Comparative quality assessment of traditional vs. improved dried Bombay duck (Harpodon nehereus) under different storage conditions: Solar chimney dryer a low‐cost improved approach for nutritional dried fish

Abstract Dried fish and fishery products provide important and affordable dietary sources of animal protein. The present study compared the sensory, microbial, and biochemical quality of bombay duck (Harpadon nehereous) dried fish produced using improved versus traditional open‐air drying method. The sensory, microbial, and biochemical quality was determined following quality index, total plate count (TPC), and Association of Official Analytical Chemists (AOAC) method, respectively. The sensory quality results indicate highly acceptable dried fish with the improved method compared to traditional method at both initial and storage condition. The microbial load (APC, TEC, TViC, and TSC) of improved dried fish was within internationally permissible microbiological standards for food whereas traditionally dried fish exceeded the permissible limit. The biochemical composition (protein, lipid, ash, and moisture) of improved dried fish had better nutritional value than the traditional dried fish. After 60 days storage time, samples packaged under HDPE conditions exhibited excellent organoleptic characteristics, lower microbial load, and lower biochemical decomposition than samples packaged in LDPE. The above results confirmed that the solar chimney dryer produced superior quality of dried fish compared to the traditional open‐air drying method both initially and after storage, while packaging dried fish under HDPE conditions extended shelf life further.

Sun drying is a traditional method for preservation of fish and other perishable foods which is commonly employed in any developing countries around the world (Lamidi et al., 2019). Solar drying technologies do have considerable potential for application in smallscale food processing units in rural areas, as alternative to expensive mechanical drying technologies under controlled conditions.
Existing commercial fish drying in Bangladesh mostly involves drying in the open air, making use of energy from sunlight without any mechanical processing (Balachandran, 2001). Traditionally, sun drying is done directly on sand at the sea beach and/or on mats woven from hogla (Typha angustata) leaves without or with minimum washing and gutting of raw materials (Nowsad, 2007).
In view of public health and food safety aspects, the quality of fish and fishery products is a major concern for local and global consumers (Huss et al., 2003). Traditionally dried fish is associated with hygiene, quality, and food safety (chemical contamination) problems arising at different steps along the supply chain from preparation to consumption (Begum et al., 2017). Overall, low quality or spoiled raw materials, lack of personal hygiene, use of harmful chemicals, and improper processing, packaging, and storage conditions are the major constraints for quality management, resulting as a consequence in low market value of dried fish (Majumdar et al., 2017;Nowsad, 2007). Long and uncontrolled drying times and nonuniform exposure to temperature and sun leave the product exposed to contamination, resulting in a low quality of dried products. Commercial processors often apply several harmful insecticides (nogos, nuvacorn, endrin, malathion, dichloro diphenyl trichloroethane, and basudin) in dried fish to prevent insect infestation and microbial contaminations during drying and storage which are hazardous for human health (Begum et al., 2017).
Several studies have been conducted on sensory, nutritional, and microbial quality aspects, of sun-dried Bombay duck from coastal area of Bangladesh (Jamil et al., 2017;Paul et al., 2018;Pravakar et al., 2013;Rasul et al., 2020;Siddique & Aktar, 2011). The physical and organoleptic qualities of most of the traditional sun-dried products in Bangladesh are not attractive to the consumer (Hasan et al., 2006). Microbial contamination is another major cause of spoilage and deterioration of nutritional quality of dried fish. Among the biochemical constituents (moisture, crude protein, lipid, and ash) of dried fish, moisture content has the largest impact on its nutritive value and microbial load. Even properly dried fish deteriorates if it is not well packaged and stored. Effective packaging and storage reduce oxidation, rehydration, microbial, and biochemical spoilage and protect the product from physical damage and nutritional losses (Omodara, 2015). As a low-cost and improved technology, the UC Davis solar chimney dryer is designed to provide efficient drying of food materials even in hazy or partially cloudy conditions, without external energy inputs other than direct sunlight. The design of the dryer is such that it creates a continuous airflow, which increases the speed of drying and rapidly removes moisture. The dryer can be constructed from locally available materials that reduce the dryer construction cost and drying time, while minimizing insect infestation and pollution with dust (Horticulture Innovation Lab, 2019).
Considering the above factors, the purpose of the present study was to compare the sensory, microbial, and biochemical quality and nutritional properties of fish dried using an improved low-cost solar drying technology, the UC Davis solar chimney dryer, against quality and nutritional properties of traditional open-air dried fish, under different packaging conditions and storage periods.

