Comparison of hermetic bags with polyethylene-lined polypropylene bags for prolonged storage of faba bean (Vicia faba L.) seeds in Ethiopia

Abstract

Poor storage facilities expose seeds to deterioration, resulting in low seed quality under a prolonged storage time. We compared the performance of five bag types: Two hermetic bags (Super GrainPro bag [SGB] and Purdue Improved Crop Storage bag [PICS]), with three non-hermetic bags (woven polypropylene bag lined with polyethylene [WPP lined with PE], polypropylene bags [PP], and jute bags) for the storage of faba bean seeds for 24 months. Faba bean seed quality was assessed under ambient conditions after 0, 6, 12, 18, and 24 months of storage. Parameters investigated include seed moisture content, thousand seed weight, germination percentages, vigor index I and II, speed of germination, percentage of insect-damaged seed, and insect counts. Results indicated that there was a significant (P ≤ 0.01) interaction of bag types and duration for all tested parameters. Hermetic bags and WPP lined with PE maintained germination percentage above 90% during 12 months of storage. The vigor index remained above 1600 mg% for hermetic bags and WPP lined with PE during 12 months of storage. However, non-hermetic bags exhibited a rapid drop in vigor index beginning from 6 months of storage. Hermetic bags and WPP lined with PE allowed minimal insect breeding and hence decreased the seed damage caused by insects. The present results demonstrated that in addition to the PICS bags and SGB, the WPP lined with PE can be used for short-period storage for up to 12 months without incurring a significant loss of seed quality. Hence, we recommend the introduction of WPP lined with PE as faba bean seed storage technique under conditions where hermetic bags are inaccessible to resource-poor farmers.

INTRODUCTION

Faba bean (Vicia faba L.) is a cool-season legume with several agroecosystem benefits, particularly when grown through crop rotation and intercropped with cereals like wheat and barley (Rawal & Navarro, 2019). In 2020, the annual global production of pulses was about 90 million tons. Of this, faba bean accounted for about 5.7 million tons. China (30.4%) and Ethiopia (18.9%) shared about 50.3% of global faba bean production (FAOSTAT, 2022). Faba bean takes the largest share of the area (504,570 ha) and production (1,070,636.5 t) of the pulses in Ethiopia (FAOSTAT, 2022).

Depending on the prevailing intrinsic and external factors, post-harvest losses of grain legumes are estimated to be about 9% for the United States and 40–50% for many developing countries (Ikegwu et al., 2022). The extent of postharvest loss in Ethiopia can be expressed as economic losses amounting to 10% of the country's annual budget, which could have greatly contributed to the gross domestic product (GDP) every year (Teferra, 2022). Postharvest loss of seeds affects the field planting value and future household food security.

Increased moisture and temperature are major causes of mold development, leading to a decline in legume grains germination and vigor (Ikegwu et al., 2022). Vigor may have a positive influence on emergence in the field. The magnitude of the influence is, however, modified by the soil environment, season, and crop types into which the crop is planted. Burris (1976) concluded beyond the differences in initial growth rate, very few consistent responses to vigor can be shown in field performance.

During storage, grain legume pests are capable of destroying up to 33–50% of global produce. The quality degradation that resulted in loss of the quantity of legume during storage globally can reach 60% (Ikegwu et al., 2022). Loss of seed quality can be caused by several pests including mold and insects. Seed beetles (Callosobruchus chinensis L. and Callosobruchus maculatus F.) in the store are the most damaging insect pests resulting in loss of faba bean seed quality in Ethiopia (Ali et al., 2006). Adult Callosobruchus beetles do not feed on stored seed and are very short-lived, usually no more than 25 days (Fox et al., 2004). The hatched larvae, however, have to physically penetrate the faba bean seed and cause seed quality losses.

