Comparative Characterization of Trends and Patterns of Physical and Chemical Attributes of Optimal and Traditional Processed Cowpea Leaves

Seasonality in the availability of cowpea leaves has often limited their utilization and thus the promotion of preservation techniques that convert the vegetables into storable and stable forms. Te recommendations for the use of highly mechanized techniques in preservation are brought into question due to limited afordability among resource-constrained households that prefer less costly approaches. Terefore, this study used statistical techniques of principal component analysis to comparatively evaluate the trends of physicochemical quality of the two diverse approaches of processing cowpea leaves. Te study evaluated dehydrated cowpea leaves of diferent processing techniques from farmer groups and optimally processed using modern techniques for nutritional composition, phytochemical compounds, and colour changes. Sun drying techniques that excluded blanching had the least content of beta-carotene and ascorbic acid, 2.65 ± 0.95 and 21.80 ± 1.24mg/100g dry weight basis (dwb), respectively, accompanied by the most signifcant ( p < 0 . 001) deterioration of colour (7.74 ± 3.49) than techniques that included. Whereas the antinutrients declined, the diference did not signifcantly difer ( p > 0 . 05) based on preservation techniques. With factor analysis determining optimal nutritional quality for cowpea leaves at 8 weeks after emergence, sun drying had the highest loss of beta-carotene and ascorbic acid, 66.7–80.1% and 53.7%–58.3%, respectively ( p < 0 . 001), whereas mineral leaching, reduction of antinutrients, and colour changes were more pronounced in dehydration techniques incorporating fermentation as pretreatment. For the traditional preservation techniques, increasing retention of minerals resulted in aggravated losses of beta-carotene and ascorbic acid, whereas in the mechanized techniques, this was not the case. In concluding that the mechanized techniques have a better combination of attenuating losses of micronutrients, the study recommends that in promoting the utilization of traditional preservation techniques, low-cost processes like steam blanching can help improve the nutritional quality of the product.


Introduction
Te vast utilization of cowpea leaves in sub-Saharan Africa (SSA) for food and nutrition security is due to their rich nutritional composition and their production in a variety of agroecological zones [1,2]. Western and eastern Africa accounted for 85.1% and 7.8% of the 14.4 million hectares of global production area of the crop, respectively [3]. Te crop has a dual purpose of utilization for its grains and vegetables, which has made it popular among many communities in SSA [4]. Moreover, the vegetable is rich in phytochemicals with health-promoting properties that have aided the continued push for their utilization, including among urban communities [2,5]. Moreover, the crop has been exploited for nonfood uses for fodder [6,7]. Its relative importance in the agricultural sector is due to its high productivity and stability, tolerance to environmental stress, economic viability, and low environmental impact coupled with its capacity to promote environmental conservation [8]. Additionally, the crop has production fexibility to permit its production in mono and mixed cropping [9].
Cowpea leaves constitute one of the most consumed African leafy vegetables in Kenya [10]. Te coastal areas of the country are among the regions with the highest production and consumption of the vegetable [7]. Tus, the vegetable forms a major component of their diet. However, seasonal availability of the crop often constrains its extended utilization among households. Reliance on fresh forms often exposes communities to shortages of such vegetables, especially in areas where there is much reliance on subsistent production [11]. Communities in the arid and semiarid lands (ASALs) of the country often incorporate traditional preservation techniques in order to enhance their utilization of the vegetable [11]. Traditional processing of the vegetables ranges from sun drying techniques to hurdle technology of blanching or cooking and drying and fermentation [12]. Over a quarter of the households in coastal areas were found to be reliant on traditional processed vegetables to overcome the shortage occasioned by seasonal availability [11]. Te nutritional quality of processed products difers based on the technique utilized in processing. Whereas, Kirakou et al. [5] recommended blanching and fast-drying techniques, including solar drying for use in the processing of cowpea leaves due to their maximum nutrient retention; Owade et al. [11] established that sun drying, a more afordable technique, is the most utilized in the cowpea leaves value addition in the coastal and eastern arid and semiarid lands (ASALs) in Kenya. Terefore, it is not sufcient to be dismissive of these technologies as less efcient ways of availing vegetables for consumption despite the limited practice among communities.
Tis research study contributes to the promotion of the adoption of value-added techniques among producing households to enhance the all-season availability of the vegetable. Te study sought to establish the trends and patterns in the retention and degradation of physicochemical attributes in value-added cowpea leaves subjected to either optimal or traditional processing techniques. Te goal of this study was to use comparative statistical approaches to evaluate optimal and traditional processing techniques of cowpea leaves. Tis approach presents an objective way of mapping the diferences and similarities in the physicochemical quality of vegetables subjected to different processing techniques. Te study will shape nutrition information that is disseminated in nutrition interventions that promote value-added practices, especially in resourceconstrained settings in SSA.