| UC Davis solar chimney dryer construction
A full-sized dryer up to 12 feet long, consisting of three components, namely chimney, drying table, and trays, was constructed according to specifications in the construction manual (Horticulture Innovation Lab, 2019) provided online by the designer. First, the chimney was made up of a vertical frame of four tall poles, raising at full height about 6 ft (2 m) above the top of the drying bed. The width of the chimney, here 2 ft wide, should be equal to the width of the drying table and trays. The chimney was lined with clear polythene plastic sheet, and the top opening was covered with nylon mesh. The chimney was anchored to the ground, by digging in the poles to stabilize the structure.
Second, the drying table, measuring 12 ft long and 2 ft wide, was constructed by stretching black plastic or dark fabric tightly over the top, sides, and ends of the table frame and stapled into place.
Third, the trays were made using four (4) strips of wood, each measuring 2 ft in length. The floor of the trays consisted of iron mesh, cut in squares 2x2 ft length x width, and stapled to two strips, pulled apart tightening the mesh, and then secured by the other two strips ( Figure 1a,b). The trays can be stacked in layers, up to three depending on the volume of raw materials to be dried. The table was then placed in horizontal position and connected carefully to the chimney in order to prevent any air leaks through the seals. At the junction of

| Collection and processing of raw material
For this study, samples of traditionally produced dried bombay duck (Harpadon nehereus) fish, varying from 16 to 20 cm in length and 50 to 60 fish/kg dry weight, were collected from a seasonal dried  then stored at room temperature in an insulated plastic box for further quality (sensory, microbial and biochemical) analysis.

| Quality analysis
2.3.1 | Sensory quality assessment of traditional and improved dried fish Sensory tests were carried out by 15 trained panelists (seven male and eight female), using the Quality Index method (Howgate et al., 1992) with some modification of defects characteristics related to dried fish (Table 2). Sensory quality of dried fish was calculated, and grading decisions were made based on defect points (Table 3) by using the following formula-2.3.2 | Microbiological quality assessment of traditional and improved dried fish The bacteriological study was conducted following the procedure of standard plate count (SPC) (Cappuccino & Sherman, 1992), and dried fish samples were prepared following method of ISO (1995). Total bacterial colony count was performed in the form of aerobic plate count (APC) on nutrient agar media. About one (1) gram of blended dried fish sample was mixed thoroughly with 9 ml of sterile 1.5% peptone water obtained at 1:10 dilution. Then, 1 ml of supernatant was taken from the centrifuge tube and a 10-fold serial dilution of the sample was performed with 0.9% physiological saline. Aliquots of 0.1 ml of the serial dilutions were inoculated (triplicate) onto nutrient agar media for APC. The plates were incubated in the incubator (JSGI-10T, JSR, Korea) in an inverted position at 37°C for 24-48 hr.
APC was calculated by using the following formula: For specific bacterial count, 0.1 ml of the stock solution onto eosin methylene blue (EMB), xylose lysine deoxycholate (XLD), and thiosulfate citrate bile salts sucrose (TCBS) agar media were used for total E. coli count (TEC), total Salmonella count (TSC), and total Vibrio count (TViC), respectively. After 48-72 hr incubation, the colony colors on respective selective media were used for identification.
The colorless or pale pink colonies on XLD agar were identified as

| Statistical analysis of experimental data
The obtained data were subjected to analysis of variance (ANOVA), and mean comparisons were carried out by Duncans' multiple range test using SPSS package (SPSS 16.00 for windows, SPSS Inc., Chicago, IL, USA) software. Significant difference was defined at p < .05.

