The fisheries of Timor-Leste: A 4-year time series analysis covering the COVID-19 pandemic In partnership with Photo credit: Jennifer KingWorldFish The fisheries of Timor-Leste: A 4-year time series analysis covering the COVID-19 pandemic Authors Sarah Jørgensen Veillat,1 Alexander Tilley,2 Joctan Dos Reis Lopes,2 E. Fernando Cagua,2 Lorenzo Longobardi,2 Jeppe Kolding.1 Affiliation 1 University of Bergen, Norway 2 WorldFish Citation This publication should be cited as: Veillat SJ, Tilley A, Dos Reis Lopes J, Cagua EF, Longobardi L and Kolding J. 2022. The fisheries of Timor-Leste: A 4-year time series analysis covering the COVID-19 pandemic. Penang, Malaysia: WorldFish. Program Report: 2023-01. Acknowledgments This report was adapted from the master’s thesis of Sarah Jørgensen Veillat in partial fulfillment of a master’s degree of fisheries biology and management at the University of Bergen, Norway. This work was undertaken as part of the Fisheries Sector Support Program, phase two, funded by the Royal Norwegian Embassy in Jakarta, and as part of the CGIAR Initiative on Aquatic Foods. Funding for this work was provided by CGIAR Trust Fund donors and the Aquadata work package of the CGIAR Initiative on Aquatic Foods. This publication has not gone through the standard WorldFish science review procedure. The authors gratefully acknowledge the participation and insight of all Timorese fishers who have generously contributed their time, knowledge and information to improve understanding of the fisheries of Timor-Leste. We would also like to recognize the important contribution of the General Directorate of Fisheries and Marine Resources of the Government of Timor-Leste, which sees the value of data-driven decision-making and has invested in improved data systems to tackle management and sustainability issues in Timor-Leste. Contact WorldFish Communications and Marketing Department, Jalan Batu Maung, Batu Maung, 11960 Bayan Lepas, Penang, Malaysia. Email: worldfishcenter@cgiar.org Creative Commons License Content in this publication is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0), which permits non-commercial use, including reproduction, adaptation and distribution of the publication provided the original work is properly cited. © 2023 WorldFish. Photo credits Front cover, pages 13, 19. Jennifer KingWorldFish; pages 11, 13, 26, Alex Tilley/WorldFish; pages 11, 15, 26, Joctan Dos Reis Lopes/ WorldFish; pages 13, 19, 26, Jeppe Kolding; pages 6, 19, 26, Holly Holmes/WorldFish; page 26, Hampus Eriksson/WorldFish. Disclaimer The opinions expressed here belong to the authors, and do not necessarily reflect those of WorldFish or CGIAR. The designations employed and the presentation of material in this information product do not imply the expression of any opinion whatsoever on the part of WorldFish concerning the legal or development status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention of specific companies or products of manufacturers, whether or not these have been patented, does not imply that these have been endorsed or recommended by WorldFish in preference to others of a similar nature that are not mentioned. WorldFish encourages the use, reproduction and dissemination of material in this information product. Except where otherwise indicated, material may be copied, downloaded and printed for private study, research and teaching purposes, or for use in non-commercial products or services, provided that appropriate acknowledgment of WorldFish as the source and copyright holder is given and that WorldFish’s endorsement of users’ views, products or services is not implied in any way. WorldFish information products are available on the WorldFish website (www.worldfishcenter.org). i Table of contents List of abbreviations 1 Key terms, definitions and units 2 Executive summary 3 How to use this report 4 1. Introduction 5 2. Study area 7 3. Fishing fleet and gear 11 3.1. The sampled fishing fleet 11 3.2. Fishing gear 12 4. Effort and landings 15 4.1. Geospatial fishing effort 15 4.2. Number of trips per month 16 4.3. Trip effort 16 4.4. Catch per unit effort 18 4.5. Catch composition 19 4.6. Total monthly landings 22 4.7. Seasonality 24 4.8. Total monthly landings by municipality 24 5. Market 26 5.1. Fisheries catch usage 26 5.2. Market value of catch 29 6. COVID-19 33 7. Conclusions 38 7.1. Limitations 38 7.2. Further research 38 7.3. Conclusion 38 References 40 ii List of abbreviations CPUE catch per unit effort GDP gross domestic product ICT information and communication technology IRI index of relative importance MAF Ministry of Agriculture and Fisheries SIDS Small Island Developing States SSF small-scale fisheries 1 Key terms, definitions and units Catch: The total number (or weight) of fish caught by fishing operations. Catch should include all fish killed by the act of fishing, not just those landed. The component of fish encountering fishing gear, which is retained by the gear. Catch per unit effort (CPUE): A measure of the relative abundance of a fishstock, calculated using a standardized and constant unit of effort in method or time. For example, the quantity of fish (kg) caught per 100 hooks, or the quantity of fish caught per fisher per hour. CPUE data can be fisheries dependent or independent, but requires the fishing gear to be the same. CPUE is gear-specific because of different probabilities of capture (so-called catchability), so comparing CPUE only by time across different types of gear is not possible without intercalibration. Effort: The amount of fishing gear of a specific type used on the fishing grounds over a given unit of time (e.g. hours trawled per day, number of hooks set per day, or number of hauls of a beach seine per day) to be used to compare catch rates in the same fishery. When two or more kinds of gear are used, the respective efforts must be adjusted to some standard type before being added, sometimes referred to as the effective fishing effort. Across different fisheries or gear types, a standardized unit of effort must be used, such as time fishing per vessel or person (e.g. fisher days). Fishing trip: Defined as an event when a fisher or a fishing vessel leaves the home site or port with the intention of fishing, travels to the fishing grounds or gleaning area, fishes for a certain time and returns to the site or port where its catch is landed, sorted or unloaded. Gear: A tool or method used to catch or gather aquatic foods, such as hook-and-line, trawl net, gillnet, pot, trap, spear, manual collection, etc. PDS trackers: GPS trackers manufactured by Pelagic Data Systems Inc. in San Francisco, California. PeskAAS: The national fisheries monitoring system of Timor-Leste. The name is a pseudo acronym, derived from the Tetum word for fisheries (peskas) + Automated Analytics System. Trip length: The duration of a fishing or gleaning trip (with or without a vessel), measured in time (normally days or hours) between departure and return time and date. Source: Fisheries and aquaculture ontology doi: 10.5281/zenodo.7381034 2 Executive summary Small-scale fisheries (SSF) represent almost half of global capture fisheries production, supporting the livelihoods of millions of people worldwide and the food and nutrition security of billions. However, the general lack of data from these fisheries makes it difficult to ensure they are included fairly in decisions about ocean resources and how to manage resources for sustainability. Timor-Leste has one of the highest rates of malnutrition in the world, and fisheries are an important potential growth sector to achieve national development objectives of alleviating poverty and eradicating malnutrition. This report uses a 3-year time series of catch and effort data from PeskAAS, the national fisheries monitoring system, to provide a baseline characterization trajectory of the country’s fisheries. The national fisheries fleet of Timor-Leste is exclusively small-scale, made up of approximately 4,554 smaller fishing vessels composed of about 60% motorboats and 40% canoes. Canoes are mostly single-person, while motorboats typically hold between one and 12 fishers depending on location and fishing type. The most common type of gear is gill nets. Fishing is predominantly nearshore, within 5 km of the coast, and focused on fringing coral reef habitats and the pelagic forereef zone. The presence of large saltwater crocodiles throughout the coastline of mainland Timor-Leste restricts most spearfishing activities to the island of Atauro. Seasonal changes and mixed livelihoods such as smallholder agriculture drive seasonal differences in fishing effort, with September to December showing the highest effort and February to June the least. Timor-Leste has an estimated average annual catch of 6781 t (SD ± 1157), with Atauro, Lautem, Bobonaro and Manatuto being the most productive municipalities. The catch composition in each area reflects the types of gear used and consists predominantly of small pelagic fish species (<30 cm) caught in gill nets. Atauro shows the most diverse catches, which frequently include small and large reef fish species. Large fish species (e.g. sailfish, stingray, barracuda, long tom) are sold for a higher price per kilogram than smaller species (on average 11 and 8 USD, respectively), with fishers retaining a high percentage of catches for household consumption. Of all the municipalities, Dili-based fishers sell their fish for the highest price because of their proximity to the capital and ability to save on transportation and trading. The time series of data used encompasses the COVID-19 pandemic, and our findings suggest that SSF in Timor-Leste were fairly resilient to the impact. Although they experienced a decline in the market value of the catch, it does not appear to have been as influential as in other countries, where decreases in fishing effort were seen. It is possible that the limited export of aquatic foods from Timor-Leste, and its modest tourism sector, sheltered the fisheries from shocks seen in the fishing sector of other nations. This reflects the persistence of SSF to generate food and income where formal markets and global supply chains are extremely limited. With increasing attention and investment in the country’s fisheries sector to achieve national development goals, it will be critical to consider if modernization and a focus on increasing production will benefit those with the most pressing food and nutritional needs, as well as environmental sustainability. This report provides a benchmark against which to track the progress and development of the fisheries sector in Timor-Leste, and serves as a case study for low-cost digital monitoring of fisheries in data-deficient scenarios. 