A mathematical description of evaporative cooling potential for perishables storage in India
Sangeeta Chopra a,*, Norbert Müller b, Devinder Dhingra c, Indra Mani a, Tushar Kaushik a, Ankit Kumar a, Randolph Beaudry d
a Division of Agricultural Engineering, Indian Council of Agricultural Research-Indian Agricultural Research Institute (ICAR-IARI), Pusa Campus, New Delhi, 110012, Delhi, India b Department of Mechanical Engineering, Michigan State University, East Lansing, MI, 48824, USA c ICAR, New Delhi, 110012, Delhi, India d Department of Horticulture, Michigan State University, East Lansing, MI, 48824, USA
Post-harvest losses of fruit and vegetables in India, estimated to be around 30 %, can cause major economic disaster for smallholder farmers. A significant part of these losses occur due to lack of appropriate cold storage facilities, high temperatures, and low RH of ambient air especially during summer months. The high capital required and lack of uninterrupted power supply makes it difficult for farmers to build cold stores and existing cold storages are not distributed equitably and do not have sufficient capacity to serve India’s 100 million smallholder farmers. One cooling option is the relatively inexpensive evaporatively cooled (EC) storage, which was designed to enable farmers to avert distress sale and get a fair price for produce. An EC storage, termed the Pusa EC room, was built using novel construction materials including fabric walls and insulative blocks and evaluated year-round over a period of 5 years (2017–2021) using respiratory and senescence responses of amaranth (Amaranthus spp.) to storage temperatures. Wetting of the fabric walls yielded cooling of the structure interior during the daytime, but not at nighttime. As a result of lowered temperatures, storage life was predicted to be nearly doubled relative to storage at ambient temperatures during the warm and dry spring and summer weeks, however it was not improved appreciably when the ambient temperature declined and RH increased during late summer, fall, and winter. The estimated daily reduction in respired CO2 for the leafy amaranth, used here as a model plant, was governed by a simple mathematical expression using wet bulb depression of tem-perature relative to ambient. The predictive equation can be applied to any plant material for which the metabolic response to temperature is defined and permits convenient estimation of the benefits of evaporative cooling, potentially anywhere on the globe. This relationship was used for predicting storage-life improvement for many cities of major climate zones in India using data retrieved from website https://en.climate-data.org. EC room benefits were projected to be highest for warmer, drier climates, as would be expected; however, regional climate classifications were not always found to be a good guide for siting EC rooms due to local and micro-climate variability.
1. Introduction
Postharvest losses of vegetables and fruit in India are estimated to be about 30 % (Chadha, 2001) and, when coupled with poor or variable prices, can spell major economic disaster for farmers. In part, post-harvest losses of perishables occur due to high ambient temperatures and low relative humidity and a lack of appropriate storage facilities. The need for cold storage is especially keen for smallholder farmers, who
are unable to afford high-cost cold storage investments and are forced to sell their crop immediately after harvest, often leading to market glut and depressed prices (Banik, 2017; Department of Agriculture, Coop-eration and Farmers’ Welfare (DAC&FW), 2017). Many farming families have found that low prices make it impossible for them to sustain their activities or their families. As a result, farmers are increasingly leaving farming as an occupation, resulting in social upheaval and placing the stability of the Indian food supply at risk (Dev, 2012; Goyal et al., 2016).
* Corresponding author. E-mail address: [email protected] (S. Chopra).
https://doi.org/10.1016/j.postharvbio.2021.111727 Received 1 July 2021; Received in revised form 9 September 2021; Accepted 10 September 2021
Postharvest Biology and Technology 183 (2022) 111727
2
In the search for low-cost cool storage structures, the use of evaporation-based cooling has been evaluated (Ambuko et al., 2017; Chinenye et al., 2013; Chopra et al., 2001; Chouksey, 1985; Habibunnisa et al., 1988; Hall, 1972; James and Williams, 1988; Longmone, 2003; Roy, 1989; Roy and Khurdiya, 1986; Umbarkar et al., 1998). These structures have been suggested as a better option compared to refrigerated storage structures, owing to less initial capital investment, lower (or no) energy demands, and minimal operational cost, making them accessible to lower-income farmers/entrepreneurs (Vala et al., 2014). In most config-urations, a room is built with a wetted medium (e.g., sand, brick batt, charcoal) sandwiched between outer and inner walls or metal mesh. A common design in India included the use of two closely spaced brick walls with a layer of wetted sand between them in what is termed a brick-sand-brick (BSB) design (Vala et al., 2014). As the water migrates through the outer brick wall and evaporates, heat is drawn from the structure, cooling it and any produce that is stored within.
Evaporatively cooled (EC) structures can employ pumps or fans, using a small amount of electrical energy, but they can also be run without power, relying upon the operator to haul and supply the needed water. The use of fans to promote evaporation and cooling is called "active" EC storage. Active EC rooms achieve somewhat lower temperatures than passive EC rooms while maintaining nearly 80–90 % humidity inside the structure (Basediya et al., 2013). Variations on both of these designs have been tested and proved to be useful in hot and dry regions in India (Basediya et al., 2013; Jha and Chopra, 2006; Vala et al., 2014). EC storage is considered as an environmentally friendly alternative to refrigeration due to its relatively low energy footprint (FAO, 2011).
While EC cooling can be helpful in reducing temperature, in many regions the ambient temperature or humidity is too high to permit EC storage rooms to reach temperatures that will significantly extend storability. This could be seen in the study conducted by Chopra et al. (2004), where average temperature in a 2 t CIPHET EC room with brick sand brick (BSB) wall design was only 2 ◦C lower than an ordinary room in January when humidity was elevated. Even so, in this study the RH in the EC room was demonstrated to be significantly higher than in an ordinary room containing produce, varying between 88–95 %, and extending storability primarily by reducing moisture loss.
EC room design influences the rate and extent of cooling. Chopra and Beaudry (2018) designed an innovative EC storage of 100 kg capacity with a wetted fabric wall and compared its performance with a structure of similar capacity using a traditional BSB wall. The study revealed that a fabric wall EC structure cools more rapidly and to a greater extent than one using BSB walls and more closely follows cooler night air temper-atures. The improved performance of the fabric wall was attributed to its lower (almost 1/60th) thermal mass and higher thermal transmittance compared to the BSB design, which allows it to more efficiently shed
heat. Further, the fabric wall can be constructed more quickly and is 40–60 % less costly than a BSB wall.