Study Design.
Te study was undertaken in two phases. In the frst phase, a survey was conducted among farmer groups processing cowpea leaves in Taita Taveta county, located between the latitudes 2°30′ and 4°30′ South and the longitudes 37˚36′ and 39°14′ East and Kitui county that lies between the latitudes 0°10′ South and 3°0′ South and the latongitudes 37°50′ East and 39°0′ East. Te study examined the methods of processing cowpea leaves. Optimized processing techniques were selected from a review conducted by Owade et al. [7]. Te selected dehydration techniques for cowpea leaves included solar drying, sun drying, and oven drying compared to local processing techniques that included sun and shade drying.

Sample Collection and Preparation.
A total of 30 samples of dehydrated cowpea leaves were obtained from 4 and 2 farmer groups in Taita Taveta and Kitui counties, respectively, who practised value-added practices for cowpea leaves. Samples were collected based on the processing technique: eight fresh, four shredded sun dried, two unshredded sun dried, two blanched sun dried, and two shadow dried from Taita Taveta county and four fresh, four shredded sun dried, and four unshredded sun dried from Kitui county. For this reason, the county of residence of the group was treated as a block rather than a factor. Samples were collected based on batches available during the weeklong study in the 2 areas. All samples were collected in May from Taita Taveta county and in October from Kitui county, 2020, when the leaves are most available, about 4 weeks after emergence, for at the time is when value addition of leaves are most practised. Collected samples, each weighing 2 kg, were placed in air-tight sterile polythene bags and placed in cooler boxes at −10C for transportation to the University of Nairobi Laboratories for analysis. Landraces were used. Te 30 samples were subjected to compositing where ∼200 g obtained from each batch were mixed in a plastic tub based on similarity of the processing technique and similar farmer group to minimize efects of extraneous outliers due to individual variations in sample types. A total of 12 composites were obtained and evaluated for colour changes before being frozen awaiting nutritional analysis.

Phase II
2.3.1. Experimental Designs. Tis study utilized a combination of two experimental designs: the full factorial arrangement in the evaluation of the optimal maturity stage for harvesting of the cowpea leaves (Repert. Bot. 1 : 779.1843) and the completely randomized experimental study in the evaluation of the optimal processing. In the full factorial experiment, the experimental factors were the period of maturity and the variety of the cowpeas. On the other hand, in the completely randomized study, the experimental factor was the processing technique.

Evaluation of Optimal Stage of Maturity for Harvesting of Cowpea Leaves. (i) Experimental Arrangement.
Two diferent varieties of cowpeas, Machakos 66 (M66, a dual purpose variety) and Kunde Mboga (predominantly for the leafy vegetables), were subjected to evaluation of their maturity indices and nutritional quality. Te experiment was done in three diferent blocks to eliminate the efect of extraneous factors such as gradients of the soil and moisture.
Te planting was done in two diferent planting seasons (April to August) and (September to August) in three different blocks at the University of Nairobi Field Station. Te spacing of the plants was 60 cm by 30 cm as determined by Muniu [13]. Te leaves of the diferent varieties were harvested at intervals of four weeks after emergence (WAE), transported to the laboratory, and stored at −20°C awaiting analysis for nutrient and antinutrient contents.
(ii) Study Site. Te study was done at the feld station located at the College of Agriculture and Veterinary Sciences at the University of Nairobi, Nairobi County, Kenya. Te feld is located West of Nairobi County along the latitudes 1°1 5′ S, the longitudes 36°44′E, and an altitude of 1820 m above sea level [14]. Te area has an annual rainfall of 1060 mm, which has a bimodal distribution with long rains between March and May and short rains between October and December [15,16]. Te temperature ranges from 13.7 to 24°C. Te soils of the area are deep well-drained and dark reddish-brown to dark brown [14].