3 How to use this report This report presents an analysis of fisheries catch and effort data generated through the PeskAAS fisheries monitoring system between 2018 and 2021. Due to the iterative co-design and co-development of PeskAAS with local stakeholders, monitoring started as a pilot project at five sites in 2017 and gradually increased to the national system it is today. Along this journey, the type and quantity of data collected from fishers changed slightly, and continuous vessel tracking was introduced with 100 boats in February 2018, then 400 boats later that year. Currently, PeskAAS enumerators record the catches from a sample of fishing trips at 15 landings sites across the country on a daily basis, and approximately 200 motorised and unmotorised vessels send continual geolocation data. However, to integrate older data from fewer sites, where appropriate, the data are grouped into broad geographical categories of North coast, South coast and Atauro island. In using this report, please keep in mind that the time series of data used may alter slightly according to the analysis being performed due to this iterative development of PeskAAS. For example, there was no accurate fish price information recorded prior to April 2019, or trip duration information recorded prior to June 2018. The precise time period used for each analysis is reported in the appropriate figure caption. The latest data and near real-time analytics can be viewed by visiting the publicly available dashboard at www.peskas.org. Data protection and permissions All data is provided voluntarily, and collection is conducted with the permission of fishers. They have the option to not answer or to provide further data, or they can request that the data we hold for their fishing activities be deleted. 4 1. Introduction SSF are an important part of global fish production, contributing at least 40% of global capture fisheries (FAO et al. 2022). Of the several million people engaged in SSF, over 90% live in developing countries and Small Island Developing States (SIDS), for whom fish is an essential source of food and income (Mills et al. 2011; FAO et al. 2012). In some high-income countries, where fisheries have been intensively monitored with quantitative data systems and governed by the state, monitoring has in general led to effective adjustments in management and fishing capacity so that fishstocks are recovering and/or fished at sustainable levels (Hilborn et al. 2020). The majority of the global marine SSF fleet is active in low- and middle-income countries, with 80% in Asia (FAO 2020). However, they are under increasing pressure to conform to data-driven management strategies (Béné and Friend 2011; Islam 2011; Kolding and Van Zwieten 2011; Kolding et al. 2014; Macusi et al. 2020). Fisher knowledge is critical in framing and evaluating fisheries-dependent data in a social-ecological context, but it is still rarely incorporated effectively into governance (Hind 2014). Managing fisheries is hard: it’s like managing a forest, in which the trees are invisible and keep moving around. – John Shephard Monitoring and assessing fishstocks present a significant challenge for quantitative data collecting methods, so management relies principally on fisheries-dependent data, specifically data collected from fishing activities and landings, etc. Nevertheless, the dispersed and diverse operations of fishing activities and the challenges inherent in measuring and understanding trends in mobile stocks place demands and limitations on fisheries-dependent monitoring. This is especially true for SSF, which are often informal, dynamic and even more widely dispersed. As such, scientific management strategies are often based on assumptions under limited information, which can do more harm than good if there is also limited integration of local knowledge or a lack of understanding of social and cultural elements (Kolding and Van Zwieten 2011). Filling these gaps in our knowledge should be a priority to determine the level of resilience of SSF against crises and, if necessary, strengthen them. SSF communities face formidable economic, social and geographical barriers in using technological approaches, such as the cost of equipment and the digital literacy to operate it. Yet, there has been a dramatic expansion of the development and testing of information and communication technologies (ICTs) for SSF (Fujita et al. 2018; Bradley et al. 2019; FAO and WorldFish 2020). One such ICT is PeskAAS, which was developed by WorldFish in partnership with the Timor-Leste Ministry of Agriculture and Fisheries (MAF) and has been scaled since 2017 (Box 1). Box 1. The PeskAAS fisheries monitoring system PeskAAS was co-designed with fisheries officers as a simple and cost-effective way to generate and analyze fish landings data in near real-time (Tilley et al. 2020). It is an interactive web-hosted application in which fisheries data trends are visualized with the goal of exploring data and improving decision-making processes for SSF managers. In 2019, the government formally adopted PeskAAS as the official national fisheries monitoring system of Timor-Leste. It now forms part of the fisheries management process and has improved and automated rapid, low-cost and accurate data generation for fisheries in Timor-Leste (Dam Lam 2022). For more information and to view the dashboard, please visit www.peskas.org. 5 This report aims to summarize and analyze the first 3 years of data collected by PeskAAS to provide a series of benchmarks for the fisheries sector, against which trends and trajectories can be compared in years to come. Furthermore, this report is intended to provide information for fisheries stakeholders, including fishers, government, nongovernmental organizations and civil society organizations to guide new development and capacity building initiatives in the fisheries sector. Although the impacts of the COVID-19 pandemic on fisheries are not a major focus of this report, the pandemic started during this data time series. This is thus used as a reference point in certain analyses, and a short analysis and discussion related to COVID-19 is included. A fisherman carries his net along the shore in Atauro, Timor-Leste. 6 Photo credit: Holly Holmes/WorldFish 2. Study area Timor-Leste, also known as East Timor, is one which is expected to triple by 2050 (Molyneux et of the SIDS in Southeast Asia. It is surrounded al. 2012). Most cities, including the capital, Dili, are by Indonesia to the west, north and east, and located on the north coast, where the majority of Australia to the south, with the Timor Sea in the people live: over 60% of the total population between (Figure 1). It has a coastline of 783 km lives on the north coast and about 20% on the and a total land area of approximately 14,954 km2 south coast, with the remainder living inland (Lopes et al. 2019). The island of Atauro, off the (General Directorate of Statistics 2015). Most north coast, with a land area of approximately 140 economic activity is centered around Dili, so the km2, has a population of about 10,000 (General rural population experiences higher poverty rates Directorate of Statistics 2015; Lopes et al. 2019). than people living in urban areas (Moxham and The country counts around 1.3 million citizens, Carapic 2013). Figure 1. Map of Timor-Leste and its surroundings. 7 Note: The colors show the elevation from 0 m above sea-level (green) to 2000+ m above sea-level (orange brown). The size of the points shows the size of the cities. The small map on the right side shows the placement of the country compared to the rest of Southeast Asia and Australia. Source: Geoatlas 2018. Figure 1. Map of Timor-Leste and its surroundings. 6 7 Timor-Leste has a tropical climate, with temperatures at the coast ranging from 25°C to 30°C year-round, with pronounced wet (December to May) and dry (June to November) seasons (Figure 2) (Timor-Leste National Directorate of Meteorology and Geophysics et al. 2015). This seasonality pattern of rainfall experiences between- year fluctuations, caused by the El Niño-Southern Oscillation and the Indian Ocean Dipole (Australian Bureau of Meteorology and CSIRO 2014). The mountain range separating the north coast from the south coast (Figure 1) causes different rainfall patterns on the two coasts (ADB 2014). The seas on the south coast are rougher than on the north coast because of cyclones coming from Note: Seasonal rainfall (gray bars) and sea surface temperature the Indian Ocean. The south coast has a shallow (green squares) in Timor-Leste, measured in Dili. continental slope, while the north coast has a very Source: Timor-Leste National Directorate of Meteorology and steep slope, quickly descending several kilometers Geophysics et al. 2015. down (Figure 3). The coral reefs are predominantly Figure 2. Seasonal rainfall and sea surface fringing reefs, located for the most part on the temperature in Timor-Leste. north coast and around Atauro (ADB 2014). Note: The light to dark blue scale shows the bathymetry from sea level down to 1000+ m deep. The maximum depth of the Timor Sea is said to exceed 5000 m. Figure 3. Map of Timor-Leste and its coastal bathymetry. 8 The official languages of Timor-Leste are levels of other SIDS in the Pacific, such as 30–118 Portuguese and Tetum, but there are many other kg in Melanesia, 62–115 kg in Micronesia and indigenous languages in use (Taylor-Leech 2009). 