ASSESSING IMPACTS OF ICE ON FISH LANDING BEACHES OF SUBA DISTRICT, KENYA - ABILA RICHARD O.

STRATEGY FOR

ASSESSING IMPACTS OF ICE ON FISH LANDING

BEACHES OF SUBA DISTRICT, KENYA

Consultancy Report for Africa Now

Consultants: Dr. Richard O. Abila & Mr. Kenneth Werimo

Funded by the EU

December 2006

Strategy for Assessing Impacts of Ice in Subat

Africa Now Consultancy Report Richard Abila & Kenneth Werimo

2

Summary Ice is a vital input in the fish marketing chain for reducing post-harvest losses and a useful tool for enhancing the bargaining position of fishers in the market. In Lake Victoria, the supply of ice has always been far short of demand, causing significant post-harvest losses to fishers and fish traders and contributing to the overall underperformance of the sector. Presently only industrial fish processors have ice, giving them undue advantage in the market vis-à-vis fishers. The current strategy to address this problem and reduce post-harvest losses involves establishing small and medium-scale ice producing units based around the lake, which can supply fishers and traders with ice to preserve fish at the landing sites as well as during fish transport. Suba, the leading fish producing district in the country, is set to benefit from ice supply intervention intended primarily to benefit fishers. The main objective of this study was to develop a strategy for monitoring the impacts of ice on beneficiary fisher communities. In particular the study aimed to identify impact indicators, generate baseline data on those indicators and suggest a plan for monitoring impacts. The study was conducted in eight fish landing beaches in Suba District, involving two of the four administrative divisions in the district, and covering the most important fish producing areas. The study was based mainly on primary data generated through interviews and daily monitoring of fish catch, prices and post-harvest losses at beach level and at fish processing factory. Available secondary data and information were also applied.

The study has identified indicators to be used for assessing the impacts of ice intervention and obtained baseline data (the current situation) of each of the identified indicators. Three sets of indicators to assess the impacts of ice have been presented, namely; for reduced post-harvest losses, for increased incomes from fish and for increased employment opportunities in ice and fisheries-related business. It has estimated the current level of post-harvest losses at beach and IFP level and suggested a program for monitoring changes of the impact indicators

Acknowledgement We acknowledge Africa Now which, through the support of the European Union, facilitated this study. We also express gratitude to fishers, fish traders, agents, Mbita Ice Plant, Industrial Fish Processor and other stakeholders of Lake Victoria, who voluntarily spared their time to answer our questions. Be assured that your involvement in this study will greatly contribute to the improvement of socio-economic conditions of fishers through better understanding of impacts of ice intervention, how to enhance the positive and minimize the negative impacts.



3

Table of Contents

Summary.................................................................................................................. 2

Acknowledgement.................................................................................................... 2

Table of Contents..................................................................................................... 3

1. Introduction .......................................................................................................... 4

1.1 The situation in Suba District.............................................................................. 5

1.3 Review of demand and supply of ice in Suba District......................................... 6

1.4 Causes of post-harvest losses along the fish distribution chain ......................... 8

2.0 Objectives of the consultancy........................................................................... 10

2.1 Study outputs ................................................................................................... 11

2.2 Study Area ....................................................................................................... 11

3.0 Methodology..................................................................................................... 13

3.1 Study Approach................................................................................................ 13

3.2 Assessment of post-harvest loss...................................................................... 14

3.3 Methodological approach for assessing quality loss ........................................ 15

4.0 Results ............................................................................................................. 17

4.1 Identified beneficiaries ..................................................................................... 17

4.2 Long-term impacts ........................................................................................... 17

4.3 Short-term impacts........................................................................................... 18

4.4 The Impact indicators....................................................................................... 18

4.5 Current status of the impact indicators............................................................. 18

5.0 Proposed program for monitoring impacts of ice.............................................. 29

5.1 Choice of indicators.......................................................................................... 29

5.2 Methodology for monitoring.............................................................................. 29

5.3 Monitoring sites ................................................................................................ 30

5.4 Timing of monitoring......................................................................................... 30

5.5 Suitable Ice containers..................................................................................... 30

5.6 Design of fish boxes......................................................................................... 30

5.7 Reasons for non-adoption of ice containers and fish boxes............................. 32

6.0 Conclusion ....................................................................................................... 33

7.0 References....................................................................................................... 33

8.0 Annexes ........................................................................................................... 34



4

1. Introduction

Ice is one of the most important inputs in the fish marketing chain. Sanitary and phytosanitary standards in fish trade, including the Hazard Analysis Critical Control Point (HACCP) system, are pegged on the assumption of sufficient supply of quality ice. To successfully develop and sustain a modern industrial fish processing and marketing sector largely depends on the availability of ice. However, supply of ice has always been far short of demand on Lake Victoria, causing significant post-harvest losses to fishers and fish traders and has contributed considerably to the overall underperformance of the sector. Presently the main sources of ice around Lake Victoria are the industrial fish processors (IFPs), which mainly produce ice for their own use, but even they do not have enough to cover all stages of the fish supply chain. Fishers are particularly vulnerable, as they cannot safely preserve fish, process or transport it to distant markets without ice. Besides the high post-harvest losses, lack of ice puts fishers in a weaker price bargaining position vis-à-vis the IFPs, who tend to take advantage of their possession of ice to exploit fishers.

The current strategy to address this problem and reduce post-harvest losses involves establishing small and medium-scale ice producing units based around the lake, which can supply fishers and traders with ice to preserve fish at the landing sites as well as during fish transport. To ensure fishers reap maximum benefit from this venture, they need to own these plants or hold controlling share of the capital. Recent initiatives include the Bunyala Ice Plant located in Busia district, with ice making capacity of 15-ton per day, which is owned by the Bunyala fisher community. The European Union is also supporting a program for developing physical infrastructure on selected fish landing sites around Lake Victoria, which may include establishing community-owned small-scale ice making units. Suba district is also set to benefit, with the planned expansion of the Mbita Ice Plant Kenya Ltd (MIPKL) from a capacity of 1MT to approximately 55 MT of ice per day. This has been realized through the joint support of USAID, under the Global Development Alliance facility (GDA), and that of a Nairobi-based fish processing plant, WE Tilley. In this arrangement USAID has provided a new ice-making unit capable of producing 30 MT, while WE Tilley is to donate two used machinery with capacity to produce about 25 MT per day.

The success of ice production in Suba district is expected to give direction and the impetus for establishment and expansion of independent ice production plants around the lake. Probably owing to its position as the leading fish producing district, the MIPKL project has received much goodwill, judging by the number of partner organizations associated with this venture in different roles. Besides USAID and WE Tilley, the Kenya Business Development Service (KBDS), the Kenya Fisheries Department, Africa Now, KREP-Development Agency, OSIENALA, among other organizations, have all expressed interest to work with MIPKL. In particular, Africa Now, with the financial support of the European Union, is poised to play a critical role in the development of commercial ice distribution strategy for MIPKL, by supplying insulated containers to selected fish landing beaches.



5

The ultimate success of ice production and distribution in Suba will be measured by its impacts, particularly on the vulnerable groups such as fishers and artisanal fish traders. This study aims to contribute to this effort by developing a strategy for assessing impacts of ice on beneficiary fisher communities in the district. While ice soon to be produced and distributed by MIPKL will provide a useful test case, the results of this study may be applicable to similar ice supply strategies by other parties in Suba and elsewhere around the lake.