| UC Davis solar chimney dryer
The solar chimney dryer can be assembled from locally available construction materials according to the specifications mentioned in section 2.1, at a cost of 150-200 USD. The production capacity of a dryer of this size varies between 60 and 80 kg raw fish depending on species and size of fish to be dried. A higher capacity of this dryer in terms of volume production can be obtained by increasing the length of the drying

| Sensory quality of traditional and improved dried fish
In the present study, sensory quality of traditional and improved dried fish is compared and shown/reprent in Table 4 and Figure 2.
The result indicated that the improved dried fish produced from UC Davis solar chimney dryer was rated "excellent" and highly acceptable by the panelists than the traditionally produced dried fish. and firm, tender texture, which was preferred by panelists due to its excellent and highly acceptable organoleptic characteristics. At the initial storage time (Day 0), traditional dried fish already displayed a blackish color and slight to moderately soft texture, which was rated as "moderately acceptable" for consumption by the panelists (Table 4, Figure 2). According to Pravakar et al. (2013), high quality dried bombay duck have a slightly silver to whitish color, while Huque et al. (2013) report whitish to yellowish color, and Islam et al. (2006) whitish to light brown color to indicate high quality.
However, after 60 days storage time, moderate-to-severe deterioration in organoleptic characteristics were observed in terms of color (dark brown), texture (slight to moderately soft), odor, flavor, and general appearance for both products (Table 4, Figure 2).
No insect infestation was found in dried fish under any production method, packaging materials, and storage time used. After storage (at days 60) in HDPE packaged conditions, the traditional dried fish was rated as "moderately acceptable" by the panelists. The HDPE packaged, improved dried fish after storage for the same time (60 days) was still rated to be as "highly acceptable" as at initial storage (Day 0  Islam et al. (2006) reported that quality dried fish products have a firm and flexible texture, have a natural odor and no insect infestation or broken pieces are detected. Those characteristics are consistent with odor, texture, and no insect infestation found in improved processed and packaged dried fish in this study. In addition, improved dried fish (both HDPE and LDPE packaged) was still highly acceptable after storage time. Likewise, Islam et al. (2006) reported quality dried marine fish remain good after three months storage. However, deterioration in color, texture, odor, and overall acceptability occurred in most dried fish samples after 5 to 7 months storage time . In the present study, traditionally produced dried fish were found to have deteriorated beyond acceptable quality already after 2 months. The results of this study thus indicate that solar dryer improved processed and HDPE packaged dried fish exhibits better sensorial quality over traditionally produced and LDPE packaged products after 60 days of storage time.