50–146 kg in Polynesia (Mills et al. 2013; FAO 2014; The country was only recently declared a sovereign López Angarita 2019). Agriculture is responsible state by the UN in May 2002, after having suffered for most of the food production, engaging 90% two brutal wars in 1979 and 1999 and a 24-year- of rural households (FAO 2015; Ximenes 2018). long occupation by Indonesia (Braithwaite et al. Most families involved in fishing have diversified 2012; Ingram et al. 2015). These calamities have livelihoods, with most of their time and energy left the country with high poverty rates and poor being spent on agriculture and livestock farming health services and infrastructure, and 36% of the (Mills et al. 2013). These livelihoods are, however, population experiences chronic food insecurity dynamic and vary according to season and climate because of poor quality and low quantity food (Mills et al. 2017). (Grebmer et al, 2019; National Directorate of Food Security and Cooperation 2019; CARE 2020). For the purpose of this study, the coastal During the Indonesian occupation, the commercial municipalities were grouped into three statistical fishery played a significant role in the economy areas: Atauro Island, the north coast and the south of Timor-Leste. But after the civil war in 1999, this coast (Figure 4). The landing sites of Manatuto and plummeted as the infrastructure of the country was Lautém are located on the north coast, so they destroyed (Barbosa and Booth 2009). Hereafter, SSF have been assigned to the north coast region even became the dominant fishery in Timor-Leste. though the municipalities extend between both coasts. The municipalities Ermera and Aileu do not Currently, annual per capita fish consumption have a coastline and are thus not included. These among the Timorese population is just 17 kg in areas were chosen because of their difference in coastal communities and 5.2 kg in inland areas. rainfall pattern, wave activity, marine landscape This is particularly low when compared to the and demographics mentioned above. Note: The dots represent the landing sites where catch and trip information were recorded, though not all landing sites are shown on this map, only the ones where enumerators are presently active. The colors represent the different regions: Atauro Island (gray), north coast (red) and south coast (blue). Source: Tilley et al. 2020. Figure 4. A map of Timor-Leste and its municipalities. 9 Box 2. Summary statistics of recorded fishing trips. The dataset used in this study consists of a total of 123,181 recorded fishing trips identified on the basis of 72,328 GPS tracks and 56,629 direct interviews with fishers at the landing site recording catch data. Of these fishing trips, 4.68% (5776) contain both GPS tracks and catch data. Atauro is the most represented municipality, approaching 30,000 recorded fishing trips, of which 62.9% were interviews recording catch data (Figure 5). All others counted fewer than 15,000 observations. Ainaro is the least represented municipality, counting 822 fishing trips, of which 88.3% were interviews recording catch data (Figure 5). Note: The percentage on the bars represents the proportion of recorded fishing trips containing catch data coming from direct interviews. Figure 5. Number of fishing trips recorded for each municipality. 10 3. Fishing fleet and gear 3.1. The sampled fishing fleet invest in motorboats. By contrast, the south coast is characterized by predominantly paddle canoe Of the 2277 fishing boats registered in 2018 fishing, which is likely a result of the isolation, poor by MAF, 323 were used to report catch and/ road infrastructure and higher poverty rates of or tracking data to PeskAAS (Table 1). The 323 that part of the country (Braithwaite et al. 2012; sampled boats include both motorboats, with Moxham and Carapic 2013; Ingram et al. 2015). either an inboard or outboard motor, and paddle canoes (Table 2; Plates 1 and 2). The capital required to invest in a motorboat is out of reach for most rural dwellers. Wooden The differences seen in the characteristics of the boats can be made from locally available natural sampled fishing boats between the areas are resources, while fiberglass boats are now being likely influenced by social, geographic, economic made in Timor-Leste. The latter are the focus of and political factors. The north coast is where Dili government funding schemes for fishers, but the and most of the economic activity is centered, so entry and maintenance costs for rural fishers is still as the richer region more people appear able to prohibitively high. Region Sampled boats Registered fishing boats Boats sampled Atauro Island 89 348 25.6% North coast 129 1463 8.8% South coast 105 466 22.5% Total 323 2277 14.2% Note: Information on the number of registered fishing boats was obtained from the MAF (unpublished data). Table 1. Total number of sampled and registered fishing boats per region. Plate 1. Paddle-driven canoes in Timor-Leste. Plate 2. A motorboat in Timor-Leste. 11 Photo credit: Alex Tilley/WorldFish Photo credit: Joctan Dos Reis Lopes/WorldFish Region Sampled Boat type Number Median boat Boat material Motor type boats of boats length (m) (fiberglass, wood, (inboard, outboard, unknown) unknown) Atauro 89 Motorboats 48 (54%) 8.0 3 (6.3%), 2 (4.2%), 24 (50.0%), 46 (95.8%), 0 (0%) 21 (43.7%) Paddle canoes 41 (46%) 5.0 1 (2.4%), NA 31 (75.6%), 9 (22.0% North coast 129 Motorboats 103 (80%) 6.0 34 (33.0%), 35 (34.0%), 55 (53.4%), 57 (55.3%), 14 (13.6%) 11 (10.7%) Paddle canoes 26 (20%) 4.5 0 (0%), NA 21 (80.8%), 5 (19.2%) South coast 105 Motorboats 30 (29%) 5.0 4 (13.3%), 12 (40.0%), 20 (66.7%), 14 (46.7%), 6 (20%) 4 (13.3%) Paddle canoes 75 (71%) 4.0 0 (0%), NA 65 (86.7%), 10 (13.3%) Total 323 Motorboats 181 (56%) 6.5 41 (22.3%), 49 (26.6%), 99 (53.8%), 117 (63.6%), 44 (23.9%) 18 (9.8%) Paddle canoes 142 (44%) 4.0 1 (0.7%), NA 117 (82.4%), 24 (16.9%) Note: Information on the number of registered fishing boats was obtained from the MAF (unpublished data). Table 2. Vessels reporting catch and/or tracking data to PeskAAS. 3.2. Fishing gear Where %Wi equals the percentage weight and %Ni the percentage number of each species i The different types of fishing gear used on of the total catch, %Fi equals the percentage Timor-Leste are gill nets, long lines, hand lines, frequency of occurrence of each species in the spear guns, seine nets, cast nets and fishing traps, total number of fishing trips, and S represents as well as manual collection (Figure 6). the total number of different species. This index can either be shown as a percentage (%IRI) or To compare the different types of gear used, a as a rectangle, where the percentage number percentage index of relative importance (%IRI) of fish (%N), the percentage weight (%W) and was used. IRI is originally used to calculate the the percentage of the frequency of occurrence relative importance of a species within a fishery, (%F) are separately displayed relative to all other by taking into consideration the weight and species. Although the original function of the abundance (numbers) of fish species caught and equation is to calculate the index of importance the frequency of occurrence all at once (Kolding of the species caught in a fishery, here it is also 1989). The equation expressed as a percentage applied to the relative gear use in the fishery. reads as follows: %𝐼𝐼𝐼𝐼𝐼𝐼𝑖𝑖 = (%𝑊𝑊𝑖𝑖+%𝑁𝑁𝑖𝑖)∗%𝐹𝐹𝑖𝑖 ∑𝑆𝑆 ∗ 100 [1] 𝑗𝑗=1(%𝑊𝑊𝑗𝑗+𝑁𝑁𝑗𝑗)∗%𝐹𝐹𝑗𝑗 12 Gill nets are the most used type of gear by number total catch in weight and 13% of the total catch in of trips on both the north coast and Atauro (Figure numbers are caught by hand lines. 6), representing the highest proportion of both catch in weight and number of fish caught. On the The differences in the constellation of the fishing north coast, gill nets make up 92% of the catch in gear used in the different areas are influenced weight and 98% of caught fish. The percentages by social, economic, cultural and geographical are lower on Atauro, but gill nets still make up 62% factors, as well as climate and natural factors. Atauro of the catch in weight and 76% number of fish and the north coast are ecologically similar, with caught. The other types of gear used on Atauro are thin, fringing reefs close to the coast and a steep spearguns, seine nets and hand lines, collectively forereef slope descending to more than 4 km deep. yielding about 34% of the total weight of the catch However, open ocean fishing gear is rarely found and 23% of the total catch in numbers. Hand lines on Atauro, as fishers prefer reef and nearshore are used somewhat frequently, for 14% of total pelagic fishing grounds (Mills et al. 2013). The choice fishing trips, but only result in a small proportion of of fishing boats and seasonal weather conditions the yield. also limit gear choice. The south coast experiences rough sea conditions for half of the year, so its On the south coast, long lines are the most fishing frequency is affected, as well as its range frequently used gear by number of trips, making and gear used in paddle canoes. Atauro is the only up 76% of the total catch in weight and 60% of area where spearfishing is common, as throughout the total catch in numbers. Gill nets still make up the mainland coastline there are large and lethal 18% of the total catch in weight and 27% of the saltwater crocodiles that regularly attack and kill total catch in numbers, while only 5.6% of the fishers and coastal dwellers (Brackhane et al. 2018). Plate 3. Gill nets used in Timor-Leste. Plate 4. Hand lines used in Timor-Leste. Plate 5. Spearfishing in Timor-Leste. Plate 6. Fish traps used in Timor-Leste. 13 Photo credit: Alex Tilley/WorldFish Photo credit: Jeppe Kolding Photo credit: Jennifer King/WorldFish Photo credit: Alex Tilley/WorldFish Note: The IRI considers the relative weight and number of fish caught per gear type, and the relative frequency of occurrence (%FRQ) of a gear type. Figure 6. Gear composition using the %IRI of the landings (April 2018 to December 2021). 14 4. Effort and landings 4.1. Geospatial fishing effort for fishing. The same is probably the case for the activity seen in the western part of the north coast, Solar-powered GPS units were installed on part of where they are headed to Indonesia or to trade the Timor-Leste national fishing fleet to track the with Indonesian fishers at sea. number of trips and to visualize geospatial fishing effort (Plate 7). These GPS units are produced by Pelagic Data Systems Inc. in San Francisco, California, and are hereafter referred to as PDS trackers. During a fishing trip, these trackers automatically record the location every 5 seconds. Most of the fishing trips seem to be concentrated inshore around the coast and around the landing sites (Figure 7). As most marine productivity is in the surf zone, this is where a high concentration of fish can be found (Alongi et al. 2012). It also demands less labor and time and requires less fuel. There is some activity recorded between the island of Atauro and the mainland as well as to the Plate 7. Solar-powered GPS unit Indonesian island of Wetar, north of Atauro. These (PDS-tracker) being installed on a fishing boat. are most likely taxiing and trading trips, and not Note: Data from 80,163 trips using the tracking information sampled by the PDS trackers from May 2019 to August 2022. Figure 7. Heat map of Timor-Leste visualizing the geospatial fishing effort. 