Evaluating the performance of an EC room is not as simple as controlled refrigeration rooms because storage structures like EC rooms undergo short- and long-term fluctuations in both temperature and hu-midity (Mahangade et al., 2020). However, these “imperfect” storages can be evaluated and compared using the senescence (e.g., rate of abscission, chlorophyll loss, and leaf yellowing) of a model plant such as amaranth. Mahangade et al. (2020) found that, in amaranth, measures of senescence are linearly related to cumulative respiration. They concluded that the fabric wall EC room improved storability by approximately 13.3 % relative to ambient conditions during a spring-summer evaluation period in the year. Cevallos and Reid (2000) similarly found that the vase life of flowers was linearly related to respiratory activity.
To date, most work evaluating the performance of EC on perishable storage has focused on periods of relatively low humidity and daytime measurements. To more fully appreciate the utility of EC for perishables storage, the impact of day and night, seasonal variation and regional variation would be helpful. We were unable to find published work that provided a critical evaluation of EC storage as it relates to these vari-ables. To our knowledge, there are no year-round evaluations of EC rooms or assessments of the suitability of EC storage across climatic regions. The aim of the study was to critically evaluate the performance of the Pusa EC room for year-round utilisation in New Delhi, and to establish parameters to predict the efficacy of a 2-tonne EC room in various climatic regions of India. We express efficacy as the impact of EC on reduction of respiration and improvement in storage life of stored produce using amaranth as a model system. The mathematical expres-sions developed can be used to estimate the utility of EC storage on any perishable if the metabolic sensitivity to temperature is known.
2. Materials and methods
2.1. EC structure at IARI
A fabric wall EC room (Pusa EC room, Fig. 1) with the dimensions of 3 m × 3 m × 3 m and capacity of approximately 2000 kg of fruit and vegetables was built at the Indian Agricultural Research Institute (IARI), New Delhi (28◦37’52.5"N, 77◦09’05.2"E). The room was oriented so that one wall faced due south. The foundation of the EC room was built from conventional bricks. The perimeter foundation base was 0.75 m below ground level and 0.7 m wide at the bottom to 0.25 m wide at ground level. At each corner, a reinforced concrete pillar was erected. Each pillar was made by stacking autoclaved aerated concrete (AAC, Aerocon, HIL LTD, Hyderabad, India) blocks (5 cm × 24 cm × 64 cm) to make a hollow column with a 14 cm × 24 cm cavity. The cavity held four iron reinforcing bars and was filled with concrete to create a column to
Fig. 1. Pusa EC room (3 m × 3 m × 3 m) built at Indian Agricultural Research Institute (IARI), New Delhi (28◦37’52.5"N 77◦09’05.2"E). The left picture is the complete structure with nylon (aramid) felt walls and right picture is the internal structure showing the painted steel mesh, which provides some support for the fabric walls and the autoclaved aerated concrete covered pillars, ceiling, and floor.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
3
support the roof (Supplementary Fig. 1). AAC blocks were chosen to minimize heat ingress by virtue of their low thermal conductivity (Chopra and Beaudry, 2018). The roof was a 3.7 m × 3.7 m reinforced concrete slab. The floor was built on a layer of 5-cm thick brick batt, pounded and levelled while adding slight moisture, topped with 3.8-cm thick gravel:sand:cement mixture of ratio 12:8:1 and covered with a thin layer of “gholua” (smooth flowing cement water mixture). AAC blocks were also cemented to all exposed concrete portions of the roof using thinset mortar. A door was installed on the north side of the structure supported by a frame of reinforced concrete beams. A mesh (5 cm × 5 cm) of 2-mm thick welded iron wire was wrapped around the walls and secured to the ceiling and floor and was used to support the fabric walls. The outer walls of the Pusa EC room were composed of a 3-mm thick nylon felt (aramid needle felt, 540 g m− 2, Tianjin Tsingke Environ-mental Protection Technology Co., Ltd, Tianjen, Taiwan). Felt selection was based on the ability of the fabric to disperse water horizontally by capillary action (i.e., wicking). Fifty-two 30 cm × 30 cm fabric samples were evaluated for this property (Supplementary Table S1). To test wicking capacity, fabrics were suspended vertically and deionized water dripped on one corner and the horizontal movement of water deter-mined after one to two hours. To meet our needs of a fully wetting a fabric wall, a horizontal movement of 15 cm was needed at a point 20 cm from the top of the fabric swatch. Of the fabric samples evaluated, only aramid nylon felt met this criterion. Despite extensive searches, no source of aramid felt was found in India, although that would be pref-erable for supplying smallholder farmers in country.
To wet the fabric wall, a pocket was sewn into the top of the fabric into which a 6-mm diameter polyethylene garden dripline (XFD0612100, XF dripline, Rain Bird, Azusa, CA, USA) with emitters (2 L h− 1) at 305 mm intervals was inserted. Filtered water was supplied to the dripline from a 2000-L water tank elevated so that its base was 1 m above the dripline. Flow was regulated using a nylon needle valve. The roof of the structure was shaded by asbestos roof panels angled 23 degrees from the horizontal with the flat plane of the panels oriented southwards. Welded iron trusses supported the shade panels. The cost of the fabric was around 10–15 % of the total for Pusa EC room, which cost approximately US $4000 to construct, including labor. Approximately 1000 kg of water was placed in the Pusa EC room to simulate a 50 % product load in the room. The water was contained in approximately 500 1-L and 250 2-L poly-ethylene terephthalate (PET) bottles. The water-filled PET bottles were placed into plastic lugs and stacked in the Pusa EC room. Use of water bottles provided a consistent thermal mass throughout the study without the wastage and cost of using perishable products.
2.2. Instrumentation
Temperature data for the Pusa EC room and the immediate external environment were collected using T-type (copper-constantan) thermo-couples. Temperatures monitored include the wet and dry bulb temper-ature of the air inside the Pusa EC room (Twet.EC and Tdry.EC, respectively), the wet and dry bulb temperatures of the ambient air (Twet.amb and Tdry.
amb, respectively), and the temperature of water bottles used to simulate fresh produce. One thermocouple was used to log the air temperature of the room and was hung in the center of the Pusa EC room. The water bottle temperature was recorded by placing thermocouples in three different water bottles placed at middle and two corners of the Pusa EC room. The thermocouple cable was secured in the bottle by screwing the cap over the wire and positioning the bottle upright to prevent leakage. Wet bulb temperature was determined using a thermocouple placed in a 0.5-cm-wide, 10-cm-long water-soaked cotton wick (Baker B6031, Baker Instruments, Wilmington NC, USA) wetted from a water reservoir made from a 15-mL conical PET tube with a hole in the cap and positioned approximately 200 cm away from the chamber walls. Data were logged every 15 min using a datalogger (CR10X; Campbell Scientific, Logan, UT, USA) and relay multiplexer (AM16/32B, Campbell Scientific). Temper-ature data was downloaded from the datalogger onto a computer with
PC400W software (Campbell Scientific). The instrumentation setup was powered by a 12-V battery, which was charged using a 60-W solar panel and a charging regulator (PS12, Campbell Scientific). Temperature data were collected at various intervals over 5 years from February 2017 through June 2021.