In storing legume grains, the usage of suitable packaging materials are very crucial in minimizing post-harvest losses. In village areas of both developing and developed nations, grains including pulses are still kept in traditional storage facilities, which are fabricated with natural materials or woven threads for grains storage. Typical examples are underground pits, thatched roof, plastic container, and basket silos (Ikegwu et al., 2022). Farmers in Ethiopia are using traditional storage structures such as gotera and polypropylene bag for the storage of faba beans (Endshaw & Hiruy, 2021). Major types of traditional storage structures in Ethiopia are gorbo, gotera, and gumbi. Gorbo is made of vertical timber poles or twigs plus a thatched roof and resting on the ground or on cross-pole platforms. Gumbi is made of a mixture of mud, tef straw, and cow dung and is either made of a single piece or rings stacked one above the other so that the vessel can be taken to pieces and reassembled elsewhere. Gotera is made of interwoven sticks or bamboo and resting on cross-pole platforms or on the ground and has conical grass roof (Tadesse, 1997); earthenware pots and gourds, bark, sacks/bags, roof storage, underground pits, and small store houses are also some the examples of traditional storage strictures in Ethiopia. The traditional storage facilities may predispose the grain to different deterioration agents and cannot guarantee the protection of stored grains for longer durations (Kumar & Kalita, 2017). Thus, efforts should be made to provide farmers with proper storage methods to reduce the loss of seed quality during storage. However, limited information is available on the use of improved storage methods including hermetic containers for prolonged storage of faba bean seed in Ethiopia.

Hermetic storage is an alternative to synthetic pesticides for the storage of agricultural commodities (Jayas & Jeyamkondan, 2002; Murdock et al., 2012). Various hermetic options are being used globally for both commercial storage, namely, silo storage methods and cocoons (Navarro & Navarro, 2014), and small-scale storage, namely, metal silos, plastic silos, hermetic bags, GrainSafes, and plastic containers (Walker et al., 2018). Several branded products such as Super GrainPro bag (SGBs), Purdue Improved Crop Storage (PICS), Ecotact, ZeroFly, AgroZ, and AgroZ Plus storage methods are also recommended for use (Olorunfemi & Kayode, 2021). However, those technologies are rarely tested for prolonged storage of faba bean seed. Besides, reports are rarely available on prolonged storage studies such as 2 years using hermetic bags.

On the other hand, commercially available hermetic storage technologies have low adoption by farmers due to the required initial investment (Tenna et al., 2020). Besides educating farmers on the cost and benefits of using hermetic bags, alternatives should be available for resource-poor farmers to protect their seeds under short-time storage. Woven polypropylene bags lined with polyethylene sheets can be considered an alternative for resource-poor farmers with limited access to finance. Thus, the present study was aimed at comparing the quality of faba bean seed stored in two hermetic bags (Super GrainPro bag [SGB] and Purdue Improved Crop Storage bag [PICS]), with three non-hermetic bags (woven polypropylene bag lined with polyethylene [WPP lined with PE], polypropylene bags [PP], and jute bags) for the storage of faba bean seeds for 24 months storage duration.

MATERIAL AND METHODS Experimental materials

Faba bean variety Ashebeka was produced at Kulumsa Agricultural Research Center (KARC) farm in Kulumsa, Ethiopia, in the 2017 main production season. The seed farm is geographically located at 8°01′10″ N latitude and 39°09′11″ E longitude at an elevation of 2200 m a.s.l. The major soil type in the study area is classified as Vertic Luvisols. The seed was sown at end of June and harvested at the end of November (153 days) 2017 at moisture content of around 25%. The harvest was kept in the field until seed moisture was dropped to around 13%, and then the seed was threshed manually. The total rainfall received during the production months (June to November) was 699 mm with the highest total of 198 and 270 mm in August and September. The average daily minimum and maximum temperatures in these months were 10.1 ± 1.4°C and 21.7 ± 1.5°C, respectively, with lower in October and December.

A 750-kg pure pre-basic seed of the Ashebeka variety was used for the experiment. The seeds were first sieved to remove extraneous materials (stones, dusts, weeds, husks, insect debris, etc.) that were small in size. Second, cleaning was made by hand to make sure that the seed is free from any impurities and insect damage. After cleaning, seeds were carefully homogenized by shovel on a smooth and clean cemented floor before being placed into five different bag types described below.