Evaluation of Optimal Dehydration of Cowpea Leaves.
Optimal processing techniques of cowpea leaves have higher retention of the physicochemical quality when hurdle technology, a combination of pretreatment and dehydration, is used [7]. Te study employed a completely randomized experimental design with the investigative factor being the dehydration technique. Kunde Mboga variety of cowpea leaves was harvested at optimal maturity, washed, cut, and divided into 2 batches. Te frst batch (15 kg) was steam blanched for 2 min followed by immersion in ice-cold water and divided into 6 equal parts. Two parts each were dried using a forced air oven drier (at a temperature of 60°C for six hours), solar (at a maximum temperature of 70°C), and the open sun on a raised platform. Drying was done till the leaves attained a moisture content below 15%. Te second batch was divided into 4 equal parts of 2.5 kg, and sugar and salt were added to each portion at 5% and 2%, respectively, as established in our earlier study [17]. Fermented vegetables were dried in a forced air oven (at a temperature of 60°C for six hours) and solar dried (at a maximum temperature of 70°C) till moisture content below 15% was attained. Dried cowpea leaves were evaluated for colour changes and then stored at −20°C awaiting evaluation of nutrient and antinutrient composition.  21-1968 [20]. Standardization of the dichlorophenolindophenol (DCPIP) reagent was accomplished by titrating it three times with 2 mL of standard ascorbic acid solution (0.02% in 5% metaphosphoric acid). Te titration was done until a rose-pink colour persisted for >5 s. A blank of 5% metaphosphoric acid was titrated three times. To a 10 g sample of the dehydrated cowpea leaves, 60 mL of 5% metaphosphoric acid was added and fltered using gravity through glasswool into a 100 mL volumetric fask. Tis was made to volume and 10 mL was placed into a 100 mL conical fask and titrated against DCPIP. Te titre of the dye was determined as per equation (1). Te amount of ascorbic acid in the dried vegetables was determined as per equation (2).

Analysis of Physicochemical Attributes of Processed
where n is the mg of ascorbic acid per ml of titrated standard solution, in this case, it is (mg of ascorbic aci d × 2)/50, a is the titre of the standard used, and b is the titre of the blank.
where x is the titre volume used for the sample, c is the titre used for the blank, f is the mg of ascorbic acid equivalent to 1 ml of DCPIP solution, e is the assayed volume (2 ml), v is the volume of the initial assay solution (10 ml), and y is the volume of sample aliquot (10 ml).

2.4.3.
Determination of Beta-Carotene Content. Beta-carotene was determined calorimetrically using the spectrophotometry method adopted through modifcation of the methods described by Biswas et al. [21].
Preparation of a standard curve: A stock solution of betacarotene (5% purity, Sigma Aldrich) was dissolved in acetone to make a concentration of 1 mg/ml. Te stock solution was used to make working solutions of 32, 16, 8, 4, 2, 1, 0.025, 0.125, 0.062, 0.03, and 0 μg/ml. A standard curve was generated on a UV-VIS spectrophotometer (Hitachi U-2900, Tokyo, Japan). Te concentration was expressed in mg/ml. All standards were protected from the light by covering them with aluminium foil.
Sample preparation: Dried samples of the vegetables (1 g) were mixed with 5 ml of chilled acetone and left at 4°C for 15 minutes with occasional shaking, vortexed at high speed for 10 minutes, and centrifuged at 1370 × g for 10 minutes. Te supernatant was collected in a tube and the extraction was repeated until a clear supernatant with no colouration was obtained. Te supernatant was flled to a volume of 50 ml. Te supernatant was passed through a Whatman paper no. 42 and the absorbance read at 450 nm. Te concentration of beta-carotene was calculated as per the following equation.
where b is the beta-carotene in mg/g, C is the concentration determined as per the calibration curve, V is the volume of the extract in ml, and M is the weight of the sample used in extraction.