1.1 The situation in Suba District

Suba is the most important fishing district in Kenya, landing about 40,000 MT a year, which is equivalent to about 25% of the country’s fish output. Of this, Nile perch constitutes about 50%, while tilapia is 15%. The most recent comprehensive survey of fishing capacity in Suba District was conducted in April 2006 by the Lake Victoria Fisheries Organization in collaboration with the Kenya Fisheries Department and the Kenya Marine and Fisheries Research Institute (KMFRI). The report shows that there are about 15,600 fishers in the district, an increase of about 34% from 2004 (Table 1), and representing about 34% of fishers in Kenya. There are 4,910 fishing crafts in the district, indicating an increase by 37% in the past two years. Of these, 811 are motorized crafts, which have tremendously increased by 61% since 2004. Correspondingly, gillnets have tremendously increased nearly five times in the past two years.

The district currently has 108 fish landing beaches, indicating a slight increase by 4% since 2004, and representing about 33% of all landing sites in Kenya. Of these, only 32 beaches have a fish banda, representing a mere 30%, while there is no working cold room on any beach. Only two beaches have a landing jetty, 10 have a dry fish store while five are supplied with electricity. There has been an appreciable improvement in electricity connection since 2004, when only one beach had electricity, although the current level of access to electricity is still unacceptably low. Comparatively, the number of beaches accessed by all weather roads increased from 23 in 2004 to 30. Only 3 beaches have potable water, indicating a decrease by 5 in the past two years, while 73 beaches had public toilet facilities.

This situation puts Suba in a very disadvantaged position in its efforts to deliver high quality fish. Certain basic facilities and services are needed to support production and successful marketing of the large quantity of fish landed in the district. However, as shown above, such vital facilities like fish banda, landing jetty, fish store and cold room are seriously inadequate. Other services like electricity supply, all weather roads, potable water and sanitary facilities are also critically insufficient. Lack of these important facilities and services mean that the district stands to experience higher post-harvest losses and lower fish prices than in the ideal situation. Without ice fishers have fewer alternatives, particularly on the choice of markets, and most likely have to sell fish to IFP agents that have ice. Therefore, one way to improve fishers’ socio-economic condition is to ensure they have access to ice.



6

Some of these basic facilities are needed to support the production and distribution of quality ice. Lack of these facilities therefore translates into higher costs of ice production and distribution in Suba. For example, lack of cold rooms in landing beaches means that ice containers have to be provided as a component of an ice distribution strategy, thereby causing significant escalation of the costs. Unavailability of potable water implies higher costs of purification and treatment of water for making ice. In the same way, low access to electricity will seriously impair ice production and storage. The very fast increase in number of motorized crafts, however, presents a good opportunity for efficient distribution of ice. Likewise, growth in numbers of fishermen, fishing crafts and gears present potentially higher demand for ice. Table 1. Fishing facilities in Suba District (2000-2006)

* The Frame survey report gives this figure without explanation, but could be an error

Source: Frame Survey Results (2000, 2002, 2004 & 2006)

1.3 Review of demand and supply of ice in Suba District The most recent and reliable study on the status of demand and supply of ice in parts of Suba District was conducted in April-May 2006 through a consultancy commissioned by USAID (through its agencies ACDI/VOCA and KBDS). The study, though covering only beaches in Mfangano and Mbita/Rusinga divisions, developed a strategy for commercial ice distribution and generated baseline data and information on beaches that were likely to buy ice produced by MIPKL. The report indicates that on average about 78 MT of ice is currently supplied to Suba District every day, out of which nearly all the ice (96%) is brought in by six fish processing plants located in Kisumu (East African Seafood, FP 2000 and Peche), Nairobi (WE Tilley), Migori (Prinsal Ltd.) and Homabay (Capital Fish

2000 2002 2004 2006 No. of fishers 14,782 16,727 11,639 15,585 Landing sites 97 100 104 108 No. of fishing craft 4,051 3,267 3,575 4,910 Bandas (fish sheds) 19 24 12* 32 Cold rooms working 1 - 1 - Jetty 4 1 2 2 Fish stores 4 1 2 10 Beaches supplied with electricity - - 1 5 Beaches with toilet facilities - 49 62 73 Beaches with potable water - 5 8 3 All weather roads 12 36 23 30 No. of crafts using engine 308 335 503 811 Total no. of gillnets 86,464 37,527 28,797 157,609 No. of small seines 5,182 674 1,674 1,235



7

Company). The ice is held in insulated trucks owned by the fish processing plants and is for preserving only fish already purchased by them for transportation between the landing beaches and fish processing factories. Coca Cola Company supplies about 2 MT on average, for the sole use by sellers of their soda brands, while MIPKL produces just about 1 MT a day, which is the only ice that can be officially bought by any party, including artisanal fish traders, for transporting tilapia to urban centres (Table 2). Thus, fishers and BMUs have completely no access to ice, while a negligible quantity is available to artisanal fish traders. Once MIPKL becomes fully operational, the ice supply to Suba District is expected to increase by 55 MT to about 130 MT per day. This assumes that IFPs will continue producing and supplying ice to Suba District at current level.

Table 2. Ice supply to Suba District (April 2006)

Supplier Estimated Supply (MT/Day) % of Total

IFPS 75 96%

Coca Cola Co. 2 2.5%

MIPKL 1 1.5%

Total 78 100%

Source: Karuga and Abila (2006)

On the demand side, the actual and/or potential consumers of ice are; • Fishermen – for use during fishing expeditions and transportation; • Beach management units – for storage of fish at the beaches • IFPs and their agents – for transportation of fish to and at the factory • Independent Traders – for transportation of fish, mainly tilapia, to various local and

urban market destinations • Kiosk and beer bar owners - for cooling of soft drinks and beers

Ideally, ice is required at all levels of the fish supply chain from fish capture to market. Thus, prior to the fish processing factory, ice is required in the fishermen’s boats, at the landing sites during storage, in fish collector/ traders’ boats and in the fish trucks transporting fish from the landing beaches to fish factories. The ideal icing ratio is one unit of ice per unit of fish, although in practice much less ice is used.

The USAID consultancy report estimated the potential demand for ice in Suba District at between 200 MT and 250 MT, assuming that ice is used at the recommended rates at all levels of the fish supply chain from the fishing ground to the fish factory. In reality, however, ice is unlikely to be used at some levels, particularly by fishermen in their fishing boats. This reduces the actual demand for ice in Suba District, although even at



8

full operational capacity of MIPKL it is unlikely that it will meet the estimated potential demand. The high demand for ice puts a big challenge on the system in place to ensure that what is available is distributed and used efficiently among competing parties (fishers, BMUs and traders). Furthermore, in a free market arena the excess demand is likely to raise price of ice above its cost of supply. This may have the opposite effect as it raises the cost of supplying fresh fish to the factory or the market. It is therefore difficult to accurately predict what impact ice will have on low income fishers in the long term. Fishers will use ice only if they perceive the returns from iced fish as above the additional cost of purchasing ice.