15 Photo credit: Joctan Dos Reis Lopez/WorldFish 4.2. Number of trips per month 4.3. Trip effort The total number of trips per month was To calculate an effort unit (the trip effort), the calculated using the trips sampled by the PDS duration of a trip (hours) was multiplied with the trackers on 454 vessels. Only a small percentage number of fishers (including women and children), of the Timor-Leste fishing fleet is being resulting in an effort unit of fisher-hours. tracked. As such, the monthly number of trips is estimated by averaging the number of trips The average trip effort (fisher-hours) of the three a boat makes over a month and multiplying regions appears to be quite similar (Figure 9a). The by the number of boats for each municipality, mean trip effort mostly ranges between 5 and 15 assuming that all boats are equally active. fisher hours, with a peak reaching up to 30 fisher hours per month in the south coast. There seems There are considerably more fishing trips taken to be no clear trends in trip effort over time among per month on the north coast than on Atauro and the regions, except for a slight decline in the north the south coast (Figure 8). This is a direct reflection coast (Figure 9b). of that area having the highest number of fishing boats registered and being where the majority of the country’s population live (General Directorate of Statistics 2015). Note: The gray shaded area defines the pandemic period starting from the COVID-19 lockdown on March 28, 2020. Figure 8. Estimated total number of trips per month (January 2018 to December 2021). 16 Note: The regions are plotted (a) together in one plot and (b) separately, where the y-axis has free scaling. The colored shaded areas represent the 95% confidence interval. The gray shaded area defines the pandemic period starting from the COVID-19 lockdown on March 28, 2020. No trip duration data was available prior to June 2018. Figure 9. The mean monthly effort (fishers * hours) of a fishing trip (June 2018 to December 2021). 17 4.4. Catch per unit effort For this report, a catch per unit effort (CPUE) was similar, though the north coast has more high used, where the effort was standardized to one peaks (Figure 10). The south coast appears to fisher and one trip hour (fisher-hour). The catch generally have a lower CPUE than Atauro and was calculated in kilograms. After 2019, the weight the north coast. The CPUE of Atauro seems calculation was done on species level, whereas to increase up until the COVID-19 lockdown before 2019 it was more general. starts, after which it stays constant. The north shows no clear change of CPUE over time, The average CPUE per month (kg/fisher-hour) whereas the CPUE of the south coast shows a for Atauro and the north coast are quite slight increase after the start of the lockdown. Note: The regions are plotted (a) together in one plot and (b) separately, where the y axis has free scaling. The shaded areas around the time series represent the 95% confidence interval. The lines are the local polynomial regression. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. No trip duration data was available prior to June 2018. Figure 10. The mean catch per unit effort (June 2018 to December 2021). 18 4.5. Catch composition Looking at the catch composition of the four most used types of gear, it shows that gill nets catch The fishery of Timor-Leste has a diverse catch primarily small pelagic species, both in weight composition (Plate 8) (Tilley et al. 2019). At least and number (Figure 12). The weight of the catch 53 different species were caught and identified, of hand lines is dominated by large pelagics excluding species classified under “unknown” and and large reef species, whereas small pelagics “other.” To handle the data and keep the results dominate the catch in numbers. When long lines easier to follow, the species were aggregated into are used, pelagic fish are primarily caught, of 10 functional groups: small pelagics, large pelagics, which large fish account for most of the weight of small reef species, large reef species, sharks and the catch and small fish for most of the catch in rays, crustaceans, shrimps, cephalopods, mollusks numbers. The catch of spear guns consists largely and unknown. of reef species. Again, the large fish account for most of the weight of the catch and small fish To compare the difference in functional groups for most of the catch in numbers. Cephalopods caught between the areas, the %IRI was used are caught in almost one-fifth of the fishing trips (see equation 1). where spear guns are used. Atauro shows the most even catch composition The choice in gear type greatly influenced by functional groups of the three sampling the catch composition of each area. The high regions (Figure 11). It is dominated by small frequency of occurrence of reef fish in the catch pelagics in weight and numbers, while small reef of Atauro reflects the scarcity of pelagic fishing species are the most frequently caught. The catch gear on the island, as well as the use of spear guns. composition of the north and south coasts are Larger fish are caught more on both the south both dominated by small pelagic species, though coast and Atauro than the north coast, which is large pelagics are also caught relatively frequently likely because fishers there use long lines, hand and make up 13% of the weight of the catch in the lines and spear guns more often. north and 9% in the south. Plate 8. Fish caught in Timor-Leste, including parrot fish, snappers, surgeonfish, breams, mackerel scad, barracuda and sardines, among others. 19 Photo credit: Holly Holmes, Jennifer King/WorldFish and Jeppe Kolding 4.5.1. Catch composition by region This section presents information on the composition of fisheries catches across all gear types in each of the three main regions. Note: A total of 24,064 fishing trips for Atauro, 27,574 for the north coast and 12,166 for the south coast. The IRI considers the weight, the number and the frequency of occurrence (%FRQ) of a functional group. Figure 11. Catch composition using the %IRI of the landings (January 2018 to December 2021). 20 4.5.2. Catch composition by gear type This section presents information on the composition of catches according to each gear type. Note: The IRI considers the weight, the number and the frequency of occurrence (%FRQ) of a functional group. Figure 12. Catch composition using the %IRI of the landings by gear type (January 2018 to December 2021). 21 4.5.3. Length composition of functional groups The small pelagics, reef species, and the cephalopods caught on Atauro appear to be marginally bigger than on the north and south coasts, whereas the large reef species caught on the south coast seem slightly bigger than on Atauro and the north coast (Figure 13). Note: Violin plots of the length (cm) of the catch for every functional group. The y-axis has free scaling for every functional group. The dot represents the median and the variance bars the first and third quartile. Figure 13. Length composition of the functional groups. 4.6. Total monthly landings remains constant within each area throughout the The monthly total catch was estimated by sampling period. multiplying the average catch per trip by the average number of trips per boat and by the The estimated total monthly landings (t) on the number of boats registered in each municipality. north coast are generally higher and with more There is no data available on the number of peaks than on Atauro and the south coast (Figure registered vessels per year, so estimates are based 14a). These peaks can also be seen in north coast on an assumption that the total number of boats CPUE trends and could illustrate windfall catches 22 of sardines, which are seasonal and the primary of fish aggregating devices could also play a role in species of small pelagics caught on the north the fluctuations, as fishing near one increases the coast. These fish often migrate in large schools, chances of encountering big schools of pelagic which can result in big catch rates if one comes fishes (Tilley et al. 2019). across such a school (Hunnam et al. 2021). This assumption is strengthened by high catches Landings appear to increase over time on Atauro occurring during the wet season, when sardines up until the COVID-19 lockdown, which reflects are mostly present (Hunnam et al. 2021). The use the trend seen in the CPUE also (Figure 14b). Note: The regions are plotted (a) together in one plot and (b) separately, where the y axis has free scaling. Bold lines and shaded areas represent the local polynomial regression with 95% confidence interval. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. Figure 14. Estimate of the total monthly catch (January 2018 to December 2021). 