Determination of Mineral Content.
Te mineral (calcium, zinc, and iron) content was determined in duplicates using an atomic absorption spectrometer as per the AOAC [18] methods. A 2 g sample of cowpea leaves was ashed at 550°C, followed by boiling in 10 ml of 20% hydrochloric acid in a beaker. Te boiled solution was fltered into a 100 ml standard fask and then read using atomic absorption spectrometry (Buck Scientifc 210 VGP, USA).

Determination of Oxalate Content.
Te oxalate content of the traditionally preserved cowpea leaves was determined in duplicate as per the procedures by AOAC [18] methods. About 1 g of preserved cowpea leaf samples were weighed into a 100 ml conical fask. To it, 75 ml of 3 mol/l of H 2 SO 4 was added and the solution was stirred using a magnetic stirrer. Te solution was fltered through Whatman flter paper no. 1, and the fltrate was collected in a 250 ml conical fask. From this sample fltrate, 25 ml of it was titrated against hot (80-90°C) 0.1 N KMnO 4 solution, with a persistent faint pink colour (30 seconds) indicating the endpoint. Te oxalate content was calculated as 1 ml of 0.1 N of KMnO 4 is equivalent to 0.006303 g of oxalate.
2.4.6. Determination of Nitrate Content. Te nitrate content of the traditionally preserved cowpea leaf and cowpea vegetable samples was determined in duplicate by modifcation of procedures described by Gaya and Alimi [22]. Samples of the vegetable were ground using a mortar and pestle and to 10 g of the ground samples, 70 ml distilled water was added followed by 2.5 ml of 4% NaOH. Te mixture was heated at 80°C for 25 minutes, with occasional shaking during heating. Tereafter, the resultant solution was fltered into a 100 ml volumetric fask through a futed flter paper and flled to mark with distilled water to form a mixture 2. About 4 ml of mixture 2 was pipetted into an icecold test tube followed by the addition of 1 ml of 1% Ag 2 SO 4 , 7 ml of 98% H 2 SO 4 , and 1 ml of 5% phenol solution to form mixture 3 that was left to stand in the dark for 20 minutes while occasionally shaking. Mixture 3 was transferred into a 50 ml separating funnel and toluene was added (mixture 4) and further shaken for 5-10 minutes to mix. Te upper phase of mixture 4 (organic phase) was retained, while the aqueous phase was discarded. Te organic phase was washed twice with 10 ml of distilled water, and each time, the aqueous phase was discarded. Te organic phase was extracted further by the addition of 10 ml of 10% Na 2 CO 3 and shaken for a minute. Te extract was collected in a test tube. Te absorbance was read at 407 nm in a UV-VIS spectrophotometer (Hitachi U-2900, Tokyo, Japan). Standard curves were generated by varying the concentrations of sulphuric acid, Na 2 CO 3 , and the phenol and reaction time of standard nitrogen nitrate solution. Te quantity of nitrates was calculated as shown in equation (4): where C is the concentration of the nitrates in the samples as per the calibration curve, S is the volume of fltrate used to read the absorbance, W is the weight of slurry used, and F is the total volume of the fltrate.

Determination of Total Phenolic Compounds.
Te total phenolic content of the preserved samples of cowpea leaves was determined using the Folin-Ciocalteu procedure that was adopted through modifcation of the methods described by Abong et al. [23]. A 5 g sample of the vegetables was subjected to extraction by adding 5 ml of methanol followed by a twenty-four-hour extraction at 25°C. Te extract was centrifuged at 3226 × g for 10 min, and the resulting supernatant was used to determine the total phenolic content.
To an aliquot of 1 ml of methanolic extract in a 10 ml volumetric fask, 2.5 ml of tenfold dilution of Folin-Ciocalteu reagent (1 : 10 dilution with distilled water) was added, followed by 2 ml of 7.5% (w/v) sodium carbonate solution. Te mixture was topped to volume and incubated at 45°C for 15 minutes. Te samples were read against a standard calibration curve of gallic acid monohydrate prepared by obtaining 0.25, 0.5, 1.0, 1.5, and 2.0 mg/ml, followed by a similar treatment as the methanolic extracts. Te calibration curve of the standard was in mg/ml with an R 2 of 0.995. Distilled water was used as the blank. Te samples were read at 765 nm using a UV-VIS spectrophotometer (Hitachi U-2900, Tokyo, Japan), and the total phenolic content was expressed as mg per gallic acid equivalent (GAE) per gram as per equation (5): where P is the total phenolic content in mg/g, C is the concentration determined as per the calibration curve, V is the volume of the extract in ml, and M is the weight of the sample used in extraction.