1.4 Causes of post-harvest losses along the fish distribution chain

The main objective of a fish supply chain is to deliver all the landed fish to the final consumer in the best quality so that it fetches the best price. Due to inefficiencies in the system, this goal is often not attainable, resulting in post-harvest losses. Activities and fish quality/safety factors affect the level of post-harvest losses at various stages along the fish supply chain. The fish distribution chain of Lake Victoria from harvesting to arrival at the processing factory, usually called the upstream supply chain, may be divided into six main handling stages (unit operations) based on quality risk assessment as shown in Figure 1.

Fig 1. Stages of fish handling



9

i) Fishing ground

Fishing is normally carried out in designated fishing grounds chosen by the fishermen. The Fisheries Department and KMFRI regularly monitor any signs of contamination (biological or chemical) of the fishing waters. This is a quality assurance initiative to ensure that fish is harvested from non-polluted waters, thus minimizing the risk of contaminating the fish before landing. The specific quality concerns are; a) Handling on board fishing vessels: After landing fish from the gillnets, it is usually

thrown to the bottom of the canoe and covered with wet material such as papyrus leaves, nets or sisal bags and transported to the landing site.

b) Transport to landing: typically refers to the transport of fish by collector boats or by

other means from a fishing canoe or remote landing, to a first point of sale. Fish collectors who buy their fish from inside the lake or from beaches in Uganda and Tanzania tie the fish together in bundles and place in insulated containers, usually with inadequate ice.

ii) At landing This involves the time from when the fish is landed on the beach to the time when it is sold to the next stage for processing and/or for fresh fish mongering. Spoilage at this stage is due to poor preservation techniques and may cause fish to be downgraded and sold for a lower price. Theft and damage due to poor handling are causes of physical loss. The specific quality concerns at this stage are; a) Handling at the landing site: On arrival at the landing site fish is unloaded and

ferried to the fish banda either by wheelbarrows, baskets or trays. From the canoes fish are sometimes dragged or washed in the lake water which may impact negatively on the quality of fish as this is a Critical Control Point.

b) Handling at the fish banda: Fish is normally placed on the concrete tables and

sometimes on the floor at the banda, where weighing and selling is done by officials of the Co-operative or Beach Management Units (BMU). Fish may be taken and iced immediately if the buyers, mostly fish processors, are present. In case buyers are not present or the price being offered is not acceptable, fish is left exposed on the ground or table without any form of preservation. Later the fish is taken up by the insulated trucks where icing is done and transported to the processing factory.

Results from this study show that it takes approximately 8-12 hrs for a fishing trip (from time of setting the net, hauling and returning to the landing sites). The storage time of Nile perch without ice is estimated to be 12 hrs, and fish will still be suitable for distribution as grade ‘A’ if iced within this time. Usually fish stays at the landing for up to 12 hrs. This means that if fish is not iced at the landing site then a higher proportion will be rejected. This therefore means that the critical stage in the fish supply chain that should get priority for icing is at the landing site.



10

iii) Factory level handling a) Transport to processing factory – Sometimes fish trucks stay on the beach for

more than three days so as to accumulate enough fish to take to a factory. Delays at this stage may lead to unnecessary spoilage, especially if no ice is used.

b) Processing - fish are often processed to extend shelf life and a number of different

methods may be employed. Insect infestation, damage during the process and poor handling and packing can cause both physical and quality loss during processing.

iv) Wholesale and retail market handling

a) Transport to wholesale - fish may be transported to a wholesale market by various means. Both physical and quality loss can occur during transport due to damage and spoilage. Loss may be incurred by the processor who still owns the fish or by a trader who has purchased the fish from the processor.

b) Wholesale - fish normally pass through a wholesale marketing stage where they

are sold in bulk and then sold on, in smaller lots, to retailers. Storage of cured fish at this stage can be for several months and this can leave the product susceptible to insect infestation and attack by other pests.

c) Transport to retail - this refers to loss that can occur during transport of the fish

between the wholesale market and a retail stage.

d) Retail - the point in the chain at which fish are mainly sold to the final consumer. Retail stage loss occurs because of contamination due to repeated handling, spoilage or insect infestation.

v) Export level handling

a) Transport to export - This stage deals with the loss that can occur during transport to an exporter or during export, usually from the fishing ground, at the landing or processing stages. Loss can result due to poor icing and delays or damage during transport .

b) Export - fish may be processed in a factory before export or simply bought by a

trader and taken directly out of the country as fresh fish. Where fish is exported directly and not via a processing factory fast means of transport to end consumer is critical. Spoilage is most likely to result from inadequate icing.

2.0 Objectives of the consultancy

The main objective of this study is to develop a strategy for monitoring the impacts of ice on beneficiary fisher communities. In particular the study aims to identify impact indicators, generate baseline data on those indicators and suggest a plan for monitoring impacts.



11

2.1 Study outputs

In line with above objectives the consultants have;

i) Identified indicators to be used in assessing the impacts of ice intervention to post-harvest losses

ii) Obtained baseline data (the current situation) of each of the identified indicators

iii) Estimated level of post-harvest losses at beach and IFP level

iv) Suggested a program for monitoring changes of the impact indicators

2.2 Study Area

The study was conducted in eight fish landing beaches in Suba District, distributed in two of the four administrative divisions in the district, and covering the most important fish producing areas. They included beaches in five islands and Mbita Mainland. The involved landing beaches were;

• Mbita Mainland - Mbita Beach • Rusinga Island- Nyagina Beach • Mfangano Island - Sena, Yokia and Wakula beaches • Takawiri Island beach • Remba Island beach • Ringiti Island beach

The beaches for this study were purposely selected from the 15 – 18 beaches that have been earmarked to participate in the pilot commercial ice distribution strategy by MIPKL. The selection criteria was to ensure fair geographical distribution by involving all islands and Mbita Mainland and the inclusion of the major fish landing beaches which are likely to provide the most important market for ice. Additional data was obtained from one IFP based in Kisumu.



12

Fig 2. The study area

Mfangano

Remba

Ringiti

Takawiri

Rusinga

MIPKL

MBITA



13

3.0 Methodology

3.1 Study Approach

This study was based mainly on primary data generated through interviews and daily monitoring of fish catch, prices and post-harvest losses at beach level and at fish processing factory. Available secondary information, In particular, the report of the USAID consultancy study and frame survey report for 2006, also provided useful secondary data and information.

To meet the objectives of this study, the consultants carried out three key tasks;

1. Field visits and conducting interviews with key informants, including BMU leaders, fishers, artisanal fish traders, agents and IFPs.

2. Participatory preliminary monitoring of the identified impact indicators to determine their status during the months of September and October 2006.

3. Discussing indicators with key stakeholders to create awareness and incorporate their input.

The consultants made four field visits, each lasting 4 - 6 days, to selected landing beaches. The first visit was to select landing sites for the study and identify impact indicators through a participatory process involving all key community-level stakeholders. In this process, the consultants, based on their expertise, experience and literature search, came up with a set of possible impact indicators which they brought up for individual and group discussion at beach level. Principally the discussions involved individual fishers, artisanal fish traders, BMUs and IFP agents so as to identify and prioritize suitable impact indicators that were most applicable to fisher communities. Based on the results of the first visit, the consultants designed a structured questionnaire for interviewing fishers and artisanal traders to get the baseline information on the current status of the baseline indicators and a data collection sheet for monitoring impact indicators.