23 4.7. Seasonality A pattern of seasonality is observed for the total Instead, the vast majority stated that crops and number of trips taken per month as well as for the livestock were their main economic activity (Mills total monthly landings in all three regions (Figures et al. 2013). It would therefore be reasonable to 15 and 16). The fishing activity decreases around assume that the seasonality of agricultural activities April and increases again in August, reaching a has an influence on fishing activity. The harvests peak in December to February. for the two primary crops, maize and rice, happens from February to April and April to July (FAO 2021). This pattern of seasonality is likely a result of The period with the highest food insecurity is in the weather conditions, as well as the diverse the months before the harvests, which is also when livelihoods of most Timorese households. The we see the highest fishing activity on the mainland fishers on the south coast make fewer trips when (Gorton 2018). On Atauro, however, the fishing the seas are at their roughest, from April to August/ culture is stronger, and up to half of fishers consider September (Tomascik 1997). The seasonality in fishing to be their main livelihood (Mills et al. 2017). fishing activity seen on Atauro is in line with the findings of Mills et al. (2017), who found that 4.8. Total monthly landings by municipality there was reduced fishing activity in July and the months around August because of the rough The municipalities with relatively high fish seas caused by the southeasterly monsoon. production are Atauro, Baucau, Lautém, Bobonaro and Manatuto (Figure 17). The municipalities with Fishing is for many households not their only or relatively low fish production are Ainaro, Covalima, even primary source of income (Mills et al. 2013). Dili (without Atauro) and Manufahi. Note: Box plots of the estimated number of trips per month (2018–2021). The y-axis has a free scale for each region. The shaded areas represent the 95% confidence interval, the lines local polynomial regression fitted to the data. Figure 15. Seasonality of trips per month. 24 Note: Box plots of the estimated total catch (t) for every month (2018–2021). The y-axis has a free scale for each region. The shaded areas represent the 95% confidence interval, the lines local polynomial regression fitted to the data. Figure 16. Seasonality of total monthly catch. Note: Bold lines and shaded areas represent the local polynomial regression with 95% confidence interval. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. Figure 17. Estimate of the total monthly catch for the 12 municipalities (January 2018 to December 2021). 25 5. Market 5.1. Fisheries catch usage to traders is the safest option, though it is less profitable than if sold directly to consumers. When interpreting the data of the catch and Fishers will try to sell as much of their catch as the anticipated market value, it is important to they can and keep what is left for household keep in mind that the information is acquired consumption or for sharing and bartering. before the first point of sale. This means the fisher is stating the price that they intend to According to statements from fishers, on average sell it for based on experience and current 94.2% of the sampled catch on Atauro is sold at (informal) market price information. It is a the market, 5.5% is kept for own consumption and prediction made by the fisher and does not 0.3% is deemed to be partially sold and partially necessarily align with actual values received. An kept. On the north coast, 53% goes to the market, assumption we make is that a fisher will gradually 2.7% is used for own consumption and 44.3% is adjust their price prediction based on recent both. On the south coast, 90.6% is sold on the sales, and it will not be purely aspirational. market, 1.9% is kept for own consumption and Fish catch is usually sold on the nearest main 7.5% is both. In the long term, market sales have road to the landing site, on the water (e.g. to shown a slight increase in all the regions, especially Indonesian fishers/traders) or to a local trader, Atauro, whereas own consumption trends were who then transports and sells the fish at the quite stable in all the regions (Figure 18). market (Plate 9) (Población 2013). Selling catch a c e b d Plate 9. Fishers and traders selling their fish directly from the boat (a), on the road (b), at the market (c), on the beach (d) and by motorbike (e). 26 Photo credit: Hampus Eriksson/WorldFish (a), Alex Tilley/WorldFish (b) (e), Holly Holmes/WorldFish (c), Jeppe Kolding (d) Note: The shaded areas around the time series represent the 95% confidence interval. The lines are the local polynomial regression. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The y-axis has free scaling. Figure 18. Destination of catch over time in mean weight per week of the sampled catch (January 2018 to December 2021). 27 An interesting difference between the regions therefore can be more precise in their answers. It is that a large part of the catch on the north could also be because of a difference in the nuance coast is documented as being partially used for of the question asked by the enumerators. own consumption as well as partially being sold. Although this is likely also the case on Atauro and the Most of the fish being kept for own consumption south coast, it is not being documented in the same are smaller fish of large species and small fish manner. Why there is this difference in documenting species, with some exceptions, such as the cobia between the regions is not clear. It could be that (Rachycentron canadum) and the moray eel fishers on the north coast are more realistic with the (Gymnothorax spp.) (Figures 19 and 20). Large fish outcome of their sales, or that fishers from Atauro and fish species are often easier to sell, as they and the south coast have a better understanding are in demand by the restaurants and the urban of which fish will sell and which will not and population (Venugopa and Shahidi 1995). Note: The y-axis shows individual species grouped by functional group, and the x-axis shows the percentage of the catch in weight (kg). Figure 19. Bar plot of the destination of the catch per species. 28 Note: (a) Percentage of the catch and (b) weight (kg) of the catch versus the size (cm) of the fish caught. The catch is divided into 12 length groups. Figure 20. Bar plot of the destination of the catch per length groups. 5.2. Market value of catch The catch from hand line fishing appears to yield the highest price (USD) per kilogram of To analyze the market value of the catch, we the different gear types, followed by long lines looked at the subset of catches that were stated by (Figure 22). Gill nets and spear guns generate the fisher as being “for sale.” the lowest price per kilogram. This corroborates the price-size relationship, given that most large In the PeskAAS dataset, the market value of the pelagic and reef species will be caught on hooks. total catch is estimated by fishers, not the value per species. So to be able to estimate the market The high value of large fish is interesting, value per functional group, the data was filtered to because in poor, rural areas smaller fish are only include trips where the catch consisted of a often more sought-after for various reasons, but single fish species (77.23% of total trips). predominantly because they can be bought for less money and are less wasteful when cold Size appeared to drive price to a certain extent, as storage is limited or unavailable. When very large large pelagic and reef species yielded the highest fish are landed, the fisher or trader might cut up price (USD) per kilogram of the different functional the fish into smaller pieces to increase the chance groups (Figure 21), followed by small pelagics and of selling it. The high value of large fish stated reef species. The lowest priced items were sharks by fishers could be a reflection of this, and not and cephalopods. A consistent positive trend in necessarily imply that large fish are purchased price per kilogram is evident in the north coast, for whole, in single transactions. both the reef species and the large pelagics. 29 Note: The shaded areas around the time series represent the 95% confidence interval. The lines are the local polynomial regression. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The y-axis has free scaling for every functional group. Figure 21. Mean market value (USD) per kilogram of the catch of the most frequently caught functional groups (January 2019 to December 2021). 30 Note: The shaded areas around the time series represent the 95% confidence interval. The lines are the local polynomial regression. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The y-axis has free scaling for every gear type. No available data prior to April 2019. Figure 22. Mean market value (USD) per kilogram of the catch by gear type (April 2019 to December 2021). The low and steady price structure of the Atauro Before COVID-19, the price per kilogram of catch market (Figure 24) likely reflects one of three was slightly higher on the south coast, followed possibilities: (1) the lack of market information by the north coast and lastly Atauro (Figure 24). available to fishers, meaning they do not know This could possibly be because the south coast is how much their fish are worth elsewhere, (2) home to larger fish species, which yield a higher higher relative competition between fishers price per kilogram. However, this is still somewhat on Atauro compared to mainland coastal counterintuitive. The south coast has poorer communities because of the higher density infrastructure and is located farther from the places of fishers and fewer markets, and (3) lower with high demand, such as Dili, so the traders likely purchasing power by local consumers, as people pay less as they need to compensate for the costs on Atauro are generally more isolated from paid of transporting and storing the produce (Población employment opportunities and so have a lower 2013). This is also the case for Atauro. While fish cash flow relative to the mainland. from the island have a higher standing among the Dili-based middle class (Mills et al. 2017), the On the north coast, a consistent positive trend in fish still need to be transported and sold through price per kilogram is evident (Figure 24), involving traders, resulting in a lower value for fishers. both reef fish and large pelagic fish. This is Dili-based fishers appear to sell their fish for the consistent with the growing population and relative highest price compared to other municipalities, wealth in Dili, the increased availability of fish, and which strengthens the theory that value is lost in potentially an increased awareness of the health transportation and trading (Figure 24). However, benefits of eating fish through externally funded the regions might not be easily comparable, as behavior change campaigns (Tilley et al. 2022). there have been differences in how the PeskAAS 31 enumerators record the pricing between the areas, During COVID-19, both the north and south coasts and over time. Lastly, there is additional uncertainty had a consistent increase in the price per kilogram of in asking fishers the value of their fish, as their catch. However, although the trend kept increasing responses could be aspirational and optimistic in the north coast, the south coast showed a quick rFaitghuere t h2a3n. Mbaesaend mona rrkeecte vnatl umea (rUkeStD p) rpiceers k. ilogram of the dcarotcph i n(J tahneu avreyr 2ag01e9 p troic eD (eFcigeumreb e2r4 )2. 