Determination of Flavonoid Contents.
Te favonoid content of the samples was determined using the aluminium chloride colourimetric procedure by modifying the procedures described by Abong et al. [23]. A standard calibration curve was generated using a catechin solution. From a stock solution of 100 μg/ml (w/v of methanol) of catechin, aliquots of 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0 ml were obtained and transferred into fve 10 ml volumetric fasks containing 4 ml of water, followed by addition of sodium nitrite and left to rest for fve minutes. After fve minutes, 0.3 ml of 10% (w/v) aluminium chloride was added and allowed to rest further for six minutes. To the rested mixture, 2 ml of 1N sodium hydroxide was added and flled to volume. Te standard curve was calibrated in mg/ml with an R 2 of 0.995. Te methanolic extract obtained as per the extraction procedures for determining total phenolics was subjected to treatment similar to catechin standards. Te concentration of total favonoids was determined in milligrams of catechin equivalents per gram (mg. CE. g −1 ) as per equation (6): where F is the total favonoid content in mg/g, C is the concentration determined as per the calibration curve, V is the volume of the extract in ml, and M is the weight of the sample used in extraction.

Determination of Antioxidant Activity.
Te antioxidant activity of the leaves was determined using the 2, 2-diphenyl-1picrylhydrazyl (DPPH) procedure by modifying the methods described by Abong et al. [23]. Methanolic extract of preserved samples of cowpea leaves was prepared by mixing 0.25 g of the sample with 10 ml of 80%(v/v) of methanol, with overnight extraction in a shaker. About 1 ml of the methanolic extract, standard Trolox solutions (0, 5, 10, 25, and 50 μg/ml), and blank were pipetted into boiling tubes and 0.002% of DPPH (prepared using absolute methanol) was added to each. Te mixture was shaken briefy and read immediately upon the addition of DPPH at 515 nm in a UV-VIS spectrophotometer (Hitachi U-2900, Tokyo, Japan). A standard calibration curve of Trolox was used to calculate the antioxidant activity of the preserved cowpea leaves in μM Trolox equivalents (TE) per 100 g dry weight.

Determination of Colour
Changes. Te L * , a * , and b * and chroma and hue angles of the dried cowpea leaves were determined as per the procedures described by the manufacturer (PCE Instruments, 2014). Using the CSCQ3 software, the hue, chroma, and ∆E were calculated based on equations (7)-(10). Te value of L * represented the lightness of the vegetable samples (more positive values have lighter colour intensity), the value a * represented the measure of redness (positive), greyness (zero), or greenness (negative), and the value b * represented the measure of yellowness (positive), greyness (zero), or blueness (negative).
Hue angle(Ho) � arctan(b/a)(for + a and + b values), Hue angle(Ho) � arctan(b/a) + 180(for -a and + b values or for -a and -b values), 2.6. Statistical Analysis. Statistical analysis of the data was done using the R language for programming (ver. 4.0.3, [24]). A one-way analysis of varaince with blocking was used to test for mean diferences induced by local processing techniques on the physical and chemical qualities of cowpea leaves. For means that were signifcantly diferent, p < 0.05, Tukey's honest signifcant diference (HSD) in the Agricolae package was used to separate them. One-way ANOVA, without blocking, was used to test diferences in means of physical and chemical attributes of optimally preserved cowpea leaves and means were separated using Tukey's HSD. Principle component analysis was used to map patterns of nutrient retention in the samples. Te data for optimization of the maturity stage of cowpea leaves were analyzed using two-way ANOVA. Akaike's Information Criterion of the AICmodav package was used to select the model that best explained the variation of the nutritional composition of cowpea leaves, and Tukey's HSD of the Agricolae package was used to separate means.