During the second visit fishers and traders were interviewed on the sampled landing sites using questionnaires. The survey involved a total of 59 fishers and traders. At the same time the consultants initiated the first preliminary monitoring of impact indicators lasting 7 – 10 days (for September 2006) on each of the beaches. BMUs were actively involved in monitoring the impact indicators. In the third visit, the consultants in close collaboration with the BMUs, conducted the second preliminary monitoring of impact indicators lasting 7 – 10 days (for October 2006) on each of the beaches. The fourth visit was for verification of results and discussing the findings with the communities and other stakeholders. Following the beach level visit, monitoring was conducted at an IFP in Kisumu for 17 days to determine quantities of fish brought in per day and what proportions are rejected as low quality.



14

Immediately following field work, data was entered in SPSS package and analyzed to produce statistical outputs, while non-parametric data was collated and interpreted, for this report.

3.2 Assessment of post-harvest loss The key immediate impact of an ice provision intervention is the reduction of post-harvest losses. Post-harvest loss can be defined and assessed from various aspects, most commonly by the following indicators; i) Physical loss: fish or fish product that is discarded and not sold for whatever

reason, be it due to spoilage, or damage. Commonly used measures for physical loss are;

a) Percent physical loss: the proportion of fish that is lost (physical loss) at a given stage of the chain expressed as a proportion of the total weight of fish entering the stage. Fish can be physically lost from the chain as a result of insect infestation, spoilage, damage, theft.

b) Weight lost (kg) - the weight of fish physically lost at that stage.

ii) Quality loss: quality deterioration can lead to a loss in value. Fish is downgraded and sold for a lower price than it could have been sold for if the quality was better. This is commonly measured by;

a) Percent sold at reduced price: the proportion of fish sold for a reduced or low price, because of spoilage or damage (quality loss), expressed as a percentage of the weight of fish entering that stage

b) Weight at low price - the weight of fish sold for a reduced price iii) Value loss: The value of the loss is the difference between the price attained

and the price the fish would have sold for had its quality been good. a) Reduced price ratio: the price of low quality fish expressed as a proportion

of the price of good quality fish. b) Best price - the price of good quality fish per kilogram, expressed in local

currency. The price refers to the good quality fish as it is sold on to the next stage in the chain. In order to calculate the price, data may be required on the kg weight of traditional units by which fish are sold.

c) Reduced price - the price that low quality fish are sold for.

iv) Total loss This incorporates the sum of the value of physical and quality loss for the whole chain expressed as a percentage of the value of the fish that enters the chain before any loss occurs using the value of the fish at that stage. It is commonly measured in terms of;

a) Total losses considered as a % of initial value of catch - the total loss in value terms expressed as a percentage of the value of the fish that enters the chain before any loss using the value of the fish at that stage.



15

b) Total losses considered as a % of maximum value of fish - the total loss in value terms expressed as a percentage of the value of the fish assuming no loss and using the best price attained.

c) Gross total losses in (local currency) - the loss in local currency once any revenue from by-products has been deducted.

3.3 Methodological approach for assessing quality loss The DFID Post Harvest Fisheries Research Programme has developed a standard method for assessing quality state of raw Nile perch, called the Field Based Loss Assessment Method (FBLAM). The method gives a systematic description of various grade categories of Nile perch (from grades “A” to “C”), the quality attributes of each grade based on a clearly defined criteria of organoleptic assessment and use value of each grade (Table 3). Points are awarded for each of the organoleptic attributes attained, and the total score used to decide the overall grade. Its noted that even though grades “B” ad “C” are both low quality grades, they are often merged and rarely are any fish thrown away as being unfit for human consumption. In practice even though grades “B” ad “C” are both low quality grades, grade “B” is increasingly being accepted for processing especially due to the high demand of fish. Therefore there are now only two quality grades i.e. grade “A” for export quality and grade “C” which, though rejected for export processing, is now purchased, iced and taken to urban centers such as Nairobi, Kisumu and Mombasa for local distribution.



16

The scheme used to assess Nile Perch freshness based on that developed by Gram et al (1987)

Table 3: Quality assessment of raw Nile perch at 20°C-30°C Outer appearance

Description Skin Gill-colour Gill-odour Eyes Texture Score Grade

Freshness allowing for export

Natural brilliance, pink-silver, scales firmly attached

Red-purple, maroon, no slime

Fresh, seaweeds

Transparent, yellow sheen

Firm, hard (rigor), elastic

9-8 A

Reduced freshness allowing for local or regional distribution if kept in ice.

Brilliance, few dull parches, scales firm.

Reddish, little pinkish slime

Natural, fresh

Reddish, convex

Firm, elastic 7-6 B

Insufficient freshness; as such unfit for human consumption. Sometimes limited consumption after frying.

Greyish, darkening loss of metallic brightening

Bleached parches, slime

Light rotten Red milky, flat

Soft 5-4 C

Unfit for human consumption.

Yellow slime, greenish fins Dull, loose scales

Green, slime Rotten, sour, strong off-odours

Sunken, turbid

Soft, buttery, fingerprints leave impressions

3-1 Reject



17

4.0 Results

4.1 Identified beneficiaries

This study has identified the main potential beneficiaries of ice supply intervention in Suba District as; IFPs, BMUs, individual fishers and artisanal fish traders transporting tilapia and grade “C” Nile Perch to urban markets. Given a choice, the IFPs would prefer to buy ice closer to the landing beaches, rather than produce it in their fish processing plants, thereby reduce the production and transport costs. However, since supply of ice is far short of demand, it is important that current short-term interventions aim to benefit BMUs, fishers and artisanal traders, so as to improve their bargaining position. Since ice is to be introduced gradually, only a few beaches will benefit in the short term. For instance, the pilot commercial ice distribution strategy by MIPKL will only serve 13 -15 beaches in the first year and, depending on the success, gradually roll out to more beaches, eventually extending to all 108 beaches in the district in the long term.

4.2 Long-term impacts A number of benefits are envisaged as a result of provision of ice. In the long term the following positive impacts may be realized; i) Poverty alleviation – due to increased incomes, reduced post-harvest losses,

better prices for higher quality fish and access to better markets. ii) Food and nutrition security – due to higher incomes and reduced post-harvest

losses iii) Healthy fisheries – If fishers and other stakeholders perceive the fisheries to be

of very high value, it may be in their interest to take more stringent management measures to conserve stocks.

iv) New employment opportunities – new opportunities in ice and fish related

business. In the contrary, certain negative long-term impacts may also result, including; i) Food insecurity – Some communities around the lake have depended on Nile

perch rejects as a major source of protein. Reduced post-harvest losses mean less availability of reject fish.

ii) Overfishing – Higher prices can be an incentive to increase fishing capacity, for

example, fish for longer hours or use more nets. The actual impacts, therefore, can be known only after careful assessment over long period of time. These long-term impacts cannot be monitored within the time frame envisaged for this study.



18

4.3 Short-term impacts The main concerns of this study are the short-term impacts. These are the impacts that can be realized within a realistically short time from when the ice is provided to the fish landing beaches, and therefore, can be monitored within the envisaged time frame. These short-term impacts contribute to the long-term impacts discussed above.