2.3. Daily, weekly, and annual temperature calculations and mathematical model for respiration of whole leafy amaranth stems - a model plant
The average daily (dry bulb) temperature differential between the air temperature of the Pusa EC room and ambient (Tamb-TEC) was calculated. To evaluate the impact of daytime and nighttime on the degree of evaporative cooling, the 15-min interval temperature data was further split into daytime (d) and nighttime (n) temperatures, where daytime was considered from 8:00 am to 6:00 pm and nighttime was from 6:15 pm to 7:45 am. From these data, the average daily daytime and night-time dry bulb temperatures were determined for the Pusa EC room (Td
EC
and TnEC, respectively) and for the ambient air (Td
amb and Tnamb, respec-
tively). Representative data are provided in detail over a 13-day period from 28 March to 9 April 2017. In this example, the walls were wetted for 7 consecutive days and water was then withheld for 6 days to capture data typical of wet and dry periods instituted periodically over the course of the study. The difference between outside and inside air for daytime and nighttime periods and between outside air and produce for the whole day were calculated from the daily average temperatures. An independent T test for the effect of wet and dry fabric on EC room temperature was performed for the period specified.
Relative humidity (RH, %) was calculated using the empirical rela-tionship between wet and dry bulb temperatures (Vaisala, 2013). The wet bulb depression was calculated for the ambient environment as the difference between the ambient dry bulb and wet bulb temperatures. The days when the fabric was wetted (periods of evaporative cooling) and when the fabric was dry (no evaporative cooling) were recorded at various intervals over the experimental years from 2017 to 2021.
The respiration rate (rCO2, g kg− 1 h− 1) for leafy shoots of the model plant amaranth (Amaranthus spp.) was calculated from temperature data using the Arrhenius equation [Eq. (1)] as developed by Mahangade et al. (2020). The apparent energy of activation (Ea) and Arrhenius factor (A) for calculating rCO2 were 55.8 kJ mol–1 and 2.16 E9 g kg− 1 h− 1, respectively (Mahangade et al., 2020) such that:
rCO2 = 2.16 × 109 × e (− 55.8/RT) (1)
where R = universal gas constant (0.0083144 kJ mol− 1 K− 1) and T =temperature (K).
Cumulative respired CO2 (cr, g kg− 1) for ambient conditions (cramb) and for the conditions within the Pusa EC room (crEC) was calculated by integrating rCO2 over 15-minute intervals over the course of a 24-h (daily) basis and for daytime and nighttime periods for each day. The percent reduction in cumulative respired CO2 relative to ambient due to evaporative cooling in Pusa EC room was calculated by Eq. (2).
Percent reduction in cumulative respired CO2 (%) = (cramb − crEC) × 100/ cramb (2)
Mahangade et al. (2020) demonstrated that cramb is linearly related to storage life of leafy amaranth stems. Thus the reduction in respired CO2 could be used to calculate the extent to which evaporative cooling could extend storage life in a relative sense. The fold change in storage life and relative change in storage life was calculated using Eqs. (3) and (4), respectively, for whole days and for daytime and nightime periods for wetted and dry fabric, separately.
Fold change in storage life = cramb / crEC (3)
Relative change in storage life = (cramb − crEC) / crEC (4)
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
4
2.4. Relationship between reduction in respired CO2 (%) and wet bulb depression
The relationship between reduction in respired CO2 and wet bulb depression was determined over the experimental period of 2017–2021 for wetted fabric on a daily basis. The data were empirically fitted using commercial software (TableCurve 2D, v. 2.0, Jandel Scientific, San Rafael, CA). The equation incorporates the difference in wet and dry bulb tem-peratures to capture the simultaneous impacts of humidity and temper-ature. Using the exponent of this difference permits the value generated to increase asymptotically. The equation was selected because it is the simplest expression that crosses the origin when the difference between the two variables (dry bulb and wet bulb) is zero, and asymptotes to a
maximum of 100 as the difference between the two variables (wet bulb depression) increases to its maximum.
Percent reduction in respired CO2 = 100 × (1− EXP (− K × (Tdry.amb − Twet.
amb)) (5)
where K is a variable that reflects the temperature sensitivity of the res-piratory response of the commodity being stored, Tdry.amb is the dry bulb temperature (◦C), and Twet.amb is wet bulb temperature (adiabatic satu-ration temperature, ◦C). From this expression, the fold change in storage life, the relative change in storage life, and the relative improvement in storage life could be calculated.
Fig. 2. Temperature (A) and RH profiles (B) in the Pusa EC room (3 m × 3 m × 3 m) and in the ambient environment taken at 15-min intervals over a representative 13-day period from 28 March to 9 April 2017 in Delhi, India. Simu-lated produce refers to 1000 kg of polyethylene terephthalate bottles filled with water. Shaded area indicates period when the outer nylon felt fabric walls were wetted. Non-shaded area in-dicates period when water was withheld from the nylon felt. Diagonal pattern indicate a rain event beginning on the evening of 4 April through the morning of 5 April, 2017.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
5
2.5. Weekly temperatures and RH from IARI weather station
We retrieved the daily wet and dry bulb data of the ambient envi-ronment from the IARI meteorological database system from Agromet observatory (28◦38’23"N, 77◦09’27"E, altitude: 228.61 m), Division of Agricultural Physics, IARI, New Delhi, (https://www.iari.res.in/index. php?option=com_content&view=article&id=450&Itemid=443) from 2017 to 2021. The weather station site was only 750 m from the IARI Pusa EC room site. We have used their ambient wet bulb data for calculating wet bulb depression, as our ambient wet bulb data at Pusa EC room site were incomplete.