| Bacteriological study of traditional and improved dried fish
In the present study, the APC, TSC, TEC, and TViC of traditional and improved dried fish under different packaging conditions and storage time are presented (Table 5). At the initial storage time (at Day 0), the APC, TSC, TEC, and TViC in traditional dried fish were 7.72 × 10 7 CFU/g, 5.0 × 10 4 CFU/g, 4.8 × 10 5 CFU/g, and 3.9×103 CFU/g, respectively, which exceeds the permissible limit (lCMSF, 1986). The higher APC, TSC, TEC, and TViC in traditionally produced dried fish might be due to poor hygienic sanitary standards and condition, improper handling, and low quality of raw materials used. In the traditional process, cross-contamination and unhygienic circumstances used for drying, insufficient use of salt, and insufficient drying time may result in poor quality dried fish.
In comparison, prior to storage (at Day 0), UC Davis solar chimney dryer produced improved dried fish contained APC and TViC at levels of 4.32 × 10 4 CFU/g and 1.3 × 10 2 CFU/g, respectively, were within the permissible limit for dried fish (lCMSF, 1986). TEC and TSC were absent/too few to count (TFTC). The permissible limit of TSC for cooked or dried fish is 1x10 5 CFU/g at 37°C (Surendran et al., 2006). Good quality dried fish should be totally free from Salmonella spp (IS, 2001). The APC, TSC, TEC, and TViC counts in improved dried fish were thus all kept within acceptable limits due to proper hygiene and sanitation practiced during handling and preparation of dried fish using the UC Davis solar chimney dryer. The result of the present study was similar with the study of Patterson and Ranjitha (2009) who found higher E. coli counts in commercially sun-dried fish than experimental sun-dried fish. The bacterial loads found in the current study were similar to the study of Huque et al. (2013) who found lower APC in improved solar tunnel dryer produced dried Bombay duck fish than in traditionally sun-dried fish. Pravakar et al. (2013) reported much lower bacterial loads, at 3 × 104 cfu/g, in traditionally produced sun-dried bombay duck fish than those found in traditional dried fish in the present study. This variation in result might occur due to variation in location, processing/preprocessing, and personnel hygiene involved between the studies.
After storage time (at Day 60), traditionally dried fish packaged in HDPE conditions had APC, TSC, TEC, and TViC values of 3.9 × 10 5 CFU/g, 3.0 × 10 3 CFU/g, 2.8 × 10 3 CFU/g, and 1.8 × 10 3 CFU/g, respectively, which were considerably lower than the value at Day 0. The higher oxygen barrier capability of HDPE packaging materials might result in lower microbial growth in the respective system. However, at the same 60 days storage time, traditionally dried fish samples packaged in LDPE showed higher F I G U R E 2 Differences in quality of dried fish produced from traditional and improved solar drying method under LDPE and HDPE packaging at 60 days storage time

Traditional drying
Improved solar drying LDPE HDPE LDPE HDPE Dried fish from both method Grinded dried fish Grinded dried fish APC 5.6 × 10 6 CFU/g, TSC 4.7 × 10 4 CFU/g, TEC 3.6 × 10 4 CFU/g, and TViC2.7 × 10 4 CFU/g values than those packaged in HDPE packaged conditions. In comparison, the APC in HDPE packaged improved dried fish was 2.3 × 10 3 CFU/g, where no TSC, TEC, and TViC were observed. LDPE packaged improved dried fish had APC, TSC, TEC, and TViC at 4.8 × 10 4 CFU/g, 0.5 × 10 1 CFU/g, 1.0 × 10 2 CFU/g, and 0.9 × 10 2 CFU/g, respectively, which were considerably higher than in HDPE packaged samples. Both traditional and improved dried fish had lower APC, TSC, TEC, and TViC under HDPE packed conditions than under LDPE packaged conditions. The higher thickness and sterile condition of HDPE provides a higher oxygen and moisture barrier that inhibits growth of aerobic bacteria, resulting in lower bacterial count in the HDPE packaged product than in the LDPE sample. Todd (2003) reported that HDPE packages protect against dust, insects, bacteria, and germs and also provide a barrier against moisture and oxygen. Relekar et al. (2014) observed no total Coliform organisms in solar tent dried improved ribbon fish at initial stage and during entire 120 days storage, while total Coliform organisms were detected in dried ribbon fish samples collected from a local market. In the current study, initially Coliform group TEC was detected in traditional dried fish samples, however, lower than permissible limit and too few to count (TFTC) TSC, TEC, and TVC were observed in chimney dryer produced improved sample and at HDPE storage condition.