021). 32 Note: The shaded areas around the time series represent the 95% confidence interval. The lines are the local polynomial regression. The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. Figure 23. Mean market value (USD) per kilogram of the catch (January 2019 to December 2021). Figure 24. Mean market value (USD) per kilogram of the catch by municipality (January 2019 to December 2021). 32 6. COVID-19 The global COVID-19 pandemic has had substantial crucial role of SSF as a source of food and income economic, social and health impacts on the entire in SIDS (Kelleher 2012), it is imperative to know world (World Bank 2020a; WTO 2020). While how the pandemic and associated domestic and most SIDS benefit from a remote location, which international movement restrictions have impacted can delay and limit the spread of COVID-19, this these fishing communities and their activities. also implies that foreign assistance during the crisis will be more problematic (Filho et al. 2020). In Timor-Leste, we observed a significant reduction Furthermore, foreign aid and tourism, which can of the total catch in Atauro (Figure 25 and Table be a vital part for the economy and well-being 3), with COVID restrictions being responsible for of SIDS, have dwindled because of international a decrease of 30.26 t of fish immediately after the travel and trade restrictions (Feeny and McGillivray lockdown, and a month-by-month reduction of 2010; Pratt 2015; Filho et al. 2020). Considering the 6.33 t relative to pre-COVID-19 conditions. Area Term Estimate Pvalue Trend before lockdown 6.891 >0.001 *** Atauro island Lockdown effect -30.26 0.026 * Trend after lockdown -6.33 >0.001 *** Trend before lockdown 0.3996 0.98 North coast Lockdown effect -152.1 0.356 Trend after lockdown 3.375 0.853 Trend before lockdown 2.128 0.146 South coast Lockdown effect 11.86 0.429 Trend after lockdown -1.328 0.428 Table 3. Interrupted time series analysis (ITS) of the total catch per month (January 2019 and December 2021). 33 Note: The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The red lines represent linear regressions split by the beginning COVID-19 restrictions. Figure 25. Monthly time series of the total catch (January 2019 to December 2021). Overall, the proportion of catches deemed for sale the entire data timeline, there was a significant by fishers decreased after the COVID-19 restrictions decrease in fish value in the north coast and Atauro, (Figure 26 and Table 4). On the north coast, from but the south coast remained level. The north coast April 2020, there was an immediate drop of 13.7% recorded a per month decrease of USD 0.96 and of fish deemed to market. In Atauro, this was 11.8%. Atauro a USD 0.25 drop. The combined effect of In contrast, the restrictions did not seem to affect social distancing rules, working from home and south coast fish sales, perhaps because of poor road movement restrictions between districts resulted infrastructure and the traditional reliance on more in far fewer people at markets and roadside stalls. localized market networks. When comparing pre- Furthermore, the tourism sector experienced a 95% and post-COVID periods up until December 2021, decline in revenue because COVID-19, so many all three areas experienced a decrease in the catch businesses depending on international travelers had deemed for the market over the longer term, about to partially or completely close (Rajalingam et al. 3.5% for the north coast and Atauro and 1% for the 2021). The fish favored by restaurants are often the south coast. larger species, and a collapse of the tourism sector is likely to affect the demand and thus the value The most apparent impact of the COVID-19 of these fish. As such, Timor-Leste experienced an pandemic in the PeskAAS dataset appears to be overall economic shock because of the pandemic on the market value of the fish. As an immediate and earlier political uncertainty. As a result, the effect, COVID-19 restrictions led to decreases in all country’s gross domestic product (GDP) is expected areas, though not significant (Figure 27 and Table to decrease 6.8%, leaving the local people with less 5). Comparing pre- and post-COVID periods, across money to spend (World Bank 2020b). 34 Area Term Estimate Pvalue Trend before lockdown 3.188 >0.001 *** Atauro island Lockdown effect -11.76 0.054 . Trend after lockdown -3.042 >0.001 *** Trend before lockdown 3.487 >0.001 *** North coast Lockdown effect -13.71 0.01 * Trend after lockdown -3.869 >0.001 *** Trend before lockdown 0.7309 0.008 ** South coast Lockdown effect 0.02715 0.989 Trend after lockdown -0.992 0.001 ** Note: Trend before lockdown refers to the indicator trend before the lockdown. Lockdown effect refers to the mean increase or decrease of the indicator immediately after the lockdown. Trend after lockdown refers to changes in the slope of the time series after the lockdown. Table 4. ITS of the catch stated for sale (January 2019 to December 2021). Note: The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The red lines represent linear regressions split by the beginning of COVID-19 restrictions. Figure 26. Monthly time series of the catch deemed for market (January 2019 to December 2021). 35 Area Term Estimate Pvalue Trend before lockdown -0.2381 >0.001 *** Atauro Island Lockdown effect -0.2606 0.623 Trend after lockdown 0.2591 >0.001 *** Trend before lockdown -0.1167 0.476 North coast Lockdown effect -1.27 0.493 Trend after lockdown 0.9621 >0.001 *** Trend before lockdown 0.3264 0.087 . South coast Lockdown effect -1.92 0.184 Trend after lockdown -0.4418 0.034 * Note: Trend before lockdown refers to the indicator trend before the lockdown. Lockdown effect refers to the mean increase or decrease of the indicator immediately after the lockdown. Trend after lockdown refers to changes in the slope of the time series after the lockdown. Table 5. ITS of the catch price (January 2019 and December 2021). Note: The gray shaded area defines the COVID-19 pandemic period, starting from the lockdown on March 28, 2020. The red lines represent linear regressions split by the beginning of COVID-19 restrictions. No available data prior to April 2019. Figure 27. Monthly time series of the fish market value (January 2019 to December 2021). 36 The pandemic and the resulting lockdown do the same household or family, so these restrictions not appear to have had as large an impact on are unlikely to have affected them as much. the fisheries of Timor-Leste as they had on other SSF according to some studies, where the loss of SSF are often considered to be a buffer or a revenue and decline in fishing activity has been safety net against income shocks, in their role substantial (Bennett et al. 2020; Knight et al. 2020; as supplier of high-nutrient low-cost food for Belton et al. 2021; Campbell et al. 2021; Ferrer lower-income consumers and as a labor buffer et al. 2021; Fiorella et al. 2021; Mangubhai et al. (Béné et al. 2010). Most SSF, especially those in 2021; Plagányi et al. 2021). This seems to suggest low-income countries, have low entry costs, and that fisheries are more resilient in Timor-Leste, result in limited assets being tied up in fishing which could be the result of the wholly domestic (Allison and Ellis 2001), which gives individuals consumption of national capture fisheries and more opportunity to move in and out of fishing the continued reliance on traditional systems activity at various periods in their lives (Jul Larsen and largely informal transactions. Other SSF et al. 2003). During the pandemic, Timor-Leste around the world lost a large part of their revenue experienced greater food insecurity than during because of the disruption of the international other years, partially because of the associated export market and the global supply chains restrictions and partially from droughts and crop (Knight et al. 2020; Belton et al. 2021; Plagányi pests (MAF et al. 2020). So it could be expected 2021). This will not have affected the fisheries that households engaging in both fishing and of Timor-Leste in a similar manner, as fish are agriculture would turn more to fishing when the almost entirely consumed domestically, with only crops are poor. This is reinforced by the fact that minimal export of seaweed and some informal people felt that agricultural activities were more trade with Indonesia (Barbosa and Booth 2009). affected by the restrictions than fishing activities Furthermore, while COVID-19 had substantial (MAF et al. 2020). impacts on the tourism in the country, the sector itself accounts only for 0.5% of the GDP, whereas in 40% of SIDS it accounts for more than 20% (UNWTO 2013; Rajalingam et al. 2020). In most literature cases reported, the decline in fisher incomes because of COVID-19 resulted in a decrease in their fishing effort, as the costs of continuing to fish (fuel, bait, ice) started to outweigh the diminishing revenue. This does not appear to be the case in Timor-Leste (Ferrer et al. 2021; Mangubhai et al. 2021). Another, more direct impact of the pandemic and the restrictions on the fishing activity of most small-scale fishers is the required social distancing and the fear of catching the virus (Okyere et al. 2020; Mangubhai et al. 2021). Many fishers had to restrict the size of their crew to be able to maintain the necessary distance on the boat, resulting in a decrease in overall effort (Mangubhai et al. 2021). Landing sites are often crowded, and it is difficult to keep 2 m apart from one another during the landing of the fish (Okyere et al. 2020). The fear of getting sick, which is a genuine concern for many in developing countries where healthcare is poor, causes some fishers to stop fishing altogether (Lau et al. 2020). However, the Timor-Leste fishing fleet is predominantly one- or two-person canoes or small motorboats (López-Angarita 2019), and crews are often from 37 Conclusions 6.1. Limitations This report is the first detailed analysis of the data collected by the PeskAAS monitoring system in Timor- Leste. There are various automatic and manual validation and data flagging steps incorporated as part of the PeskAAS workflow (Longobardi et al. 2021), and additional statistical and manual checking for specific analyses undertaken. However, because of the nature of the collection system (field based enumerators and fisher reporting), some errors might still be present. Furthermore, as a result of the continued development of the PeskAAS system over time with government partners, there were fewer enumerators and vessel trackers initially. As such, the number of records in the first year is low relative to later years, making outliers more prevalent and trends harder to determine. Some of the trends seen in the results could also be caused by sampling bias. All the recorded information, except for what comes from the PDS trackers, is based on choices and estimations made by the fishers and the enumerators. If there is not a clear protocol and uniform routine on how to sample the data, then this will likely result in biases between the stations. The choice on which fishing trips are sampled is also likely to include inherent bias, as enumerators will prefer to work with fishers that they know are reliable and cooperative. The accuracy of the records should, however, have increased over time because of training and experience. 