Physicochemical Qualities of Traditionally Processed Cowpea Leaves.
Tere was a signifcant (p < 0.05) diference in the proximate composition of cowpea leaves based on the processing technique (Table 1). Whereas the crude fat content (4.3 ± 0.3 g/100 g dry weight) was high in blanched and sun-dried leaves than in fresh and other preserved samples, there was a decline in the crude ash content (p < 0.05). Tis low crude ash content is pronounced with signifcantly (p < 0.05) low mineral, iron, and calcium contents in the blanched and sun-dried leaves ( Table 2). Te leaching of minerals explained the declining sodium, iron, and zinc contents in the blanched dehydrated vegetables. Te use of water rather than steam blanching aggravates the loss of the minerals in water [25]. On the positive end, the moisture levels reported in the traditionally preserved products were within the recommended limit by specifc standards that permit up to 15% [26]. Te moisture content established in this study was also within the range of documented studies by Owade et al. [7]. It is imperative to maintain moisture below 15% in order to prevent quality deterioration occasioned by microbial growth due to less optimal moisture content [27]. Hag et al. [28] established a critical limit of ≤14% for the growth of microorganisms in dehydrated African leafy vegetables. Te utilization of artisanal traditional processing for preservation resulted in a signifcant loss (p < 0.05) of micronutrients. Sun-dried leaves had the least amount of beta-carotene and ascorbic acid. Tis is caused by losses induced by photo-oxidation activity catalyzed by UV radiation [29]. UV-induced oxidation converts the beta-carotene from the provitamin A form to derivatives with less vitamin A activity. Additionally, exposure to factors such as heat that induces drying and oxygen also accelerates the oxidation of both beta-carotene and ascorbic acid [30]. Without blanching, the losses are aggravated due to increased oxidation of the two micronutrients with antioxidant activity [25,29]. Whereas the antioxidant activities of the preserved samples signifcantly (p < 0.001) decreased with the application of traditional preservation techniques, the antinutrient content in the vegetables remained invariably high (Table 3). Moreover, degradation in colour also occurred with a signifcantly high deviation (p < 0.001) Journal of Food Quality occurring in preservation techniques that did not combine blanching (Table 4). Te colour coordinates for b * , a * , and chroma and hue angles were signifcantly diferent from the fresh vegetables. It is recommended that in processing, such leaves are subjected to blanching as a pretreatment as it attenuates loss of beta-carotene and antioxidants and improves colour retention [25,31]. Pretreatment like blanching is known to improve colour retention through attenuating continued oxidation of coloured pigments [25].