The bag types were bought from a local market and suppliers in Addis Ababa. Locally manufactured sisal jute bags (Jute bags), polypropylene bags (PP bags), and fertilizer bags (WPP lined with PE) with a capacity of 50 kg were bought from the local market in Asella, Ethiopia. All hermetic bags were newly purchased 1 month before usage. The fertilizer bags are UV-stabilized woven polypropylene (WPP) bags with polyethylene (PE) inner lining (ca. 70 μm thickness). Super GrainPro bags (SGB) of 100-kg capacity were obtained from local agent HiTEC Trading PLC, Addis Ababa, Ethiopia. Super GrainPro bags are produced by GrainPro™ Company (Concord, Massachusetts, USA) and consist of a single layer of 78 μm of a plastic film made of two polyethylene films between which is sandwiched a plastic layer that is highly impermeable to oxygen (DeBruin et al., 2014). The Purdue Improved Crop Storage bags (PICS bags) consists of one woven polypropylene bag surrounded by two layers of high-density polyethylene (HDPE), each 80 μm thick (Baoua et al., 2013), and were obtained from local agent Shayashone Trading PLC, Addis Ababa, Ethiopia. Hermetic bags were carefully examined for any leaks, before filling, and in case there were leakages; they were not used in the experiment.

Thirty kilograms of seed from homogenized seed were filled into each bag, and the bags were tied properly using sisal strings for non-hermetic and plastic cable for hermetic bags to prevent gas movement. Each bag type was independently replicated for each storage duration (6 months) to avoid bag opening effects.

Treatments and experimental design

Two factors, each having five levels, that is, five bag types and five storage periods (an initial 0-month, 6, 12, 18, and 24 months) were used in the experiment. Faba bean variety Ashebeka produced at Kulumsa Agricultural Research Center and harvested in November 2017 was used for the storage experiment. The study was laid out in 5 * 5 * 4 with five storage periods (including 0-month storage), five bag types, and four reps per bag type (=100 samples total, that is, 20 samples at 0-month of storage before bags were filled and 80 samples [bags] after bags were filled) factorial experiment in a completely randomized design. All bags were placed on top of each other in a warehouse (at one corner of the hall) under ambient conditions (Figure 1) with daily minimum temperature of 12.1 ± 1.7°C and daily maximum temperature of 23.9 ± 2.3°C based on 671 records from March 2018 to February 2019.

FIGURE 1. Average monthly maximum and minimum temperature and relative humidity in the seed storage environment.

Sampling

Submitted sample, from each bag, was obtained by hand halving method as described by ISTA (2014). Seeds from the bag were poured evenly on the clean smooth surface and thoroughly mixed into a pile. The pile was divided in to half and halve each half again—giving four portions—and halve each portion again—giving eight portions; four portions were randomly combine and retained and the other four were removed. This procedure was repeated using the retained portions until, approximately but not less than, a kg of seed was obtained. A seed sample of 1 kg was obtained from each experimental bag at the time of sampling. Sampling was conducted at every 6-month intervals beginning from the beginning of the storage experiment. The initial sample (0-month storage) was taken for initial seed quality analysis before the inception of the storage study. Sampled packages were removed from the experiment right after samples were taken (i.e., samples taken from new bags every time). The seed samples were used for the assessment of seed quality and enumeration of insects. Each sample was partitioned into pure seeds, insects, and impurities (broken seed, inert matter, dust, insect wastes) using 4-, 1.7-, and 1-mm test sieves (Endecotts Limited, UK).

Data collection Seed moisture (%)

Seed moisture was determined using DRAMINSKI GMM mini moisture meter (DRAMINSKI S.A., Poland), which was newly purchased at the outset of the experiment, and it was calibrated a year after. Seed moisture contents were measured immediately after opening the bags (before all other tests) to avoid moisture adjustment with the external environment. Four measurements per sample were taken and expressed in (%) at one decimal place. Samples were transferred to the moisture meter using a small plastic box. Each measurement consisted of about 270 ml of seed (to fill the measurement chamber of the moisture meter).

Germination percentage (%)

Germination tests were made using the pure seed proportions of the samples. Two hundred seeds were counted per sample and replicated in four, 50 seeds in each replication. The seeds were sown in moistened sand in plastic germination boxes and kept at 20 (°C) in the germination room for 14 days. Normal seedlings were counted according to rules for testing seeds (ISTA, 2014) and expressed as percentage germination.

Seed vigor indices

The seed vigor index was based on the germination test conducted in Section 2.4.2, and it was calculated according to Kalsa and Abebie (2012) determining the germination percentage, seedling dry weight, and seedling length of the same seed sample. Ten randomly selected normal seedlings were used to determine the average seedling length and average seedling dry weight.