6.2. Further research Over the past decade there has been a substantial rise in interest in SSF, with research and conservation efforts focused on protecting or restoring the critical nutritional and livelihood role they have in rural coastal and inland communities. As the International Year of Artisanal Fisheries, 2022 is in some ways the culmination of this growth, and the emerging work on Illuminating Hidden Harvests (FAO et al. 2022) aims to visualize the currently undervalued and invisible effects and benefits of SSF. The informal nature of SSF, their geographical and economic isolation, and the sheer diversity of fisheries, contexts and cultures involved has historically made it a very data-poor sector (Kolding et al. 2014). However, cheaper and smaller and more environmentally rugged technologies have enabled real-time sampling of SSF even in hostile and corrosive marine environments. Furthermore, the development of open-source software allows for the rapid scaling and adaptation of tools and methods from one context to another. This data obtained from PeskAAS creates crucial environmental and social benchmarks for the Timor-Leste fisheries sector, and provides decision-makers with information to support accountable management policies in the fisheries sector. This report only describes and analyzes a part of the fishery in Timor-Leste. PeskAAS includes large amounts of information that were not included in the analysis for this report. Recently, a new nutritional dashboard was included into PeskAAS that visualizes the potential contribution of fisheries catches to recommended daily intakes of six important micronutrients and fatty acids (Iron, Zinc, Calcium, Selenium Omega-3 fatty acids and vitamin A) based on models developed by Hicks et al. (2019). Further research is underway to explore management mechanisms to incorporate nutrition information into fisheries decision-making, such as developing stock assessment models from length-frequency information for key species identified as important in Timor-Leste fisheries, and for local nutrient yield. 6.3. Conclusion This study is the first to provide a detailed description of the SSF of Timor-Leste. There are variations in the characteristics of the fisheries within the country, caused by social, cultural, economic, geographic and climate differences. The north coast is the wealthiest and most densely populated area. As such, fisheries there invest in more expensive fishing equipment and are responsible for most of the landed fish, primarily small pelagic species. The south coast presents a more rudimentary fishing fleet dominated by 38 paddle canoes. It also experiences harsher and more variable weather conditions at sea without being well equipped against it, resulting in a discernible seasonal pattern of the fishing activity. Fishing is usually not the main economic activity of households on the mainland, as a diversified economy with agriculture and livestock farming are the preferred occupations. Atauro has a stronger fishing culture, as more actors identify themselves primarily as fishers. Timorese fishers sell their catch locally, and what they cannot manage to trade they will keep for their own consumption or local distribution. Compared to other comparable SIDS and their SSF, Timor-Leste appears to have been less affected by the impact of the COVID-19 pandemic. Fishing activities do not seem to have significantly changed, and the impact is mostly observed in the decline of the market value of the catches. However, the country does not rely on tourism and export as much as many other SSFs, which could explain the lesser impact. As the fisheries of Timor-Leste are being developed and promoted, attention should be given to preserving the resilience of the fisheries and creating safety nets for possible future crises so that food security will not be put in danger. Although the PeskAAS software and monitoring program are still developing and improving, they contain a lot of information that has enabled a description and analysis of a previously little-known fishery. Timor- Leste is a pilot study area for a wider implementation of PeskAAS. Hopefully, this study can be used to further develop and improve the monitoring system so that it can be deployed in other countries to increase our knowledge of SSF around the world. 39 References Allison EH and Ellis F. 2001. The livelihoods approach and management of small-scale fisheries. Marine Policy 25:377–88. Alongi D, Amaral A, Carvalho N, McWilliam A, Rouwenhorst J, Tirendi F, Trott L and Wasson R. 2012. River catchments and marine productivity in Timor-Leste: Caraulun and Laclo catchments: South and north coasts. Charles Darwin University. Darwin, Australia. [ADB] Asian Development Bank. 2014. State of the Coral Triangle: Timor-Leste.Mandaluyong City, Philippines Australian Bureau of Meteorology; [CSIRO] Commonwealth Scientific and Industrial Research Organisation. 2014. Climate variability, extremes and change in the western tropical Pacific: New science and updated country reports. Pacific-Australia Climate Change Science and Adaptation Planning Program Technical Report, Australian Bureau of Meteorology and Commonwealth Scientific and Industrial Research Organisation, Melbourne, Australia. Barbosa M and Booth S. 2009. Timor-Leste’s fisheries catches (1950–2009): Fisheries under different regimes. In Zeller D and Harper S eds. Fisheries catch reconstructions: Islands, Part I. 39–52. Fisheries Centre, University of British Columbia. Fisheries Centre Research Reports 17 (5). Belton B, Rosen L, Middleton L, Ghazali S, Mamun AA, Shieh J, Noronha HS, Dhar G, Ilyas M, Price C, et al. 2021. COVID-19 impacts and adaptations in Asia and Africa’s aquatic food value chains. Marine Policy 129:104523. Béné C, Hersoug B and Allison EH. 2010. Not by rent alone: Analyzing the pro-poor functions of small-scale fisheries in developing countries. Development Policy Review 28:325–58. Béné C and Friend RM. 2011. Poverty in small-scale fisheries: Old issue, new analysis. Progress in Development Studies 11:119–44. Bennett NJ, Finkbeiner EM, Ban NC, Belhabib D, Jupiter SD, Kittinger JN, Mangubhai S, Scholtens J, Gill D, Christie P 2020. The COVID-19 pandemic, small-scale fisheries and coastal fishing communities. Coastal Management 48:336–47. Brackhane S, Webb G, Xavier FME, Gusmao M and Pechacek P. 2018. When conservation becomes dangerous: Human-crocodile conflict in Timor-Leste: Crocodile dispersal. Journal of Wildlife Management 82:1332–44. Bradley D, Merrifield M, Miller KM, Lomonico S, Wilson JR and Gleason MG. 2019. Opportunities to improve fisheries management through innovative technology and advanced data systems. Fish and Fisheries 46:332. Braithwaite J, Charlesworth H and Soares A. 2012. Networked governance of freedom and tyranny: Peace in Timor-Leste. ANU Press. Campbell SJ, Jakub R, Valdivia A, Setiawan H, Setiawan A, Cox C, Kiyo A, Darman, Djafar LF, Rosa E et al. 2021. Immediate impact of COVID-19 across tropical small-scale fishing communities. Ocean and Coastal Management 200: 105485. CARE. 2020. COVID-19 crisis: Analysis of risk in Australia’s neighborhood. Melbourne, Australia: CARE. K. von Grebmer, J. Bernstein, R. Mukerji, F. Patterson, M. Wiemers, R. Ní Chéilleachair, C. Foley, S. Gitter, K. Ekstrom, and H. Fritschel. 2019. 2019 Global Hunger Index: The Challenge of Hunger and Climate Change. Bonn: Welthungerhilfe; and Dublin: Concern Worldwide. Dam Lam R, Dos Reis Lopes J, Lozano D, Da Costa D F, De Fátima Belo M, Da Silva J, Da Cruz G, Rossignoli C. 2022. An assessment of the impact of digital monitoring of small-scale fisheries on management and wellbeing outcomes in Timor-Leste. Penang, Malaysia: WorldFish. 40 [FAO] Food and Agriculture Organization. 2014. Global Blue Growth Initiative and Small Islands Developing States (SIDS). Rome: FAO. [FAO] Food and Agriculture Organization. 2015. Agriculture trends in Timor-Leste from 2010–2015. Rome: FAO. http://www.statistics.gov.tl/wp-content/uploads/2019/10/FAO_TL_infographics_S_English-language.pdf [FAO] Food and Agriculture Organization. 2020. The state of world fisheries and aquaculture 2020: Sustainability in action. Rome: FAO. doi:10.4060/ca9229en [FAO] Food and Agriculture Organization. 2021. Special report: 2021 FAO Crop and Food Supply Assessment Mission (CFSAM) to the Democratic Republic of Timor-Leste. Rome: FAO. doi:10.4060/cb5245en [FAO] Food and Agriculture Organization; WorldFish. 2020. Information and communication technologies for small-scale fisheries (ICT4SSF): A handbook for fisheries stakeholders. Bangkok: FAO; Penang, Malaysia: WorldFish. [FAO] Food and Agriculture Organization; Duke University; WorldFish. 