corn silages can be explained by the increase in acetic acid concentration with storage time.However, the fact that relocated silage is more stable cannot be considered beneficial because the loss of nutrients and reduction in the concentration of NFC were higher in silage with longer ST.This indicates that the relocation of silage increased the degradation of nutrients, even under anaerobic conditions in the silo, as a function of ST.The IVDMD value was also lower for silage with an ST of 64 d, showing that the observed reduction in the NFC reduced the IVDMD value of the silage that was relocated and subsequently stored.Thus, it is important to know the storage period and chemical composition of relocated silage before formulating livestock diets.
The decreased DM content of the silage with longer storage (Table 5) is related to nutrient losses in general, which can be observed by the reduction in DMR values with increasing ST.When evaluating the storage time of corn silage, Sariçiçek et al. (2016) also observed a decrease in the DM concentration of silage after long periods of storage.The reduction in DMR after long-term storage could be explained by slight modifications in the fermentation profile and an increase in acetic acid concentration.Non-fiber carbohydrate is the main source of fermentable substrates in silage, while WSC, glucose, fructose, sucrose, and fructan are also important sources of energy for microorganisms in the rumen (Bernardes et al., 2018).A reduction in the availability of these compounds in silage decreases their nutritive value.

Conclusions
The relocation of well-preserved corn silage for up to 60 h has no effect on the fermentative profile and nutritive value of the silage.However, dry matter content increases as a function of relocation time owing to aerobic exposure during relocation process.Increased storage time of the relocated corn silage, up to 128 days, has no effect on silage preservation.However, it decreases the nutritional value of relocated silage.Storage period after relocation has a greater impact on the nutritional value of relocated corn silage than time of exposure to air during relocation. 7

Table 1 -
Chemical composition (g/kg of dry matter) of corn silage before relocation and temperature (°C) and pH of the silage piles after relocation time (RT; Experiment 1) FC -fresh-chopped corn plant; ND -not determined; DM -dry matter; OM -organic matter; EE -ether extract; CP -crude protein; NDF -neutral detergent insoluble fiber; WSC -water-soluble carbohydrates; IT -initial temperature (moment when silages were exposed to air); FT -final temperature (moment when silages were recompacted -after their respective times of exposure to air); MaxT -maximum temperature; MinTminimum temperature; SEM -standard error of the mean.

Table 2 -
Microbial populations (log 10 cfu/g of fresh weight), fermentation products (g/kg of dry matter), aerobic stability (h), and dry matter recovery (DMR; %) of corn silage ensiled for 30 days, subjected to relocation times (RT), and stored again for 30 days (Experiment 1) HmaxT -time to reach the maximum temperature; MaxT -maximum temperature; pH-12d -pH measured after 12 days of aerobic exposure; DMR -dry matter recovery; SEM -standard error of the mean; L -linear effect; Q -quadratic effect. 1 Non-relocated silage stored for 60 days after ensiling.2Comparedwith a significance level of 5%.‡ Differs from the control (NR) by Dunnett's test.

Table 3 -
Chemical composition (g/kg of dry matter) and in vitro dry matter digestibility (IVDMD) of corn silage stored for 30 days, subjected to relocation times (RT), and stored again for 30 days (Experiment 1) DM -dry matter; OM -organic matter; EE -ether extract; CP -crude protein; NDF -neutral detergent insoluble fiber; NFC -Non-fibrous carbohydrates; NH 3 -N -ammonia nitrogen; TN -total nitrogen; WSC -water-soluble carbohydrates; SEM -standard error of the mean; L -linear effect; Q -quadratic effect. 1 Non-relocated silage stored for 60 days after ensiling.2Comparedwith a significance level of 5%.‡ Differs from the control (NR) by Dunnett's test.

Table 4 -
Microbial composition (log 10 ufc/g), fermentation products (g/kg of dry matter), aerobic stability, and dry matter recovery (DMR; %) of corn silage stored for 150 days and subjected to storage times (ST) after relocation (Experiment 2) HmaxT -time to reach the maximum temperature; MaxT -maximum temperature; pH-7d -pH measured after seven days of aerobic exposure; DMR -dry matter recovery; SEM -standard error of the mean; L -linear effect; Q -quadratic effect. 1 Non-relocated silage.2Comparedwith a significance level of 5%.‡ Differs from the control (NR) by the Dunnett's test.

Table 5 -
Chemical composition (g/kg of dry matter) and in vitro dry matter digestibility (IVDMD) of corn silage stored for 150 days and subjected to different storage times (ST) after relocation (Experiment 2) DM -dry matter; OM -organic matter; EE -ether extract; CP -crude protein; NDF -neutral detergent insoluble fiber; NFC -Non-fibrous carbohydrates; NH 3 -N -ammonia nitrogen; TN -total nitrogen; WSC -water-soluble carbohydrates; SEM -standard error of the mean; L -linear effect; Q -quadratic effect. 1 Non-relocated silage.2Comparedwith a significance level of 5%.‡ Differs from the control (NR) by Dunnett's test.