2.6. Fold change in storability and relative change in storage life
The performance of the Pusa EC room was evaluated by estimating the fold change and relative change in storage life of amaranth, calcu-lated as previously described. The daily fold change in storability was calculated from 15-min interval data acquired on-site throughout the 5- years of the study and, from this, was used to determine the average daily fold change in storability averaged over each week and each month of the year. The fold change in storability was also calculated using IARI weather data for wet and dry bulb from the Agromet observatory. This was done to determine whether weather station data could reproduce Pusa EC room performance parameters on weekly and monthly bases. The ’predicted’ (weather-station derived) values of the daily fold change in storage life for amaranth were regressed against the performance data from the research site. Similarly, the relative improvement in storage life was also determined using publicly available daily average climate data for each month for New Delhi from https://en.climate-data.org/asia/ind ia/ (accessed 26 May 2021), based on European Centre for Medium- Range Weather Forecasts (ECMWF) database. The on-line global climate model is based on 1.8 billion data points and a resolution of 0.1–0.25 grade and uses weather data collected between 1999 and 2019.
2.7. Climate-data of major climatic zones of India
The daily temperature (Tdry) and RH data, averaged for each month, for major climatic zones of India, was retrieved from https://en. climate-data.org/asia/india/ (accessed 26 May 2021) as noted previ-ously. The daily average temperature and RH for each month were converted into wet bulb temperature (Twet) using the following empirical relation for atmospheric pressure (Stull, 2011).
The efficacy of Pusa EC room was compared on the basis of total benefit for the whole year [i.e., cumulative yearly storage-life improvement (days per year)] and the proportion of this predicted improvement attributable to the three months of a year having the highest amount of evaporative cooling. The cumulative yearly storage- life improvement was calculated using Eq. (7).
Cumulative yearly storage life improvement
= (∑Dec
JanDaily relative change in storage life) ×
(36512
)
(7)
3. Results and discussion
3.1. Diurnal patterns in temperature and relative humidity
The 15-min interval data (i.e., 96 data points per day) from the selected representative period of March-April 2017 revealed clear daily patterns for temperature and RH (Fig. 2). Further, the patterns for temperature and RH in the Pusa EC room differed when the fabric wall
was wet and dry. In this representative example, the exterior (ambient) temperature peaked at midday, declining thereafter to a minimum at around 6:00 a.m. the next morning. When the fabric wall was wetted, the interior temperature of the Pusa EC room peaked before the ambient temperature reached its maximum, declining as the exterior tempera-tures were still rising; when the fabric wall was dry, the interior tem-perature peaked simultaneously with the ambient temperature. The minimum interior temperatures for wet and dry walls were reached simultaneously with, or slightly later than, the minimum ambient tem-perature. Following a rain event on 4 April, and coincident with the dry fabric treatment period, the ambient temperatures declined slightly relative to the first 7 d; however, the interior temperature maxima and minima were still higher than during the warmer period when the fabric was wet.
In the representative example, the ambient RH peaked at nighttime (60–85 %, exclusive of the rain event) and declined to a minimum (30–35 %, exclusive of the rain event) at midday or slightly thereafter. In the rain event, the RH increased throughout day and night, returning to its earlier pattern approximately one day later. When the fabric was wet, the RH of the interior remained high (88–95.8 %), throughout day and night, with no discernible pattern, even though the interior temperature fluctuated between 22 and 28 ◦C. However, when the fabric was dry, the RH was lower (53–60 %) and fluctuated with the ambient RH. This humidity range in for the Pusa EC room with wetted walls is comparable to that obtained by Chopra et al. (2004) in a BSB-type EC room. Elevated humidity in the interior of the Pusa EC storage was expected; a high relative humidity in EC storage is a well-established and attractive feature of the technology (Basediya et al., 2013; Liberty et al., 2013).
In the representative example, evaporative cooling reduced interior temperatures relative to ambient during the day, but not during the nighttime. Under conditions of evaporative cooling, when ambient temperatures were at their warmest during the daytime, the maximum temperature in the Pusa EC room was 14.2 ◦C cooler than the ambient temperature (Table 1). However, without evaporative cooling, the room was only 4.7 ◦C cooler than the ambient (Table 1).
The amount of cooling during the night was limited; the temperature within the structure was higher than the exterior temperature at the nighttime temperature minima. However, evaporative cooling yielded temperatures that were nearer to ambient at this point averaging 1.4 ◦C warmer, compared to 4.3 ◦C when there was no evaporative cooling (Table 1). The lack of nighttime cooling is due to the increase in the RH during this part of the day. A region with lower nighttime humidity would likely yield better performance from evaporative cooling.
The temperatures in the Pusa EC room for the period evaluated here were higher than those obtained by Ambuko et al. (2017) for the zero-energy brick cooler and the evaporative charcoal cooler. This was likely because the ambient temperatures experienced in India (20–42 ◦C) were higher than those in the Kenya experiment (15.3–29 ◦C). The
Table 1 Effect of fabric wall wetting on the difference between ambient (amb) and Pusa EC room air temperatures during daytime (d) and nighttime (n), and on the temperature of simulated produce at 15-min intervals over a representative 13- day period from 28 March to 9 April 2017 in Delhi, India by independent T tests. Produce was simulated using 1- and 2-L bottles of water.
Temp. difference
Fabric Mean ± SD t df Sig. (2- tailed)
Tdamb - T
dEC* Wet 14.12 ± 1.55 a
7.802 11 0.000 Dry 4.71 ± 2.73 b
Tnamb - T
nEC* Wet − 1.44 ± 0.85 a − 3.622 11 0.004
Dry − 4.34 ± 1.92 b T24h
amb - T24hproduce** Wet 5.10 ± 0.526 a 5.062 5.685 0.003
Dry 1.12 ± 1.86 b
*Levine’s test assuming equal variances, **Levine’s test assuming unequal variances.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
6
temperature reduction due to evaporation, while difficult to compare because of diurnal fluctuations, appeared to be greater for the Pusa EC room, however. Ambuko et al. (2017) recorded an average temperature reduction of 7–8 ◦C at the daytime maxima compared to the 14.2 ◦C reported here. This difference is likely because humidity was higher during the daytime in Kenya. During the nighttime, the zero-energy brick cooler did not provide effective cooling, but the evaporative charcoal cooler did, cooling below the nighttime air temperature, unlike the Pusa EC room. This better performance of the evaporative charcoal cooler at this time of day could reflect the influence of a number of factors, but likely not humidity, since the humidity in the Kenya study rose to 85–95 % at night. Influx of additional heat from the roof or soil in the Indian
Pusa EC room may explain part of the difference. For instance, the average soil temperature in the Delhi area is approximately 22 ◦C (Bansal et al., 1983), whereas that in the Nairobi area is approximately 18.3 ◦C (McCulloch, 1959) for a dry shaded surface.