2022. Small-scale fisheries and sustainable development: Key findings from the Illuminating Hidden Harvests report. Rome: FAO; Durham, USA: Duke University; Penang, Malaysia, WorldFish. Feeny S and McGillivray M. 2010. Aid and growth in Small Island Developing States. Journal of Development Studies 46:897–917. Ferrer AJG, Pomeroy R, Akester MJ, Muawanah U, Chumchuen W, Lee WC, Hai PG, Viswanathan KK 2021. Covid-19 and small-scale fisheries in Southeast Asia: Impacts and responses. Asian Fisheries Science 34:99–113. Fiorella KJ, Bageant ER, Mojica L, Obuya JA, Ochieng J, Olela P, Otuo PW, Onyango HO, Aura CM, Okronipa H. 2021. Small-scale fishing households facing COVID-19: The case of Lake Victoria, Kenya. Fisheries Research 237. doi:10.1016/j.fishres.2020.105856 Filho WL, Lütz JM, Sattler DN and Nunn PD. 2020. Coronavirus: COVID-19 transmission in pacific small island developing states. International Journal of Environmental Research and Public Health 17:1–8. Fujita R, Cusack C, Karasik R, Takade-Heumacher H and Baker C. 2018. Technologies for improving fisheries monitoring. Environmental Defense Fund, San Francisco, 71. https://www.edf.org/sites/default/files/oceans/Technologies_for_Improving_Fisheries_Monitoring.pdf General Directorate of Statistics. 2015. 2015 Timor-Leste Population and Housing Census: Data sheet. Statistics Timor-Leste. Dili, Timor-Leste: General Directorate of Statistics. Gorton C. 2018. Food and nutrition security in Timor-Leste: Challenges and prospects. Future Directions International. Hicks CC, Cohen PJ, Graham NAJ, Nash KL, Allison EH, D’Lima C, Mills DJ, Roscher M, Thilsted SH, Thorne- Lyman AL, et al. 2019. Harnessing global fisheries to tackle micronutrient deficiencies. Nature 574:95–98. Hilborn R, Amoroso RO, Anderson CM, Baum JK, Branch TA, Costello C, de Moor CL, Faraj A, Hively D, Jensen OP, et al. 2020. Effective fisheries management instrumental in improving fish stock status. Proceedings of the National Academy of Science USA 117:2218–24. Hind EJ. 2014. A review of the past, the present, and the future of fishers’ knowledge research: A challenge to established fisheries science. ICES Journal of Marine Science 72:341–58. Hunnam K, Carlos I, Hammer MP, Dos Reis Lopes J, Mills DJ and Stacey N. 2021. Untangling tales of tropical sardines: Local knowledge from fisheries in Timor-Leste. Frontiers in Marine Science 8:470. Ingram S, Kent L and McWilliam A. 2015. A new era? Timor-Leste after the UN. ANU press. Islam MM. 2011. Living on the margin: The poverty-vulnerability nexus in the small-scale fisheries of Bangladesh. Poverty Mosaics: Realities and Prospects in Small-Scale Fisheries. Springer, Dordrecht. 71–95. 41 Jul Larsen E, Kolding J, Overa R, Nielsen JR and van Zwieten P. 2003. Management, co-management, or no management?: Major dilemmas in Southern African freshwater fisheries. Fisheries Technical Paper 426/1 and 2. Rome, Italy: Food and Agriculture Organization. Kelleher K, Westlund L, Hoshino E, Mills D, Willmann R, de Graaf G, Brummet R. 2012. Hidden harvest: The global contribution of capture fisheries. Washington: WorldBank; Penang, Malaysia: WorldFish. Kolding J and Van Zwieten PAM. 2011. The tragedy of our legacy: How do global management discourses affect small scale fisheries in the south? Forum for Development Studies 267–97. Kolding J, Béné C and Bavinck M. 2014. “Small-Scale Fisheries: Importance, Vulnerability and Deficient Knowledge.” In Garcia S, Rice J and Charles A, eds. Governance for Marine Fisheries and Biodiversity Conservation. Interaction and Coevolution. John Wiley & Sons, Ltd. Chichester, UK. 317–31. Knight CJ, Burnham TLU, Mansfield EJ, Crowder LB and Micheli F. 2020. COVID-19 reveals vulnerability of small-scale fisheries to global market systems. The Lancet Planetary Health 4:e219. Lau J, Sutcliffe S and Hungito W. 2020. Lived experiences of Covid-19: Impacts on an atoll island community, Papua New Guinea. Townsville, Australia: James Cook University. Longobardi L, Tilley A and Cagua F. 2021. PeskAAS data workflow. Penang, Malaysia: WorldFish. https://digitalarchive.worldfishcenter.org/handle/20.500.12348/4929 Lopes S, Mendes HH, De Araujo VS, Martins A, de Fatima Mesquita M, de Jesus S, Lopes F, Fernandos E, Ribeiro M, De Araujo T et al. 2019. Timor-Leste em Número. Impressão. Dili, Timor-Leste: Director Geral de Estatística. López-Angarita J, Hunnam KJ, Pereira M, Mills DJ, Pant J, Teoh SJ, Eriksson H, Amaral L, and Tilley A. 2019. Fisheries and aquaculture of Timor-Leste in 2019: Current knowledge and opportunities. Program Report: 2019–15. Penang, Malaysia: WorldFish. https://worldfish.archive.knowledgearc.net/bitstream/ handle/20.500.12348/3737/Program-Report-2019-15-Timor-Leste-Fisheries.pdf?sequence=2&isAllowed=y Macusi ED, Macusi ES, Jimenez LA and Catam-isan JP. 2020. Climate change vulnerability and perceived impacts on small-scale fisheries in eastern Mindanao. Ocean and Coastal Management 189:105143. Mangubhai S, Nand Y, Reddy C and Jagadish A. 2021. Politics of vulnerability: Impacts of COVID-19 and Cyclone Harold on Indo-Fijians engaged in small-scale fisheries. Environmental Science and Policy 120:195–203. Mills, Westlund L, De Graaf G, Kura Y, Willman R and Kelleher K. 2011. Under-reported and undervalued: Small-scale fisheries in the developing world. Small-scale fisheries management: Frameworks and approaches for the developing world. 1–15. Mills DJ, Tilley A, Pereira M, Hellebrandt D, Pereira Fernandes A and Cohen PJ. 2017. Livelihood diversity and dynamism in Timor-Leste: Insights for coastal resource governance and livelihood development. Marine Policy 82: 206–215. Molyneux N, Da Cruz GR, Williams RL, Andersen R and Turner NC. 2012. Climate change and population growth in Timor-Leste: Implications for food security. Ambio 41:823–40. Moxham B and Carapic J. 2013. Unravelling Dili: The crisis of city and state in Timor-Leste. Urban Studies 50:3116–33. National Directorate of Food Security and Cooperation. 2019. The first IPC analysis report on the chronic food insecurity situation in Timor-Leste: Evidence and standards for better food security and nutrition decisions. Dili, Timor-Leste: National Directorate of Food Security and Cooperation. Okyere I, Chuku EO, Ekumah B, Angnuureng DB, Boakye-Appiah JK, Mills DJ, Babanawo R, Asare NK, Aheto DW, Crawford B. 2020. Physical distancing and risk of COVID-19 in small-scale fisheries: A remote sensing assessment in coastal Ghana. Scientific Reports 10: 1–13. 42 Plagányi É, Deng RA, Tonks M, Murphy N, Pascoe S, Edgar S, Salee K, Hutton T, Blamey L, Dutra L. 2021. Indirect impacts of COVID-19 on a tropical lobster fishery’s harvest strategy and supply chain. Frontiers in Marine Science 8:730. Población EA. 2013. Fisheries and food security in Timor-Leste: The effects of ritual meat exchanges and market chains on fishing. Food Security 5:807–16. Pratt S. 2015. The economic impact of tourism in SIDS. Annals of Tourism Research 52:148–60. Rajalingam G, Atwood K, Miranda A, Ximenes J, Santo G, da Conceição Teixeira I, Mata Dalan Institute, Pillai S, Lime L, Apse J, et al. 2021. 2020 economic impact of COVID-19 on MSMEs and Timor-Leste’s tourism sector. The Asia Foundation. Dili, Timor-Leste. Taylor-Leech K. 2009. The language situation in Timor-Leste. Current Issues in Language Planning 1–68. doi: 10.1080/14664200802339840 Tilley A, Wilkinson SP, Kolding J, López-Angarita J, Pereira M and Mills DJ. 2019. Nearshore fish aggregating devices show positive outcomes for sustainable fisheries development in Timor-Leste. Frontiers in Marine Science 6:487. Tilley A, Dos Reis Lopes J and Wilkinson SP. 2020. PeskAAS: A near-real-time, open-source monitoring and analytics system for small-scale fisheries. PLoS One 15:1–11. Tilley A, Byrd KA, Pincus L, Klumpyan K, Dobson K, Dos Reis Lopes J, Shikuku KM. 2022. A randomised controlled trial to test the effects of fish aggregating devices (FADs) and SBC activities promoting fish consumption in Timor-Leste: A study protocol. PLoS One 17:e0269221. Timor-Leste Department of Food Security, Ministry of Agriculture and Fisheries, Oxfam, Mercy Corps, United States Agency for International Development, Adra, CRS, Tomak, World Vision, FAO. 2020. Rapid food security assessment 2020: Full report – Timor-Leste. Dili, Timor-Leste: Timor-Leste Department of Food Security. Timor-Leste National Directorate of Meteorology and Geophysics; Australian Bureau of Meteorology; [CSIRO] Commonwealth Scientific and Industrial Research Organisation. 2015. Pacific-Australia Climate Change Science and Adaptation Planning Program: Current and future climate of Timor-Leste. Dili, Timor-Leste: National Directorate of Meteorology and Geophysics. Tomascik T, Mah AJ, Nontji A and Moosa MK. 1997. The ecology of the Indonesian seas. Hong Kong, China: Periplus Editions (HK). [UNWTO] UN World Tourism Organization. 2014. Tourism in Small Island Developing States (SIDS): Building a more sustainable future for the people of Islands. Madrid, Spain: UN World Toursim Organization. Venugopa V and Shahidi F. 1995. Value-added products from underutilized fish species. Critical Reviews in Food Science and Nutrition 35:431–53. Ximenes EM, Yeros MC, Gusmão C, Lopes S and Mendes HH. 2018. Timor-Leste: Annual external trade statistics 2018. Dili, Timor-Leste: General Director of Statistics. World Bank. 2020a. Food security and Covid-19: Understanding poverty. Washington: World Bank. https://www.worldbank.org/en/topic/agriculture/brief/food-security-and-covid-19 World Bank. 2020b. Timor-Leste economic report, October 2020: Towards a sustained recovery. Washington, DC: WorldBank. [WTO] World Trade Organization. 2020. Trade statistics and outlook: Trade set to plunge as Covid-19 pandemic upends global economy. Press Release. Geneva: World Trade Organization. 43 About WorldFish WorldFish is an international, not-for-profit research organization that works to reduce hunger and poverty by improving aquatic food systems, including fisheries and aquaculture. It collaborates with numerous international, regional and national partners to deliver transformational impacts to millions of people who depend on fish for food, nutrition and income in the developing world. The WorldFish headquarters is in Penang, Malaysia, with regional offices across Africa, Asia and the Pacific. The organization is a member of CGIAR, the world’s largest research partnership for a food secure future dedicated to reducing poverty, enhancing food and nutrition security and improving natural resources. For more information, please visit www.worldfishcenter.org 44