Optimization of the Stage of Maturity for Harvesting
Cowpea Leaves. Te promotion of cowpea leaves in the food security initiative hinges on their rich micronutrient and phytochemical composition [6]. Using the WSSplot, the optimal number of clusters was determined as three for the classifcation of the nutritional and antinutrient contents of cowpea leaves harvested at diferent maturity stages. Cluster one had the optimal content of protein and micronutrients (Table 5). Tis had the advantage of establishing optimal trends of increasing nutrient content while minimizing the accumulation of antinutrients. Whereas seasonal variation had no diference in loading in the diferent clusters, the variety of cowpea leaves and the stage of maturity of leaves difered in loading amongst the three clusters. Kunde Mboga variety and cowpea leaves harvested at eight weeks after emergence (WAE) had the highest loading in cluster one (Tables 6 and 7). In choosing the most optimal period of harvesting, the cluster with the highest number of positive values for the normalized means of protein and micronutrient contents and the highest number of negative values for the antinutrient contents was selected and cluster two met this criterion. Independent variable for Model_1 is the main efect of type of variety, season, and period of harvesting in weeks after emergence; Model_2 is main efect of type of variety; Model_3 is main efect of season; Model_4 is main efect of Te values are mean ± SD of duplicates. Values with diferent letters in the superscript along the column are statistically diferent. S1, blanched sun dried; S2, fresh leaves; S3, shadow dried; S4, unblanched sun dried. All the variables are in dry matter basis except for moisture content.  period of harvesting in weeks after emergence; and Model_5 is the interaction factors of type of variety, season, and period of harvesting in weeks after emergence. Seasonal variation had no efect on the micronutrient or antinutrient content of cowpea leaves. Te beta-carotene, sodium, and calcium contents were higher in the Kunde Mboga variety than in the dual-purpose variety of "Machakos 66" (Table 8). Similarly, the favonoids and total phenolics had higher concentrations in the Kunde Mboga variety than in "Machakos 66" (Table 9). Te Kunde Mboga is an improved variety that was developed for its vegetables rather than its grains, thus the better nutrition profle of the vegetables than the M66, dual-purpose, and variety. With an increasing period of growth, cultivated cowpea leaves accumulate more antinutrients. Tis is not the trend for the micronutrients and protein content Muchoki [32]; Kirigia et al. [2]. Cowpea leaves like other African leafy vegetables present the property of high content of phytochemicals including antinutrients [33]. Beta-carotene was higher in cowpea leaves harvested at 8 WAE than at 4 or 12 WAE. Zinc and calcium contents increased with increasing WAE, whereas sodium content decreased. Te interaction between variety and stage of harvest afected zinc and crude fbre contents (Table 10). At 4 WAE, cowpea leaves from Machakos 66 variety had higher crude fbre content than "Kunde Mboga." At 12 WAE, leaves from the latter had higher crude fbre content than the former (Figure 1). Zinc content in the leaves harvested from "Machakos 66" decreased over time of harvest. In the Kunde Mboga variety, it increased ( Figure 2). Whereas it is desirable to have higher micronutrient content in the vegetable, the fbre content should not be high. More mature leaves tend to have higher fbre content Ohler et al. [34], thus being tougher for consumption as vegetables [35].

Nutrient Composition of Optimally Processed Cowpea
Leaves. Te hurdle concept (combination of two preservation techniques) ofers the advantage of attenuating quality losses in the vegetables while improving the desirable product attributes such as sensory and textural properties [7,29]. Te focus is primarily on higher retention of micronutrients such as vitamins and minerals, with the loss of the latter frst depicted by crude ash content. Fermented dehydrated vegetables had signifcantly (p < 0.001) high crude ash than blanched leaves (Table 10). In optimizing the fermentation process of cowpea leaves, Owade et al. [17] added salt (sodium chloride) at a proportion of 2% (w/w), explaining the elevated crude ash level in the fermented dehydrated leaves. Te fbre content in the fermented dehydrated leaves signifcantly (p < 0.001) declined, whereas the moisture content signifcantly (p < 0.001) increased. Soluble fbre is also broken down during lactic acid fermentation Nyman [36], so the fermented leaves have lower fbre content than the blanched. Oven drying techniques achieved the least amount of moisture of all the dehydration techniques (p < 0.001), which is desirable for prolonging the shelf-life of the dried product. Dried leaves with high    Te values are mean ± SD of duplicates. Values with similar uppercase letters followed by a diferent lowercase in the superscript along the column are statistically diferent.   (23)   Journal of Food Quality moisture content encourage microbial spoilage of the product and thus will have a short shelf-life [28]. Te efect of photo-oxidation in the reduction of the labile micronutrients explains the loss of 66.7-80.1% in beta-carotene and 53.7-58.3% in ascorbic acid content in the sun-dried vegetables (Table 11). Hurdle technology combining fermentation and sun drying had the least retention of beta-carotene (19.8%) and ascorbic acid (41.7%). Combining dehydration techniques with fermentation resulted in a reduction in iron and zinc contents in the vegetables as compared to those combining dehydration with blanching. On the other hand, the sodium content of all the dehydrated leaves combined with fermentation was relatively high, more than even the fresh vegetables (p<0.001). Incorporating salting in the dehydration processes for enhanced preservation resulted in reduced beta-carotene, ascorbic acid, zinc, and iron contents, whereas sodium, calcium, and moisture contents were relatively higher [37]. Te fermentation period of 16 days enhances the leaching of micronutrients from the processed leaves into water [17].
Incorporating fermentation in the processing of dehydrated cowpea leaves signifcantly (p < 0.001) reduced the antinutrient contents of the leaves (Table 12). Te nitrates, followed by the oxalates, had the highest decline when fermentation techniques were included in the processing. It is desirable that the two compounds are low in foods due to their negative efects on the bioavailability of micronutrients [38]. It is also desirable that the techniques retain the physical attributes, such as the colour of the products. Te use of a combination of two preservation techniques in the processing of cowpea leaves seeks to minimize quality loss while improving sensory and textural properties [7,29]. Whereas all dehydration techniques induced deterioration of the colour of the preserved samples, sun dried samples processed through hurdle technology had the highest deviation (p < 0.001), see Table 13. Exposure to UV radiation during sun drying destroys the colour pigments, including the chlorophyll and the carotenoids, thus the high deviation in colour [39].