The changes in the temperature of simulated produce in the EC room lagged changes in the temperature of the room air, reflecting the greater thermal mass of the simulated produce relative to that of air (Fig. 2). Consistent with this, the temperature maxima for the simulated produce were lower and the minima were higher than room air. During the wetted wall period, temperature of the simulated produce was lower, on average, than when the walls were dry, as might be anticipated based on the shifts in air temperature (Table 1).
Fig. 3. Daily averages for estimated cumulative CO2 respired on a fresh weight basis for amaranth (A), temperature difference between ambient and room for daytime (Tamb-TEC) for daytime and nighttime periods (B), RH (C), and air temperature (D) for the Pusa EC room (3 m × 3 m × 3 m) and the ambient environment during periods when the nylon felt fabric walls were wetted (dark gray shading) or dry (light gray shading) in 2017 in Delhi, India. Non- shaded areas indicate environmental moni-toring only. Amount of CO2 respired for amaranth was calculated from previously pub-lished data (Mahangade et al., 2020).
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
7
3.2. Temperature and RH in the Pusa EC room as a function of wet or dry walls
The 15-min interval data collected during 2017 was used to calcu-late daily averages for each week to better represent seasonal patterns for temperature and RH in the Pusa EC room and in the ambient environment for a portion of the year. The ambient dry bulb tempera-ture increased steadily from the initial weeks and peaked around week 20–25 (mid May – mid June) and then steadily decreased until the end of the year (Fig. 3D). The average wet bulb temperature, which de-termines the maximum extent of evaporative cooling possible, had clear seasonal trends, reaching its maximum near weeks 27–33. The difference between ambient dry bulb and wet bulb (wet bulb depres-sion) exceeded 5 ◦C only between weeks 13–24 (mid-April – end June). The conditions inside Pusa EC room were monitored between 8th week and 33rd week in 2017. When the fabric was wet, the interior tem-perature was below the ambient temperature, but above the wet bulb. When the fabric was dry, the interior temperature rose to the ambient temperature, remaining well above wet bulb temperature. This larger sampling of data was consistent with the two weeks of data in Fig. 2, during the dry and wet periods.
The RH of the ambient air (Fig. 3C) declined after week 11 to a minimum of 45 % on the 17th week, increasing thereafter to 80–90 % by the 25th week (first week of July) coinciding with the onset of the monsoon season in 2017. During weeks 8–13 and 27–33, when the fabric was wet, the elevated ambient RH prevented evaporative cooling. As per the detailed data of Fig. 3, interior RH remained between 90–100 %, when the fabric was wet. When the fabric was dry, the interior RH was closer to ambient.
For wetted fabric, the temperature difference between ambient and EC Room was higher during the daytime than at nighttime, indicating that most of the cooling occurred during daytime and little to no cooling took place at night (Fig. 3B). The only period when there was any appreciable cooling at night was when the RH was low. When the fabric was dry, the daytime temperature difference between the interior and exterior was very low (0–3 ◦C), indicating almost no cooling. The tem-perature difference was negative during nighttime, which indicated that EC Room was warmer than the ambient.
Under evaporative cooling (i.e., wet fabric), the daily cumulative respired CO2 was reduced in Pusa EC room relative to ambient condi-tions (Fig. 3A). The reduction was greatest during the warm dry summer weeks 13–24 (mid-April – end June), reaching approximately 40 % relative to ambient. The expected reduction could be more or less for other crops depending on their metabolic sensitivity to temperature. The apparent energy of activation (Ea) for respiration rate used for this es-timate is based on that for amaranth as 55.8 kJ mol− 1 (Mahangade et al., 2020). Apparent Ea values are reported to range from 20 to 100 kJ mol− 1
for common fruit and vegetables in air, with the most frequent Ea values being between 50 and 80 kJ mol− 1 (Beaudry, 2007; Eriko et al., 2001; Tano et al., 2005). When the fabric was dry, the estimated cumulative respired CO2 was similar to that in the ambient.
We were unable to find references documenting the temperatures in an EC structure when the walls of such a structure were dry, making the relative benefit of evaporative cooling for a particular structure unclear from previous work. Here, the data demonstrate an advantage in terms of temperature control that is mainly encountered during the warmer daytime period.
3.3. Reduction in respired CO2 in the Pusa EC room as a function of wet bulb depression
The data from 5 years of data collection (2017–2021) revealed that the percent reduction in respired CO2 was related to wet bulb depression only when the fabric was wet, as is expected in response to evaporative cooling (Fig. 4). The relationship between reduction in respired CO2 (%) with wet bulb depression for wet fabric was curvilinear and described
empirically by Eq. (5) to yield a K value of 0.0481 such that: Percent reduction in respired CO2 = 100 × (1-EXP (-0.0481 × (Tdry.amb − Twet.amb)).
3.4. Fold change in storage life
The fold change in storage life is depicted for daytime and nighttime periods for each day for wetted fabric walls over the years 2017–2021 (Fig. 5). Over the five years of the study, the fold change in storage life in response to evaporative cooling was consistently greater in the daytime than at nighttime, consistent with the previous, more detailed analyses for 2017 (Fig. 3B) where most, if not all, of the evaporative cooling occurred during the day. After the first few weeks of the year, the day-time fold change in storage life increased (Fig. 5B and C) as the ambient temperature increased and RH declined (Fig. 5A). The average daytime change in storage life increased up to 2-fold during the warm dry spring and summer between weeks 13 and 24 (mid-April – end June), when the ambient temperature was at its maximum and RH near its minimum. Thereafter, as the ambient temperature declined and RH increased, the daytime fold change declined, reaching a value near 1, indicating no benefit in storability for produce in the Pusa EC room. The average nighttime fold change in storage life remained between 0.9 and 1.3 throughout the year, indicating no meaningful benefit in storability during the night.
The daily fold change in storage life, calculated from the measured 15-min temperature data, was averaged for each week over the years 2017–2021 for the Pusa EC room with wetted walls and is depicted in comparison to the predicted fold change derived from temperature and RH data from the Agromet observatory at IARI and Eq. (5) (Fig. 5B). The R2 value for the correlation of the two data sets is 0.76 (Fig. 5B, inset), suggesting that weather station data can be substituted for on-site data and used to predict the performance of the EC Room with a reasonable degree of accuracy.