Seedling length included the root length and shoot length. The average seedling length was calculated based on the 10 seedlings. After removing remnants of the cotyledons, the 10 seedlings were placed in an envelope and oven-dried at 105°C for 24 h. The oven-dried seedlings were weight and the average seedling dry weight was calculated based on the 10 seedlings.

Seed vigor index-I was calculated by multiplying germination (%) by mean seedling length (cm). Vigor index-II was calculated by multiplying the germination percentage by the seedling dry weight of the same seed sample.

Speed of germination

The speed of germination was conducted on pure seed components of the seed samples. Fifty seeds each in four replications were sown in a germination box on moistened sand and the box was covered for germination. The substratum was kept in a germination room maintained at a temperature of 20 (°C). The number of seedlings emerging daily was counted from the day of planting till the time germination ceased. Thereafter, a germination index (SGI) is computed by using the following formula as described by Kalsa and Abebie (2012).

$$\mathit{\text{SGI}}=\Sigma (\mathit{\text{Ni}}/\mathit{\text{Di}})$$ where Ni is the number of normal seedlings on day D; D = days after planting.

Adjusted thousand seed weight (TSW)

TSW was measured from the pure seed component of each seed sample. TSW was adjusted according to Hellevang's (1995) adjusted quantity formula using actual moisture at the time of test and standard moisture content. For safe storage of pulses, the optimum moisture content needs to be in the range of 9%–12% (Ikegwu et al., 2022). The standard seed moisture content (i.e., base moisture) of faba bean is 12% based on Ethiopian Standards, faba bean seed specification (ES, 2012). Moisture adjustment was made using the following equation: [Image Omitted. See PDF]

Insect damage seeds (IDS) percentage

The number of insect-damaged seeds (windowed/hole) and undamaged seeds were counted in a 1 kg sample. Insect-damaged seeds (IDSs) were expressed as a percentage of the total number of seeds in a kg of sample.

Enumeration of insects

Right after obtaining 1 kg of a sample, the number of live and dead adults was counted at each sampling occasion by sifting the 1-kg sample through test sieves. Live and dead adults separated from beans were enumerated. Immobile insects were pocked with brushes to check if they are hibernating or dead. When the movement was detected as pocked, the insects were counted as live.

Data analysis

All data were subjected to two-way analysis of variance (ANOVA) to detect significant effects of methods and durations of storage. The homogeneity of variances among different storage strategies or storage periods was tested using Bartlett's test. Insect count data were log-transformed before the analysis of variance. The ad-hoc comparison was employed to separate the means when factor effects were significant at P = 0.05, and Tukey's adjustment of P values was used. Pearson's correlation coefficients were used to determine the association of other parameters with seed quality (germination and seedling vigor index). R software Version 4.1.1 was used for data analysis. Graphs were plotted using Sigma Plot software Version 12.5 (Systat Software, Inc., Chicago, IL, USA).

RESULTS Seed moisture (%)

The average seed moisture content at the beginning of the experiment was 11.4%. A significant interaction effect (F_{16, 75} = 8.41; P < 0.01) was detected between the types of bags used and the duration of storage. During the first 6 months of storage, seed moisture ranged from 10.8% in PICS bags to 11.6% in PP bags. Seeds stored in Jute bags also demonstrated high moisture gain during the first 6 months of storage (Figure 2). There was a significant (P < 0.05) decline in moisture content as the storage period was prolonged from 6 months to 12 months. At 12 months after storage, the seed moisture was below 10% in all bag types. The lowest values (8.7% and 8.8%) were recorded for seeds stored in the PP bags and Jute bags, respectively. At 18 months of storage, the highest mean value of seed moisture content was recorded for seeds stored in the Jute bags. No significant changes in mean values of seed moisture content were recorded in any of the containers between 12 and 18 months of storage.

FIGURE 2. Mean (±SE) of seed moisture (%) as affected by storage period and types of bags.

Germination percentage

The germination percentage at the outset of the experiment was 95%. There was a significant interaction between storage duration and bag types (F_{16, 75} = 9.3; P < 0.01) in terms of germination percentage. Improved bag types such as polypropylene bag lined with polyethylene sheets, PICS bag, and Super GrainPro bags maintained initial germination percentages up to 12 months of storage (Figure 3). A significant decline in germination percentage was detected after 12 months, regardless of the bag type. A significant decline in germination percentage was observed in traditional bag types such as Jute bags and polypropylene bags.