Comparative Characterization of Retention of Physicochemical Quality of Optimally and Traditional Processed
Cowpea Leaves. Essentially, dehydrated vegetables should have a closer similarity in quality to fresh vegetables when cooked to enhance consumer acceptability of these preserved forms. In fnding the blanched solar-dried leaves as the most acceptable in the evaluation of the impact of  preservation techniques on sensory attributes, deterioration of textural properties and colour was minimized by Natabirwa et al. [40]. Artisanal techniques that employ the use of sun drying techniques excluding blanching as a pretreatment result in alteration both in textural and colour properties [41]. Te correlation maps generated through principal component analysis for the nutrient composition of locally processed cowpea leaves showed that with limited retention of beta-carotene content, the antioxidant activity and crude protein content also deteriorate ( Figure 3). Additionally, the utilization of techniques that improved the retention of the minerals (sodium, calcium, zinc, and iron) aggravated the losses of antioxidant activity and beta-carotene. However, blanching has been found to attenuate deterioration of antioxidants and colour, so their inclusion improves quality amelioration. Te optimally processed cowpea leaves that incorporated blanching as a pretreatment had higher retention of crude protein and beta-carotene ( Figure 4). Te loss of the minerals was not aggravated by the use of processing techniques that improved the retention of beta-carotene. Even with blanching, the use of hot water as in the case of traditional processing rather than steam as in the optimal techniques has the disadvantage of aggravating the leaching of minerals [5]. Limited leaching of minerals coupled with attenuation of labile nutrients such as betacarotene improves nutrient retention.    Journal of Food Quality Variables located in the same quadrant are positively correlated; those located in opposed quadrants are negatively correlated. Te distance of a variable from the origin measures the quality of representation in the 2 principal components (Dim 1 � frst principal component (micronutrients and physical attributes) and Dim 2 � (macronutrients and antioxidant activity). Variables closer to the margin of the circle are represented by the 2 principal components. L * , a * , and b * are the coordinates for the colour space and PCA represents the principal component analysis.
Variables located in the same quadrant are positively correlated; those located in opposed quadrants are negatively correlated. Te distance of a variable from the origin measures the quality of representation in the 2 principal components (Dim 1 � frst principal component (micronutrients and antinutrients) and Dim 2 � second principal component (macronutrients)), with variables closer to the margin of the circle being represented by the 2 principal components. L * , a * , and b * are the coordinates for the colour space and PCA represents the principal component analysis.

. Conclusion
Tis study concludes that the mix of techniques utilized in the traditional preservation of cowpea leaves lacks balance in the trends of retention of essential nutrients in the products. Te incorporation of mechanized techniques introduces a balance and attenuates losses of these essential micronutrients. Even so, this should not be the reason for dismissing the traditional processing techniques as a means of improving vegetable availability among households for the leaves still had signifcant amounts of beta-carotene, zinc, and iron, some of the micronutrients whose defciencies are prevalent in Africa. Tis study would thus recommend that initiatives promoting the utilization of similar traditional techniques of preservation evaluated in this study should coopt for some of the low-cost pretreatments such as steam blanching in order to improve the nutritional quality of the products.

Data Availability
Te data used and/or analyzed during the current study are available from the corresponding author upon request.

Consent
Tis manuscript was presented as a part of a thesis at the Faculty of Agriculture, University of Nairobi Kenya. Te source is cited as Owade [42].

Conflicts of Interest
Te authors declare that they have no conficts of interest.