The modeled average daily relative improvement in storage life was determined for each month of the year from fold change values for both measured (based on estimates of respiration using actual temperature measurements) and predicted [based on estimates of respiratory reduction from the local Agromet observatory wet and dry bulb tem-peratures using Eq. (5)] data sets (Fig. 6B). The R2 value for correlation of the two data sets is 0.90 (Fig. 6B, inset), which is an improvement in
Fig. 4. Reduction in the percent respired CO2 as a function of wet bulb depression for storing the model plant, amaranth, in the Pusa EC room (3 m × 3 m × 3 m) with wetted (closed symbols) and non-wetted (open symbols) nylon felt fabric walls from 2017 to 2021. Each symbol represents percent reduction for an individual day integrated (summed) over the whole day using mea-surements made at 15-minute intervals from 12:00 a.m. to 11:45 p.m. The solid line is the fitted curve for predicting the percent reduction in respired CO2, based on daily average differential between the wet bulb and dry bulb tem-peratures. The equation for fitted line is: Percent reduction in respired CO2 = 100 × (1-EXP (− 0.0481 × (Tdry amb -Twet.amb), with an r2 value of 0.55 and where Tdry.amb is the dry bulb temperature and Twet.amb is wet bulb temperature (adiabatic saturation temperature) of ambient air.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
8
fit relative to weekly data. This suggests that average monthly temper-ature and RH data taken from a weather station adequately describes the year-round performance of structures like the Pusa EC room.
Interestingly, the improvement for storage life differed between modeled climate data retrieved from https://en.climate-data.org/asia /india (Climate-Data.org, 2021; accessed on 1 May, 2021) and the local Agromet observatory at IARI (ICAR-IARI, 2021; Fig. 6B). The difference was because the RH for New Delhi retrieved from the modeled website source was lower than that from the Agromet observatory at IARI in the early summer months (Fig. 6A). The difference in RH between the two data sources may be due to the effect of surrounding vegetation on the IARI campus on the ambient environment. Even so, the pattern of relative improvement in storage life for both weather station and modeled data were similar in that the improvement in storage life was depicted as increasing from January through May, thereafter, declining till August and then rising slightly between September through December.
3.5. Performance of EC Room in cities in major climatic zones in India
The relative change in storage life was low (median between 0.11–0.22) for cities classified by Koppen and Geiger (Geiger, 1954) as humid subtropical climate (Cfa/Cwa) and tropical monsoon climate (Am) (Fig. 7A). Among the sites evaluated, the relative change in storage life was highest (median = 0.52) for Jodhpur, India, which is in a hot desert climate (BWh). More moderate improvements in storage life were found for cities in hot semi-arid climates (BSh) and Bhopal, a city belonging to tropical savanna classification (Aw) (median 0.3–0.44). The city of Bhopal is situated in the middle of India and is not near water body and has high temperatures and low humidity during summers (Fig. S2). The other four cities classified as tropical savanna had low (median between 0.11− 0.23) relative change in storage life, as these cities are along the coastline of India and are moderately hot with a relatively high humidity.
Fig. 5. Daily mean ambient temperature and RH (A), measured and predicted fold change in storability of the modeled plant, amaranth, in the Pusa EC room (3 m × 3 m × 3 m) due to evaporative cooling (B), and measured fold change in storability of amaranth during day-time and nighttime periods in the Pusa EC room due to evaporative cooling (C) over 5 years from 2017 to 2021. Week 1 is the first week of January. In the upper panel (A), the solid black line depicts daily mean temperature and dotted black line is the daily mean RH (%) of ambient air over 5 years from 2017 to 2021 from Agromet observatory IARI, New Delhi. For change in storage life calculations (B and C), only periods of wetted nylon felt walls using measurements were considered. Measured fold change in storability (B, closed symbols) are from respiratory rate calculations (Mahangade et al., 2020) based on measured temperature data for the Pusa EC structure and the ambient environment; predicted fold change in stor-ability (B, open symbols) are derived from the expression Percent reduction in respired CO2 =
100 × (1-EXP (− 0.0481 × (Tdry.amb- Twet.amb)) based on daily averages for the dry bulb air temperature (Tdry.amb) and the wet bulb tem-perature (Twet.amb) obtained from the IARI Agromet weather station (ICAR-IARI, 2021). The inset graph in the middle panel (B) re-gresses measured data against predicted data. In the lower panel (C), each point represents a portion of a day for the week specified, where open symbols indicate daytime and closed symbols represent nighttime. Daytime was from 8:00 a.m. to 6:00 p.m. and nighttime is from 6:15 p.m. to 7:45 a.m. Solid and dashed lines depict averages for each week for daytime and nighttime, respectively.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
9
The cumulative yearly storage life improvement for cities in humid tropical and tropical monsoon climates and in tropical savanna climatic regions along the Indian coastline was between 40–85 days over the course of the year (Fig. 7B). The cumulative storage life improvement was highest (around 200 d) for cities in hot desert climates, followed by 134–180 days for cities in hot semi-arid zone. The proportion of the predicted storage-life improvement for the three most favorable months for evaporative cooling relative to the whole year ranged between 0.3 and 0.5, suggesting that the benefit of evaporative cooling was not evenly spread across the year for all sites. For Bhopal, New Delhi and Lucknow, for instance, nearly half the benefit for evaporative cooling throughout the year would be predicted to occur in only three of the 12 months.
Over the course of the study, it was noted that the nylon felt discolored over time due to the build up of algae. The fabric could be cleaned using a copper sulfate based algicide according to manufacturer’s recommenda-tions. The longevity of the nylon felt is unknown, but as of publication of this article, it has been installed and wetted for over 5 years and despite the discoloration, no deterioration is visible. The EC functionality appears to be unchanged in that there were no evident differences in performance over the 5 years of the study as determined by plots of fold change in storage life (Fig. 5). Fungal growth, while not evaluated, was not evident; fungi do not degrade nylon to our knowledge. The findings of this study are not intended to suggest the use of a nylon felt fabric should be used to effect evaporative cooling, however, given that it is considerably more effective than other wall designs in cooling by evaporation (Chopra and Beaudry, 2018), the data provide a ’best case’ evaluation of the potential for evaporative cooling as a function of climatological data.
4. Conclusions
To our knowledge, this is the most exhaustive study of the potential for evaporative cooling to extend product storability, expanding previ-ous observations to include year-round assessments for multiple years and differentiating between daytime and nighttime performance. Further, this is the first study to demonstrate that wetting of the evap-orative surface is essential for the cooling effect, an outcome often assumed, but never tested to our knowledge. Further, the comparison of temperature and humidity in the same structure with dry walls as a control treatment has not been evaluated. Instead, comparisons have routinely been made only to ambient conditions.
Our data for the Pusa EC room, corroborates previous studies and demonstrates that evaporative cooling can provide a high RH and lower temperature relative to ambient when conditions for evaporation are favorable. The data from this study for temperature conditions in an EC room throughout the year clearly demonstrates the shortfall of this technology during periods of high ambient relative humidity, a fact that has also been assumed by earlier researchers (Basediya et al., 2013), but not quantified. In Delhi, elevated humidity during the nighttime essen-tially limits all of the benefit to evaporative cooling occurs during the daytime. This feature has been shown previously, but not much dis-cussed. Nighttime increases in RH markedly limit the benefits of evapo-rative cooling.