FIGURE 3. Mean (±SE) of germination percentage as affected by storage period and types of bags.

Speed of germination

The speed of germination at the outset of the study was 18.8 seedlings per day. There was a significant interaction (F_{16, 75} = 5.91; P < 0.01) between the storage period and the types of bags. Seeds stored in WPP lined with PE demonstrated a significant decline in the speed of germination after six and 12 months of storage (Table 1). The speed of germination of seeds stored in WPP lined with PE, PICS, and SGB bags did not demonstrate a significant reduction after 12 months of storage. Changes in the speed of germination at 18 and 24 months of storage were not conclusive because of undulating mean values.

TABLE 1 Mean of the speed of germination (seedlings/day) of faba bean seed stored under different methods and storage periods.

Storage type	Storage period (months)	Speed of germination (seedling/day)
WPP lined with PE	M06	17.2 abcd
WPP bag	M06	11.4 i
Jute bag	M06	13.9 fghi
PICS bag	M06	18.0 abc
Super GrainPro (SGB)	M06	18.3 ab
WPP lined with PE	M12	18.6 a
WPP bag	M12	15.6 cdefg
Jute bag	M12	17.6 abcd
PICS bag	M12	18.0 abc
Super GrainPro (SGB)	M12	18.1 abc
WPP lined with PE	M18	14.5 efgh
WPP bag	M18	12.6 hi
Jute bag	M18	14.3 fgh
PICS bag	M18	15.8 bcdefg
Super GrainPro (SGB)	M18	15.1 defgh
WPP lined with PE	M24	16.5 abcdef
WPP bag	M24	13.6 ghi
Jute bag	M24	17.3 abcd
PICS bag	M24	18.7 a
Super GrainPro (SGB)	M24	17.1 abcde
CV (%)		5.95
F16,75		5.91

Vigor indices

The mean value of vigor index I (multiple of the germination percentage and seedling length in mm) of faba bean seeds at the beginning of the storage period was 3792% mm. The interaction effect of the storage period and the methods of storage on vigor index I was significant (F_16,75 = 3.75; P < 0.01). Except for seeds stored in PICS bags, there was a significant (P < 0.05) decline in vigor index I in all bag types at 6 months of storage.

The mean value of Vigor index-II (a multiple of germination percentage and seedling dry weight in mg) of faba bean seed at the outset of the storage period was 2742% mg. There was a significant interaction effect of duration and bag type (F_16,75 = 21.22; P < 0.01) on vigor index II. Except for PICS and Super GrainPro (SGB) bags, there was a significant decline in the mean values of vigor index II during the subsequent periods of storage (Figure 4). The PICS bags and SGB maintained vigor index II as high as 2587% mg and 2958% mg, respectively.

FIGURE 4. Mean (±SE) of vigor index II (% mg) as affected by storage period and types of bags.

TSW

The mean TSW at the outset of the storage experiment was 864 g. There was a significant interaction between durations and methods of faba bean seed storage (F_16,75 = 6.88; P < 0.01) for a TSW. The TSW exhibited a significant decline during the first 6 months of storage in all bag types except the Jute bag (Figure 5). After 24 months of storage, the percentage loss in TSW of faba bean was about 13.0% in PICS bags while the loss ranged from 17.8% to 18.8% in all other types of bags.

FIGURE 5. Mean (±SE) of thousand seed weight as affected by storage period and types of bags.

IDS percentage

No damaged seeds were detected at the outset of the storage period. However, the effect of different bag types on the percentage of damaged seeds was significantly dependent on the period of storage (F_{16, 75} = 10.14; P < 0.01). Seed stored in WPP lined with PE and Jute bag resulted in a higher rate of IDS ranging from 80.61% to 85.05% and 77.12% to 80.57%, respectively, after 24 months of storage (Table 2). IDS ranged from 5.37% to 23.87% for seeds stored in a PICS bag. Seeds stored in hermetic bags and WPP lined with PE bag exhibited comparable quality in terms of seed damage (Figure 6). Storing seeds in WPP lined with PE for 12 months also resulted in comparably fewer IDS percentages (22.62%).

TABLE 2 Mean percentage of insect-damaged seeds under different methods and durations of storage.