4.4 The Impact indicators

Through a participatory process, the following indicators were identified and prioritized for assessing short-term impacts of ice intervention. The indicators have been categorized into three groups; i) Indicators for reduced post-harvest losses

a) Reduction in quantity of reject (grade C) fish b) Reduction in percentage of reject fish c) Reduction in income loss per day d) Reduction in percentage of fish value loss

ii) Indicators for increased incomes from fish

a) Quantity of grade ‘A’ fish b) % increase in income for fisher/trader/processor c) Increase in price of export-quality fish

iii) Indicators for increased employment opportunities in ice and fisheries-

related business a) Quantity of fish landed b) % increase in use of insulated containers c) % increase in no. of ice users d) % increase in number of IFPs obtaining ice from MIPKL e) % increase in price of ice f) % increase in people involved in fresh fish business

4.5 Current status of the impact indicators In order to assess the current status of impact indicators in September and October 2006, baseline data were collected using two methods. First, through interviewing individual fishers and traders using questionnaires, and, secondly, by conducting preliminary monitoring of quantities of fish landed, proportions rejected, and the prices of respective grades “A” and “C” fish during the two-month study period. The data was compiled and analyzed for each of the eight landing beaches and one IFP involved in the study and then aggregated to make generalized conclusion. Each of the indicators has been analyzed by considering five descriptive statistical measures, namely;

• Mean for September 2006 data • Mean for October 2006 data • Mean for combined September and October data



19

• Standard deviation for September and October data. This represents the average/ standard disparity of each of the data from the mean, indicating the extent of ‘variation’ in the variables between the days.

• Maximum figures for September and October • Minimum figures for September and October

The results are presented in Tables 4-15 for different indicators. Tables 4 gives analysis of the current status of quantity of fish landed in each of the beaches. Tables 5, 6, 7, 10 and 11 collectively indicate the levels of post harvest losses. Table 8 gives the current price level of export-quality fish while Table 9 gives the status of income from fishery to the BMUs. Unless where specified, the combined September/October means are considered more accurate and used as representative means for the study period. However, in some cases, reliable data was not obtained for both months, in which case, only one month data has been used. 1. Post-harvest losses The level of post-harvest losses has been assessed using four indicators;

a) Quantity of reject fish (Grade C) b) percentage of reject fish (grade C ) c) income loss per day d) Percentage loss in fish value per day

The quantity of reject fish (grade C) is presented in Table 5. These results show wide disparity in the quantity of reject fish in the 8 fish landing beaches involved in this study, chiefly reflecting the quantity of fish landed on each of the beaches and accessibility factors. Mbita and Ringiti lead in quantity of fish rejects, with an average of about one tonne per day on each of the two beaches, while the least quantity of loss is experienced in Nyagina Beach, with a mean of 45 Kg per day. The aggregated mean for all eight beaches is about 347 Kg of rejected fish per day. The quantity of reject fish may not be a very good indicator as it gives the absolute physical loss, which is likely to increase with the quantity of fish landed. A better measure of physical loss indicator is the percentage of reject fish (i.e. quantity of reject fish as a percentage of total landings). The current status of this indicator is presented in Tables 6 for fish landing beaches and Table 7 for IFP. Results show very wide variation on this indicator between the landing beaches. Nyagina Beach had the lowest percentage of post-harvest loss at 1.5% per day, followed by Mbita Beach at 4.2%. On the other hand, Yokia and Wakula, had the highest percentage of rejects at over 30% and 27% respectively. Both beaches are very small, with among the least fish catches, but they are very poorly accessible and have no ice containers. The aggregated mean percentage of rejected fish in all 8 beaches was estimated at 13.6%. The IFP data from one factory monitored over 17 days indicate a low mean reject rate of about 2%. The indicator of loss in fish income combines both the physical loss and reduction in prices due to low quality fish. It takes into account that in most landing beaches even the very low quality fish (Grade C) is not discarded, but sold to local consumers at a small fraction of the grade ‘A’ fish. Thus, the income is reduced by an amount



20

equivalent to the product of the quantity of reject fish and the price. Remba and Ringiti experienced the highest loss in fish income at an average of about Ksh 98,299 and Ksh 77,489 per day respectively, although with very high daily variation. On the other hand Nyagina, Yokia and Wakula experienced the least daily income loss at Ksh 4,400, Ksh 9,576 and Ksh 9,455 respectively. The aggregated mean loss in daily income for all eight beaches was estimated at Ksh 30,330 per day. The last indicator of post-harvest loss is the percentage loss in value of fish. This indicator represents the loss in income expressed as percentage of total potential income. It gives a measure of how the lost income compares with the maximum income that would have been obtained if all the fish was kept in top quality (grade ‘A’), and therefore fetched the best price. Nyagina and Mbita Town beaches had the lowest percentage loss in fish value at 1.3% and 2.8%, while Sena, Yokia and Wakula had the highest loss levels at 30.9%, 26.7% and 23.6% respectively. The mean aggregate of percentage loss in value of fish for all the surveyed beaches was estimated at 12.9%. 2. Total fish income The second category indicator is the total income earned by BMUs from Nile perch. This was estimated by monitoring and recording the quantity of fish landed on the beach, the quantity of high quality (Grade ‘A’) fish and its price and quantity of reject fish (Grade ‘C’) and the price. The total daily income is the sum of income generated from selling grade ‘A’ and grade ‘C’ fish (i.e. grade ‘A’ fish x price of grade ‘A’ + Grade ‘C’ x price of grade ‘C’). Use of ice is expected to gradually increase both the quantity and price of grade ‘A’ fish, corresponding to higher quality fish accessing better markets. Hence, the total fish income will be expected to increase over time. In this study, total income has been represented by three indicators;

a) Percentage of grade ‘A’ fish b) Total fish income to BMUs c) Price of export-quality fish

The percentage of Grade ‘A’ fish represents the converse of the percentage of Grade ‘C’ fish discussed above. Beaches that recorded high percentage of Grade ‘C’ fish would record low percentage of Grade ‘A’ fish. Therefore, results presented above under post-harvest losses in this aspect can be used to indicate changes in percentage of grade ‘A’ fish. The total daily fish income to BMUs is presented in Table 9. The results show that Mbita and Ringiti had the highest total daily fish incomes at an average of about Ksh 1.6 million and Ksh 1.1 million respectively. On the opposite end, Yokia and Wakula had the lowest mean daily fish incomes at about Ksh 28,415 and Ksh 33,943 respectively. These results largely reflect the quantity of fish landed on the various beaches. The aggregated mean daily fish income for all the eight landing beaches was estimated at Ksh 469,023. The last indicator in this category is the price of export-quality (grade ‘A’) fish. As fish quality improves with increased usage of ice, better markets can be accessed, resulting