This is the first attempt to develop a predictive equation for the impact of EC storage on respiratory activity and storage life of amaranth as a function of RH and temperature. The mathematical model created
Fig. 6. Daily average temperatures (closed symbols) and RH (open symbols) for each month from the Agromet observatory (IARI, 2021) and from the Climate.org website (Climate.org, 2021) for New Delhi (A) and relative improvement in storage life of amaranth in the Pusa EC room (3 m × 3 m × 3 m) (B). The relative improvement in storage life is derived from three sets of data: measured temperature data in the Pusa EC room and its immediate ambient environment (closed cir-cles), predicted based on temperature and hu-midity data from the Agromet observatory (open circles), and predicted based on temper-ature and humidity data from the Climate.org website (closed triangles) for New Delhi (A). The predicted fold change in storability (B, open symbols) are derived from the expression Percent reduction in respired CO2 = 100 × (1-EXP (− 0.0481 × (Tdry.amb- Twet.amb)), based on daily averages for the dry bulb air temperature (Tdry.amb) and the wet bulb temperature (Twet.
amb). The inset graph in the lower panel (B) regresses measured data against predicted data from the Agromet observatory.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
10
here can be expanded to all plants whose metabolic responses to tem-perature are described. Given that the apparent energy of activation of amaranth metabolism is similar to many other perishables, the assess-ments of benefits will likely not differ markedly. The weekly and monthly average daily dry bulb temperature and RH data from weather stations adjacent to the research site provided an adequate prediction of EC room performance throughout the year. The fitted relationship from Eq. (5) permits convenient estimation of the benefits of evaporative cooling, potentially anywhere on the globe, as long as data for average RH and temperature are available. In India, the potential benefit from EC storage is predicted to range widely and to be highly site-specific, with a significant portion (30–50 %) of the benefit obtained during a relatively short 3-month window during the year when temperatures are high and humidity is low. While climate-modeled responses can provide guidelines for what one might expect in terms of benefits he climatic data, actual performance may differ from what is anticipated because of microclimate effects on evaporation as evidenced by differences in the predictions from the Agromet observatory and the Climate.org sources.
Further, while the climate Koppen and Geiger climate classification system (Geiger, 1954) can serve as a general guide for expected EC benefits, temperature and RH patterns differ within single climate classes sufficiently that the classification, by itself, cannot be used to accurately predict success of EC room technology.
Authorship contributions
Sangeeta Chopra: Conceptualization, Formal analysis, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Validation, Supervision, Software, Visualization, Writing - original draft, Writing - review & editing, Transfer of Technology.
Devinder Dhingra: Conceptualization, Resources, Funding acqui-sition, Project administration, Methodology, Transfer of Technology.
Indra Mani: Funding acquisition, Project administration, Resources, Transfer of Technology.
Fig. 7. Relative change (A) and the cumulative days improvement in storage life per year (B) of amaranth in the nylon felt-covered Pusa EC room (3 m × 3 m × 3 m) due to evaporative cooling. Data are for various cities representa-tive of major climatic zones of India as described by their assigned Koppen-Geiger code (Cfa/Cwa = humid and monsoon subtropical, BSh = hot semi-arid, Csa = hot-summer Medi-terranean, Aw = tropical savanna, BWh = hot desert, and Am = tropical monsoon). Data points in the upper panel (A) represent each month of the year, boxplots depict the range between the first and third quartiles, whiskers depict minimums and maximums excluding outliers. The climate data is taken from http s://en.climate-data.org/asia/india-129/. The fraction above each histogram bar in the lower panel (B) is the ratio for the top three months relative to the whole year.
S. Chopra et al.
Postharvest Biology and Technology 183 (2022) 111727
Randolph Beaudry: Conceptualization, Formal analysis, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Validation, Supervision, Software, Visuali-zation, Writing - original draft, Writing - review & editing, Transfer of Technology.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research was supported by the USAID Partnership for Enhanced Engagement in Research (PEER, Cycle 7) grant funding from The Na-tional Academies of Sciences, Engineering and Medicine (NAS), Wash-ington, DC 20001. The authors would also like to thank NAS for their financial support and IARI, India and MSU, USA for administrative support. Beaudry acknowledges support from Michigan AgBioResearch and the USDA National Institute of Food and Agriculture, Hatch project MICL002688.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.postharvbio.2021 .111727.
References
Ambuko, J., Wanjiru, F., Chemining’wa, G.N., Owino, W.O., Mwachoni, E., 2017. Preservation of postharvest quality of leafy amaranth (Amaranthus spp.) vegetables using evaporative cooling. J. Food Quality, 5303156. https://doi.org/10.1155/ 2017/5303156, pp. 6.
Banik, N., 2017. Farmer Distress. Rajasthan Priorities, Copenhagen Consensus Center. License: Creative Commons Attribution CC BY 4.0. https://www.copenhagenconse nsus.com/sites/default/files/raj_farmer_distress_sm.pdf (Accessed on 30 June 2021).
Bansal, N.K., Sodha, M.S., Bharadwaj, S.S., 1983. Performance of earth air tunnels. Energy Res. 7, 333–345. https://doi.org/10.1002/er.4440070405.
Basediya, A.L., Samuel, D.V.K., Beera, V., 2013. Evaporative cooling system for storage of fruits and vegetables - a review. J. Food Sci. Technol. 50 (3), 429–442. https://doi. org/10.1007/s13197-011-0311-6.
Beaudry, R.M., 2007. MAP as a basis for active packaging. In: Wilson, C.L. (Ed.), Frontiers of ‘Intelligent’ and ‘Active’ Packaging for Fruits and Vegetables. CRC Press, Boca Raton, FL, pp. 31–55.
Cevallos, J.C., Reid, M.S., 2000. Effects of temperature on the respiration and vase life of Narcissus flowers. Acta Hortic. 517, 335–342. https://doi.org/10.17660/ ActaHortic.2000.517.42.
Chadha, K.L., 2001. Handbook of Horticulture. Indian Council of Agricultural Research, New Delhi, p. 1031.
Chinenye, N.M., Manuwa, S.I., Olukunle, O.J., Oluwalana, I.B., 2013. Development of an active evaporative cooling system for short-term storage of fruits and vegetables in a tropical climate. Agric. Eng. Int.: CIGR J. 15 (4), 307–313. https://cigrjournal.org/ index.php/Ejounral/article/view/2655.