Storage type	Storage period	Insect damaged seed (%)
PP lined with PE	M06	19.4 cde
PP bag	M06	91.3 a
Jute bag	M06	88.2 a
PICS bag	M06	6.0 de
Super GrainPro	M06	6.3 de
PP lined with PE	M12	24.3 cde
PP bag	M12	89.0 a
Jute bag	M12	85.3 a
PICS bag	M12	25.8 cde
Super GrainPro	M12	7.5 de
PP lined with PE	M18	46.7 bc
PP bag	M18	87.4 a
Jute bag	M18	81.5 a
PICS bag	M18	10.4 de
Super GrainPro	M18	34.6 cd
PP lined with PE	M24	73.3 ab
PP bag	M24	87.0 a
Jute bag	M24	80.4 ab
PICS bag	M24	23.6 cde
Super GrainPro	M24	75.3 ab
CV (%)		30.55
F16,75		10.39

FIGURE 6. Insect-damaged seed (IDS) of faba bean after 6 months of storage.

Enumeration of insects

The storage insect species detected in samples during the study were mainly bruchids (Callosobruchus spp). No live adult insects were detected at the outset of the experiment because the seed was properly cleaned before the storage experiment. There was a significant interaction between methods and duration of storage (F_16,75 = 3.8; P < 0.01). Dead insects largely outnumbered the live insect population at all storage periods (6 to 24 months) regardless of the bag type. The mean number of live insects between 6 and 24 months of storage ranged from 0.0 insects per kg of seed to 34.2 insects per kg of seed. The highest mean number of dead insects (4389 insects per kg of seed) was recorded in a polypropylene bag at 18 months of storage while the lowest (eight insects per kg of seed) was recorded in a Super GrainPro bag after 6 months of storage (Figure 7).

FIGURE 7. Mean (SE) of the number of adult insects per kg of seed (both dead and live).

Parameter association

The associations between parameters are depicted in Table 3 and Figure 8. Strong correlations were detected between insect count and IDS and germination percentage, vigor index I, vigor index II, speed of germination, and TSW. A strong and negative correlation (r = −0.84) was detected between vigor index II and IDS (%). Weak correlations of live insect count with seed quality parameters were detected unlike that of total insect count (Figure 7).

TABLE 3 Pearson's correlation coefficient (r) for seed quality parameters.

Seed quality parameters	Insect count (number)		Insect damaged seed (IDS) (%)
	t value	r	t value	r
Germination (%)	−6.22**	−0.53	−11.41**	−0.76
Vigor index I	−5.40**	−0.48	−11.46**	−0.76
Vigor index II	−6.97**	−0.56	−15.12**	−0.84
Speed of germination	−4.33*	−0.40	−8.89**	−0.67
Thousand seed weight	−6.27**	−0.54	−6.77**	−0.56

^*t value significant at alpha = 0.05.

^**t value significant at alpha = 0.01.

FIGURE 8. Correlation plots for different parameters. MC = seed moisture content, AdTSW = adjusted thousand seed weight, germ = standard germination, VI = vigor index I, VII = vigor index II, Spg = speed of germination, ND = insect damaged seed percentage (by number), NLiveIns = number of live insects, Total = insect count.

DISCUSSION

Previous studies indicated that hermetic bag types provide alternatives to the use of chemicals in protecting seeds from insect damage and loss of germination (Kalsa et al., 2019; Mendesil et al., 2022; Prasantha et al., 2014). The present study stressed the performance of faba bean seeds stored in hermetic bags, and non-hermetic bags over a 2-year period of storage. Seed moisture content, germination percentage, speed of germination, and seedling vigor index demonstrated comparable performance in hermetic and WPP lined with PE.

In hermetic bags and WPP lined with PE, there were steady declines in moisture content as the storage period progressed up to 18 months. This was a rare outcome under hermetic storage conditions though Prasantha et al. (2014) also reported a slight decrease in the moisture content of mungbean seed from 12% to 11% within 6 months of storage. However, the level of variation in seed moisture in hermetic bags after such a prolonged storage is rarely available except that Bakhtavar et al. (2019) reported a slight change of seed moisture in Super GrainPro bags. In our present study, the temperature difference between the outset of storage and the subsequent time might have resulted in a new hygroscopic balance as reported by Coradi et al. (2020). Improper sealing of the hermetic bags can result in an increase in moisture content, but not such a steady decrease unlike the case observed in this study.