21

in increased price of quality fish. This study recorded the prices of both export quality and reject fish, even though the latter may not be useful as an indicator of impacts of ice usage. Unlike the other variables, fish prices remained relatively very stable over the two months of study. In Nyagina, Yokia and Sena, the prices did not change at all during that period. Remba and Mbita recorded the highest mean prices at 118 Ksh/Kg and 115 Ksh/Kg respectively, while Yokia had the lowest mean price at 86 Ksh/Kg. In overall, mean fish prices ranged at between 86 Ksh/Kg and 130 Ksh/Kg, with an aggregated mean price of 108 Ksh/Kg for all eight beaches. 3. Employment opportunities in ice and fish related business Increased use of ice is expected to have a trickle down effect, particularly in generating new employment opportunities in the fishery. There will be need for new ice containers and transport facilities. Ice can also support expansion of fish trade through allowing traders to venture into new markets. In this study, the employment impact of ice is assessed through five indicators;

g) Quantity of fish landed h) Use of insulated containers i) Number of ice users

The quantity of fish landed gives an indication of the employment potential of the fishery. In reality it is unlikely that ice may be the main contributing factor to increased fish catches. However, it can provide a strong enough incentive to cause fishers to increase their fishing effort (i.e. availability of ice may cause fishers to fish longer hours or use more nets, expecting that they will incur very little post-harvest losses). Increased quantity of fish landed means greater capacity of the fishery to employ more traders and those engaged in support enterprises. This study collected baseline data of the current status of fish quantities landed daily on the eight beaches (Table 4). Remba and Ringiti recorded the highest level of landing, at an average of about 13,993 Kg and 10,940 Kg respectively. Comparatively, Yokia and Wakula recorded the least level of catches at about 432 Kg and 374 Kg respectively. The aggregate for all eight beaches was estimated at an average of 4,457 Kg. The number of users of insulated containers can give an indication of demand both for ice and the containers themselves. The current level of usage of insulated containers in the study beaches has been estimated at about 8.5% of fishers (Table 14). However, all these were in the mainland beaches of Mbita and Nyagina (Table 15). In fact, these beaches did not have proper insulated containers but they depended on insulated trucks that were ever on standby at the beaches. The availability of ice and insulated trucks on these two beaches could be the main reason for their low post-harvest losses. Similarly the number of ice users gives an indication of the capacity of business both in ice selling and fish trade. The current level of ice usage has been estimated at about 10%, all of whom are either agents or fish traders (Figure 3, Table 13). None of the fishers interviewed was actually using ice. Another possible indicator for employment impacts is the number of people involved in fresh fish trade, as this is likely to increase with increased usage of ice.



22

Table 4. Quantity of total landing per day (Kg)

Mbita

Town Nyagina Takawiri Sena Remba

Ringiti Yokia Wakula Aggregate

Mean daily landings (Sep 06) 5,769 2,845 - 620 13,253 - 432 374 3,882

Mean daily landing (Oct 06) 3,481 2,591 1,678 - 14,733 10,940 - - 6,685

Mean daily landing (Sep-Oct 06) 4,889 2,727 1,678 620 13,993 10,940 432 374 4,457

Standard deviation of daily landing (Sep-Oct 06) 1,484 1,033 958 204 6,515 4,647 77 189 5,230

Maximum daily landing (Sep-Oct 06) 7,600 5,000 3,619 876 28,339 19,397 550 712 8,262

Minimum daily landing (Sep-Oct 06) 2,242 1,320 347 326 7,334 5,501 266 74 2,176

Table 5. Quantity of reject fish per day (Kg) at beach level

Mbita

Town Nyagina Takawiri Sena Remba Ringiti Yokia Wakula Aggregate

Mean quantity of reject fish (Sept 06) 362 59 - 168 1,276 - 126 94 348

Mean quantity of reject fish (Oct 06) 23 33 92 - 938 908 - - 399

Mean quantity of reject fish (Sep-Oct 06) 232 45 92 168 1,107 908 126 94 347

Standard deviation of reject fish (Sep-Oct) 229 48 53 141 559 386 26 42 415

Maximum daily reject fish (Sept-Oct 06) 1000 200 199 351 2,342 1610 166 182 756

Minimum daily reject fish (Sept-Oct 06) 10 5 19 25 403 457 80 14 126



23

Table 6. Percentage of rejected fish per day at beach level

Mbita


Mean % rejected fish (Sep 06) 6.3 1.7 - 24 9.4 - 30.2 27.1 16.5

Mean % rejected fish (Oct 06) 0.7 1.3 5.5 - 7.2 8.3 - - 4.6

Mean % rejected fish (Sep-Oct) 4.2 1.5 5.5 24 8.3 8.3 30.2 27.1 13.6

Standard deviation of % reject (Sep-Oct 06) 3.8 0.9 0.03 17 1.5 1.9 9.9 10 11.5

Maximum % reject fish (Sept-Oct 06) 17.5 4 5.5 52 12.1 8.3 62.4 41.5 25.4

Minimum % reject fish (Sept-Oct 06) 0.3 0.3 5.5 7 3.6 8.3 16.6 11.8 6.7

Table 7. Percentage of rejected fish at factory level

No. of days monitored 17

Mean quantity of landed fish per day 6,192

Mean % of rejected fish per day 2.0

Standard deviation of % rejected fish 0.6

Maximum % rejected fish per day 2.9

Minimum % rejected fish per day 0.8



24

Table 8. Price of export-quality fish (Ksh/Kg)

Table 9. Total daily income (Ksh)

Mbita


Mean daily income (Sept 06) 685,575 314,786 - 53,277 1,564,767 - 28,415 33,943 446,794

Mean daily income (Oct 06) 346,330 293,228 159,018 - 1,668,924 1,074,931 - - 708,486

Mean daily income (Sep-Oct 06) 534,548 303,288 159,018 53,277 1,564,767 1,074,931 28,415 33,943 469,023

Standard deviation of daily income (Sep-Oct) 222,708 112,877 91,058 13,344 767,924 461,424 6,367 21,100 567,188

Maximum daily income (Sept-Oct 06) 899,000 542,400 344,981 70,560 3,309,696 1,899,839 37,420 79,400 897,912

Minimum daily income (Sept-Oct 06) 222,625 146,900 33,078 34,388 804,115 513,573 14,560 7,110 222,044

Mbita Town

Nyagina Takawiri Sena Remba Ringiti Yokia Wakula Aggregate

Mean price of fish (Sept 06) 124 113 - 113 118 - 86 116 112

Mean price of fish (Oct 06) 100 113 100 - 118 105 - - 107

Mean price of fish (Sep-Oct 06) 115 113 100 113 118 105 86 116 108

Standard deviation of fish price (Sep-Oct) 12 0 1.6 0 1.5 2.4 0 4 10.8

Maximum fish price (Sep-Oct 06) 127 113 100 113 120 110 86 130 112

Minimum fish price (Sep-Oct 06) 100 113 95 113 117 100 86 115 107



25

Table 10. Income loss per day

Table 11. Percentage loss in value of fish per day

Mbita


Mean % loss in value (Sept 06) 4.3 1.5 - 30.9 6.2 - 26.7 23.6 15.5

Mean % loss in value (Oct 06) 0.5 1.2 4.7 - 5.3 6.7 - - 3.7

Mean % loss in value (Sept-Oct 06) 2.8 1.3 4.7 30.9 6.2 6.7 26.7 23.6 12.9

Standard deviation of % loss in value (Sep-Oct) 2.5 0.8 0.01 27.9 2.1 3.4 8.7 8.7 12

Maximum % loss in value (Sep-Oct 06) 11.2 3.5 4.7 83 9.1 6.8 55.1 36.1 26.2

Minimum % loss in value (Sep-Oct 06) 0.2 0.8 4.6 6.8 2.7 6.6 14.7 10.4 5.9

Mbita Town

Nyagina Takawiri Sena Remba Ringiti Yokia Wakula Aggregate

Mean loss in daily income (Sept 06) 30,706 5,810 - 16,783 98,299 - 9,576 9,455 28,438