| Biochemical analysis of dried fish sample
The biochemical quality of traditional and improved dried bombay duck (Harpadon nehereus) fish under different packaging conditions during storage time is presented in Table 6. Initially (at Day 0), the freshly produced improved dried fish had significantly higher protein content (59 ± 0.49%) than traditionally produced dried fish  Bhattacharya et al. (2016).
In the current study, lipid content of dried H. nehereus fish ranged from 5.25% to 6.00% produced under different processing methods, packaging condition, and storage time. The result supported the study by Siddique and Aktar (2011) who reported the lipid content in dried Harpodon nehereus varied from 7.73% to 7.78% during 6 month storage period, 6.08 to 8.62% in solar dried fish sample , and 5.94 to 11.77% in salted sun-dried products (Bhattacharya et al., 2016). Moisture content in improved dried fish had also increased significantly (p < .05) after 60 days of storage from Day 0 (15.18 ± 0.32%).
The higher moisture content in the traditional product indicates that fish was not dried properly, which leads to faster deterioration of the product. Moreover, in the traditional process, dried fish processors allow incomplete drying resulting in more moisture in dried fish to gain more weight for economic benefit. For both traditional and improved produced products, HDPE packaged products had lower moisture content than LDPE packaged products (p < .05), indicating that HDPE packaging provides a higher barrier for water absorbability than LDPE package, which was supported by Todd (2003)    and 19.31 ± 0.48% in un-irradiated bombay duck (Huque et al., 2013). Differences in moisture content exhibited in the current study might be due to dipping of fish into 5%-10% NaCl salt solution (in improved method) to avoid insect infestation in comparison to 10%-25% NaCl solution practiced in traditional method to get a higher weight benefit. The hygroscopic characteristics of NaCl salt bind with water in its structure (Bermeo et al., 2020), which results in a comparatively higher moisture content in the dried fish produced in that way. Irrespective of processing and packaging method, increasing moisture content was correlated with decreasing protein content during storage. Das et al. (2018) reported a similar finding that biochemical quality deteriorated for dried fish with increased relative humidity and moisture content of the products.

| CON CLUS ION
This paper delivers proof of concept of the UC Davis solar chimney dryer that provides an improved low-cost method for production of premium quality dried fish. Organoleptically, the UC Davis solar chimney dryer produced improved dried fish was rated as "excellent" by panelists and considered more highly acceptable than traditionally processed dried fish both at initial (Day 0) and after 60 days storage time. The dried fish in HDPE packaging had longer shelf life and retained better sensory, microbial, and biochemical quality compared to LDPE packaged samples. The APC, TSC, TEC, and TViC of UC Davis solar chimney dryer produced improved dried fish were within permissible limits, and no Salmonella spp. was found both at initial (Day 0) and after 60 days of storage, while the collected traditionally processed dried fish exceeded the ICMSF permissible limit.
Between two packaging conditions, the dried fish in HDPE packaged samples had better microbial quality compared to LDPE packaged samples. Nutritionally at initial time, the UC Davis solar chimney dryer produced improved dried fish had higher nutritional value compared to traditionally processed dried fish. After 60 days storage time, the nutritional quality remained good, confirming lower deterioration and longer shelf life in HDPE packaged compared to LDPE packaged conditions. Thus, quality dried fish can be produced using UC Davis solar chimney dryer in combination with improved packaging for extended shelf life that could deliver premium quality dried fish with desirable nutritional properties at acceptable domestic and international market standards.

ACK N OWLED G M ENT
The authors gratefully acknowledge technical and logistic supports from the USAID ECOFISH-BD project and Aquaculture-Horticulture for Nutrition Collaborative Program, WorldFish Bangladesh and South-Asia office, and the special research allocation (2017-18 and 2018-19) of the Ministry of Science and Technology, Bangladesh.

CO N FLI C T O F I NTE R E S T
The authors declare there is no conflict of interest. conceptualization (supporting); methodology (supporting); supervision (equal); validation (equal); writing Review Editing (equal).

E TH I C A L R E V I E W
The study involved the human for sensory quality analysis of dried fish which was conducted following the ethics approval from the

Research and Training Center of Patuakhali Science and Technology
University, Bangladesh.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.