In non-hermetic bags, moisture slightly increased during the first and the third 6 months. This could be attributed to that both the first and third 6 months coincide with the main rainy season (June and September) at the study site. Seeds readily absorb moisture under a humid environment resulting in comparatively a higher moisture record (Baributsa & Baoua, 2022; Coradi et al., 2020). There was higher seed moisture in WPP lined with PE and Super GrainPro at 24 months after storage, which might be due to insect infestations. Kiobia et al. (2020) also reported fluctuations in moisture content in non-hermetic containers.

Increasing damages were observed in the Super Grain Pro bag as the storage time increased, especially after 18 months of storage. This might be due to that the hermetic bags are not perfectly impermeable to oxygen and moisture, or bag condition during storage (invisible damage by insects, or any other form) might have contributed to permeability of the bags (Kiobia et al., 2020). Baoua et al. (2014) also suggested that the high-density polyethylene (HDPE) layers are not perfect oxygen barriers but greatly reduce the rate of oxygen exchange with the surrounding environment. In contrast, the seed moisture in the PICS bag was lower compared to non-hermetic bag types.

In our study, hermetic bags and WPP lined with PE maintained a germination percentage above 90% for 12 months and above 80% for 18 months. Our findings are in agreement with previous studies that hermetic bags maintained seed quality during storage (Baributsa & Baoua, 2022; Coradi et al., 2020; Mendesil et al., 2022). The comparable performance of the hermetic bags and WPP lined with PE provides farmers with an easily accessible and cheaper alternative. However, in non-hermetic bags, the germination percentage was observed to drop below 90% within 6 months of storage and below 80% within 18 months of storage. Baributsa and Baoua (2022) also reported a drop in the germination percentage of soybean seeds stored in polypropylene bags. Loss of germination in polypropylene bags was faster compared to that of jute bags.

Hermetic bag types protect seeds from insect infestation and damage due to slow insect activity under oxygen limitations within the sealed environment (Harish et al., 2014). Insect infestation in Super GrainPro bag started increasing from 12 months of storage (Figure 7).  This might be attributed to bag damage by insects. Hermetic grain bag seems to be resistant to the perforation of Sitophilus zeamais, but not to Prostephanus truncatus (Ognakossan et al., 2013). In the present study, the hermetic bags and WPP lined with PE suppressed weevil prevalence for 6 months. Similarly, the percentage of IDS was significantly lower in the WPP lined with PE compared with the WPP lined with PE and Jute bags. In Ethiopia, hermetic bags have been tested and recommended for use, but the adoption is lower among farmers due to limited access to finance for initial investment (Tenna et al., 2020). While the promotion of the hermetic bags is unequivocally important, the use of a WPP lined with PE is also recommendable because these materials can be available from the local markets.

CONCLUSION

In our study, the moisture content of faba bean seeds stored in hermetic bags and WPP lined with PE decreased steadily from 11.4% of the initial moisture to 9.2% in PICS bags after 18 months. We speculated such loss of moisture in a hermetic environment is attributed to hygroscopic balance under varying temperatures. However, additional studies are required to understand the dynamics of relative humidity and gas composition within bags under such a prolonged storage period.

Hermetic bags and WPP lined with PE maintained germination percentages above 90% during 12 months of storage. The vigor index remained above 1600 mg% for hermetic bags and WPP lined with PE during 12 months of storage. However, non-hermetic bags exhibited a rapid drop in vigor index beginning from 6 months of storage.

Hermetic bags and WPP lined with PE allowed minimal insect breeding and hence decreased the seed damage caused by insects. The relationships between seed quality parameters, insect counts, and IDS were negative and significant (P < 0.01) indicating that the loss of seed quality was mostly due to the prevalence of weevils.

AUTHOR CONTRIBUTIONS

Astawus Esatu Seifu designed and conducted the experiment, Karta Kaske Kalsa contributed to data analysis and manuscript writing, Hasen Seid Aweke supervised data collection and contributed in the writ up, and Girma Debele Dibaba collected the data.

ACKNOWLEDGMENTS

The study was financially supported by the Ethiopian Institute of Agricultural Research (EIAR) under the activity code 51/05/018 from 2017 to 2019.