Mean loss in daily income (Oct 06) 1,740 3,185 7,790 - 83,161 77,489 - - 34,673

Mean loss in daily income (Sep-Oct 06) 18,840 4,410 7,790 16,783 98,299 77,489 9,576 9,455 30,330

Standard deviation of loss in daily income (Sep-Oct) 19,142 4,718 4,464 13,911 50,905 33,363 1,966 4,192 36,264

Maximum loss in daily income (Sep-Oct 06) 81,000 19,600 16,918 34,398 210,816 36,527 12,616 18,000 53,734

Minimum loss in daily income (Sep-Oct 06) 750 490 1,622 2,450 35,078 136,846 6,080 1,400 23,089



26

Table 12. Fishers’ statistics

Fishers's Descriptive statistics

58.0 .5 6.0 2.2 1.259.0 1.0 14.0 7.7 4.057.0 3.0 2000.0 253.3 383.655.0 2.0 1500.0 173.5 274.539.0 2.0 300.0 57.4 77.543.0 2.0 60.0 19.8 12.356.0 80.0 120.0 112.6 9.5

Time period it takes to transport fish from the source (Hrs)On average time period fish take stored at landing (Hrs)Quantity of fish on average landed or purchased Kgs.Quantity of grade A fish landed/purchased Kgs.Quantity of grade C fish landed/purchased Kgs.Price for grade C quality Ksh/KgPrice for grade A quality Ksh/Kg

NMini-mum

Maxi-mum Mean

Std.Deviation

Current usage of ice for fresh fish preservation

89.8%

10.2%

No

Yes

Fig 3. Ice usage by fishers



27

Table 13. Category of ice users

Usage of ice for fresh fish preservation * Position of respondent Crosstabulation

4 2 666.7% 33.3% 100.0%

53 53100.0% 100.0%

53 4 2 5989.8% 6.8% 3.4% 100.0%

Count% within Usage of iceCount% within Usage of iceCount% within Usage of ice

Yes

No

Usage of ice for freshfish preservation

Total

Fisher Agent TraderPosition of respondent

Total

Table 14. Usage of insulated containers by fishers

Use of insulated container

5 8.5 8.5 8.554 91.5 91.5 100.059 100.0 100.0

YesNoTotal

ValidFrequency Percent Valid Percent

CumulativePercent



28

Table 15: Access to insulated containers by BMUS

Beach * Use of insulated container Crosstabulation

% within Beach

100.0% 100.0%100.0% 100.0%

12.5% 87.5% 100.0%100.0% 100.0%

44.4% 55.6% 100.0%100.0% 100.0%100.0% 100.0%100.0% 100.0%

8.5% 91.5% 100.0%

RembaTakawiriNyaginaGembeMbita townSenaYokiaWakula

Beach

Total

Yes No

Use of insulatedcontainer

Total

Table 16 Dimensions and capacities of fish boxes

Type Material Volume (1)

External Dimensions (mm)

Tare weight (kg)

Capacity of fish (kg) d/

Stack nest

Plastic Plastic Plastic

30 60 100

800 x 450 x 150 800 x 450 x 270 900 x 495 x 355

2.5 3.7 5.0

17 35 56

Stack Plastic Plastic Plastic Plastic Aluminium Wood

42 60 70 90 76 60

600 x 368 x 214 813 x 480 x 178 844 x 514 x 190 850 x 515 x 260 832 x 370 x 260 812 x 470 x 178

2.6 3.5 5.0 6.0 6.0 7.0 b/

25 35 40 50 43 35



29

5.0 Proposed program for monitoring impacts of ice

The program to monitor impacts of ice has to deal with the following questions;

i) Choice of indicators: What are the impact indicators to be monitored?

ii) Method for monitoring indicators: What methods will be used to monitor impacts?

iii) Monitoring sites: How many sites will be involved in impact assessment/ monitoring?

iv) Timing: How long will monitoring be undertaken?

5.1 Choice of indicators

This study has identified a number of indicators that should be applied in monitoring the impacts of ice in Suba District. Baseline data has been obtained on all the indicators, therefore, monitoring will basically compare how (the direction and magnitude) the indicator variables have changed in relation to the baseline. The following are the indicator variables suggested to be applied in a monitoring program;

i) Indicators for reduced post-harvest losses

• Reduction in quantity of reject (grade C) fish • Reduction in percentage of reject fish at beach level • Reduction of post-harvest losses at factory level • Reduction in income loss per day • Reduction in percentage of fish value loss

ii) Indicators for increased incomes from fish

• Increase in quantity of grade ‘A’ fish • Percentage increase in income for fisher/trader/processor • Increase in price of export-quality fish

iii) Indicators for increased employment opportunities in ice and fisheries-

related business • Increase in quantity of fish landed • Percentage increase in use of insulated containers • Increase in number of ice users • Increase in people involved in fresh fish business

5.2 Methodology for monitoring

This study used a combination of two methods to get baseline data on impact indicators;

a) a survey of fishers and traders using a questionnaire

b) Daily monitoring of basic fish quantity and quality data for 7-10 days per month for two months.

The two methods have provided sufficient baseline data on the identified impact indicators. No difficulty was experienced in the application of either method. The same



30

methods applied here can be maintained. However, it is important to ensure that the process is made as participatory as possible with adequate involvement of BMUs.

5.3 Monitoring sites

To effectively take advantage of the baseline data generated in this study, monitoring impacts of ice should be done on the same eight beaches and IFP. Those beaches in this study that will be excluded in the pilot ice distribution arrangement will still need to be monitored and the data used as control.

5.4 Timing of monitoring

The fishery is subject to seasonal annual variations, hence data taken at a particular time of the year might not be comparable to a different season. It is therefore very important that the timing of monitoring should reflect the conditions at the time of the baseline survey. It is therefore recommended that monitoring be done in the period extending from September – November once a year.

5.5 Suitable Ice containers The proposed intervention will need a complementary input of insulated ice containers. The USAID consultancy report gave an insight on suitable containers for distributing ice from MIPKL. Insulated containers are critical for fish preservation without which losses in the form of quality deterioration and hence reduced selling price is inevitable. The range of containers that have been tried for ice and fish storage around the study area comprise locally-made containers made of wood and plain iron sheet; plastic containers (with or without insulation) and insulated fiberglass containers all of varying sizes depending on the specific needs and source. However, there was currently very minimal usage of any of these types of ice-storage containers at the landing beaches. Ideally, the choice of ice containers should consider a number of critical factors. These include unit cost, efficiency in ice storage, durability, appropriateness with respect to size as per the needs of fisher folks, maintenance cost and after-sale service, among others. Taking into account the average unit cost, efficiency, durability and preferences of the fisher folks, MIPKL preferred the use of fiberglass ice containers. An important advantage of the fiberglass variety is that they can maintain ice freshness for 3-4 days compared with plastic containers which are said to achieve 2-3 days. According to various ice container users, they are hardy, more durable and easy to clean.

5.6 Design of fish boxes

While large size containers are suitable to be placed on landing beaches, fishers ideally require smaller insulated boxes for keeping fish in ice during fishing trips. The boxes should be able to fit in fishers boats. Fish boxes are commonly made of wood, aluminium or plastic. Wood was the traditional material for fish boxes but is being increasingly replaced now by plastic for reasons of cost and problems of usage. Wood



31

is extremely difficult to clean properly, has a short life compared to plastic and requires frequent repair. Aluminium is easier to keep clean but is prone to damage by rough handling making them difficult to stack. Plastic (thermoplastic high density polyethylene) is easy to clean, will withstand rough handling and allows detail design requirements to be incorporated in the mould.