Chopra, S., Beaudry, R., 2018. Innovative composite wall designs for evaporative cooled structures for storage of perishable. Indian J. Agric. Sci. 88 (11), 1692–1695.
Chopra, S., Baboo, B., Kudos, S.K.A., Oberoi, H.S., 2001. Development of passive evaporative cooled room for on-farm storage of fruits and vegetables. In: Agarwal, R. S. (Ed.), Proc. Emerging Technologies in Airconditioning and Refrigeration. Allied Publishers Ltd., Mumbai, pp. 265–273.
Chopra, S., Kudos, S.K.A., Oberoi, H.S., Baboo, B., Ahmad, K.U.M., Kaur, J., 2004. Performance evaluation of evaporative cooled room for storage of Kinnow mandarin. J. Food Sci. Technol. 41 (5), 573–577.
Chouksey, R.G., 1985. Design of passive ventilated and evaporatively cooled storage structures for potato and other semi perishables. In: Proc. Silver Jubilee Convention of ISAE. Bhopal, India, October 29–31, pp. 45–51.
Climate-Data.org, 2021. India. In: Merkel, A. (Ed.), AM Online Projects, Oedheim, Germany. https://en.climate-data.org/asia/india. (Accessed on 30 June 2021).
Department of Agriculture, Cooperation and Farmers’ Welfare (DAC&FW), 2017. Doubling Farmers’ Income – Volume III, Post-Production Agri-Logistics: Maximising Gains for Farmers. Ministry of Agriculture & Farmers’ Welfare. August 2017, pp. 176. https://agricoop.gov.in/sites/default/files/DFI%20Volume%203.pdf. (Accessed on 30 June 2021).
Dev, S.M., 2012. Small Farmers in India: Challenges and Opportunities. Indira Gandhi Institute of Development Research, Mumbai, p. 34. June 2012. http://www.igidr.ac. in/pdf/publication/WP-2012-014.pdf. (Accessed on 30 June 2021).
Eriko, Y., Keiko, T., Daisuke, H., Wenzhong, H., Toshitaka, U., 2001. Effect of temperature on the respiration rate of some vegetables. Control applications in post- harvest and processing technology. In: 3rd IFAC/CIGR Wkshp. 3–5 Oct. Tokyo, Japan.
FAO, 2011. Energy-Smart Food for People and Climate - Issue Paper. Publishing Policy and Support Branch, Office of Knowledge Exchange, Research and Extension, FAO, Viale delle Terme di Caracalla, Rome, Italy, p. 66. http://www.climatechange.govt. nz/emissions-trading-scheme/participating/agriculture/allocation/. (Accessed on 30 June 2021).
Geiger, R., 1954. Klassifikation der klimate nach W. Koppen. Landolt-Bornstein – Zahlenwerte und Funktionen aus Physik, Chemie, Astronomie, Geophysik und Technik, alte Serie, Vol 3. Springer, Berlin, pp. 603–607.
Goyal, S.K., Prabha, Rai, J.P., Kumar, S., 2016. Indian agriculture and farmers-problems and reforms. Indian Agriculture and Farmers. Poddar Publication, Tara Nagar, Chittupur, BHU, Varanasi, pp. 79–87. https://www.researchgate.net/publication/ 330683906. (Accessed on 30 June 2021).
Habibunnisa, E.A., Aror, E., Narasimham, P., 1988. Extension of storage life of the fungicidal waxol dip treated apples and orange under evaporative cooling storage conditions. J. Food Sci. Technol. 25 (2), 75–77.
Hall, E.G., 1972. Precooling and container shipping of citrus fruits. CSIRO Food Res. Q. 32 (1), 1–10. http://hdl.handle.net/102.100.100/313885?index=1. (Accessed on 30 June 2021).
ICAR-IARI, 2021. Daily Weather Data, IARI Meteorological Database System. Division of Agricultural Physics, IARI, New Delhi. https://www.iari.res.in/index.php?optio n=com_content&view=article&id=450&Itemid=443. (Accessed on 30 June 2021).
James, W.L., Williams, R.W., 1988. New developments in solar powered services for remote aboriginal communities. In: Proc. of the Conference on Science and Technology for Remote Communities. 18–19 July, Murdoch University, Murdoch, W.A. Paper 3.1. https://researchrepository.murdoch.edu.au/id/eprint/22817/. (Accessed on 30 June 2021).
Jha, S.N., Chopra, S., 2006. Selection of bricks and cooling pad for construction of evaporatively cooled storage structure. J. Inst. Eng. (India), Agr. Eng. Div. 87, 25–28.
Liberty, J.T., Okonkwo, W.I., Echiegu, E.A., 2013. Evaporative cooling: a postharvest technology for fruits and vegetables preservation. Intl. J. Sci. Eng. Res. 4 (8), 2257–2266.
Longmone, A.P., 2003. Evaporative cooling of good products by vacuum. Food Trade Rev. (Pennwalt Ltd.) 47, 13–16.
Mahangade, P.S., Mani, I., Beaudry, R., Müller, N., Chopra, S., 2020. Using Amaranth as a model plant for evaluating imperfect storages: assessment of solar-refrigerated and evaporatively-cooled structures in India. HortScience 55 (11), 1759–1765. https:// doi.org/10.21273/HORTSCI15249-20.
McCulloch, J.S.G., 1959. Soil temperatures near Nairobi 1954-1955. Q. J. R. Meteorol. Soc. 85, 51–56. https://doi.org/10.1002/qj.49708536306.
Roy, S.K., 1989. A low cost cool chamber: an innovative technology for developing countries (tropical fruits storage). In: Johnson, G.I., Champ, B.R., Highley, E. (Eds.), Postharvest Handling of Tropical Fruits. Australian Centre for International Agricultural Research, Canberra, A.C.T., Australia, pp. 393–395, 1994.
Roy, S.K., Khurdiya, D.S., 1986. Studies on evaporatively cooled zero energy input cool chamber for storage of horticultural produce. Indian Food Packer. 40, 26–31.
Stull, R., 2011. Wet-bulb temperature from relative humidity and air temperature. J. Appl. Meteorol. Climatol. 50, 2267–2269. https://doi.org/10.1175/JAMC-D-11- 0143.1.
Tano, K., Kamenan, A., Arul, J., 2005. Respiration and transpiration characteristics of selected fresh fruits and vegetables. Agron. Afr. 17 (2), 103–115. https://doi.org/ 10.4314/aga.v17i2.1662.