Boxes used in the fishing industry are often subjected to abuse by rough handling and should be designed to withstand such treatment. In the water they are often subjected to fierce gravitational forces caused by vessel motion. Particular attention should be given to the design of the handles or recesses for lifting, which should be strong enough to carry the load. A stack formed from loaded boxes should be rigid and not tend to flex.

The tare weight of plastic and aluminium boxes is constant and not subject to age or environmental conditions. Wooden boxes will absorb water and can almost double in weight when saturated making accurate weighing of fish in the wooden boxes difficult and increasing transport costs. Typical box weights are given in Table 16. Fish boxes should be designed to have good drainage in order that the melt water, which contains bacteria from the fish, is allowed to drain away.

The life of any box is plainly dependent on factors of design, material, construction, degree of use and methods of handling but as a comparison aluminium and plastic boxes might last 6-10 years, respectively, whereas a wooden box might last only 2 years and require repair during that period. In the calculation of requirements of boxes over a period, pilfering and loss should be accounted for, and in this respect, plastic boxes offer an advantage over wooden or aluminium inasmuch as the material is of little use for alternative purposes.

The dimensions and capacity of a fish box should reflect factors of fish size, packing density of fish and ice. The length should be such that fish will not overhang the boxes or have to be forced into it and the depth should be such as to allow for fish and ice without crushing. The volumetric capacities and dimensions of different types of boxes are given in Table 16.

It is essential that after use fish boxes should be thoroughly cleaned of all dirt, fish slime and scales. The method of cleaning to be employed will depend on the number of boxes involved and whether or not the debris has been allowed to dry onto the box. With small numbers of boxes, a simple cold water wash may be all that Is required but if the debris has been allowed to dry on the boxes then an overnight soak In a detergent solution with scrubbing and hosing the following day will be necessary. Such treatment is not suitable for wooden boxes which cannot be satisfactorily cleaned. A mildly alkaline detergent should be used for wood, aluminium or plastic boxes at a concentration of 1-2 percent solution.



32

Stack/nest box

Stack only box

5.7 Reasons for non-adoption of ice containers and fish boxes

Insulated containers are required to extend storage period of ice and fish. It is unlikely that an ice distribution strategy can be successful without provision of suitable insulated containers to the end users of ice, in this case, BMUs and fishers. This study assessed various reasons that are likely to hinder non-adoption of insulated ice containers along the fish supply chain. Some of the reasons are;

i) Small size of fishing vessels which makes it unsafe to fit the insulated container.

ii) High costs of constructing/ purchasing insulated container.

iii) Inadequate skilled personnel to construct container



33

iv) Unavailability of some construction materials for the construction of the container e.g. Styrofoam, iron sheets etc.

v) No price premium/incentive for those who use insulated container with ice. IFPs indicated that they would pay more for quality fish, whether iced or not.

vi) Inadequate and /or irregular supply of ice.

6.0 Conclusion

This study has established the critical need of ice in Suba District. The current supply level is far short of demand and directly serves only IFPs. The study established the daily fish reject level of fish delivered to the landing beaches at between 7% and 25%, with a mean of about 14%. Of the fish delivered to IFP (which usually has been selected at the landing beaches) the mean reject level is about 2%. Interventions to supply fishers and fish traders with ice can have real positive benefits in reducing post-harvest losses and raising the bargaining position of fishers. Three sets of indicators to assess the impacts of ice have been identified, namely; for reduced post-harvest losses, for increased incomes from fish and for increased employment opportunities in ice and fisheries-related business. Baseline data has been obtained on all the indicators, therefore, monitoring will basically compare how (the direction and magnitude) the indicator variables have changed in relation to the baseline. The report has proposed a monitoring program that deals with the choice of impact indicators, method of monitoring, selection of sites and timing of monitoring.

7.0 References

Abila R.O. (2006) ‘More Ice on the Way, but is there Fish for It?’. Lake Victoria portal website (www.lakevictoria.info)

Department of Fisheries (2006). Draft Report on The Lake Victoria (Kenya) Fisheries Frame Survey 2006. LVFO/IFMP/ Ministry of Livestock and Fisheries.

Gram et al (1987)

Karuga S. and R. O. Abila (2006). ‘Commercial Ice Distribution Strategy and Key Programme Activities for Mbita Ice Plant Kenya Ltd’. Consultancy Report for USAID/Kenya Business Development Service/ ACDI-VOCA. Nairobi. May 2006.

Muhoozi L. I., Wekesa S. and Lyimo E. (2005). Regional Catch Assessment Survey Synthesis Report for June-September 2005. LVFO/IFMP. Jinja.



34

8.0 Annexes

Annex 1:

Fishing facilities in Suba District

2000 2002 2004 2006

No. of fishers 14,782 16,727 11,639 15,585

Landing sites 97 100 104 108

No. of fishing craft 4,051 3,267 3,575 4,910

Bandas (fish sheds) 19 24 12 32

Cold rooms working 1 - 1 -

Jetty 4 1 2 2

Fish stores 4 1 2 10

Beaches supplied with electricity - - 1 5

Beaches with toilet facilities - 49 62 73

Beaches with potable water - 5 8 3

All weather roads 12 36 23 30

No. of crafts using engine 308 335 503 811

Total no. of gillnets 86,464 37,527 28,797 157,609

No. of small seines 5,182 674 1,674 1,235



35

Annex 2: Questionnaires (To be filled by fishers/traders/agents/processors) 1) Date: 2)Name of Beach -----------------------------Type of beach: [1].Landing Beach [1].Collection beach [1].Urban market [1].Processing factory 3.) Name/position of Respondent: --------------------------- 4) What is the immediate source of your fish? [1].Fishing ground [1].Landing Beach [1].Collection beach 5) On average how long is fish stored at that source?-----------Hrs ------------Days 6) How long does it take to transport fish from the source of this place? ----- Hrs---- Days 7)On average what quantity of fish do you land/purchase?-------Kgs/boat 8) Indicate the quantity of each grade landed/purchased Grade A ---------------------Kgs Grade B ---------------------Kgs Grade C ---------------------Kgs 9) What price did you get for: Grade A quality --------------------- Ksh./Kg Grade B quality ---------------------Ksh/Kg Grade C quality ---------------------Ksh/Kgs 10. Do you use an insulated container? [1] Yes [1].No 11. Do you use ice for preservation of your fresh fish [1].Yes [1] No 12. How much do you pay for ice/kg?-----------------------------Ksh./Kg 13. How much are you willing to pay for ice/Kg? 10. How much are you willing to pay for ice/Kg? Fishers ----------------------- Traders -----------------------



36

Annex 3: Data monitoring sheet

Landing Beach ------------------------------Month:------------------------------- Quantity of fish landed (Kg) Price (Ksh/Kg) Day Date Total

quantity Grade A

Grade B

Grade C

Grade A

Grade B

Grade C

Total income

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

ASSESSING IMPACTS OF ICE ON FISH LANDING BEACHES OF SUBA DISTRICT, KENYA - ABILA RICHARD O.

Documents