Skip to contents

Estimates monthly total catch and revenue per district and fishing unit with FAO's ARTFISH method, as set out for OPEN ARTFISH (de Graaf, Stamatopoulos & Jarrett 2017, FAO): boats in the census × days each boat fishes in the month × catch per fishing day. generate_fleet_analysis() runs it beside the GPS tracker method, so the two come from the same data and can be compared.

Usage

raise_catch_fao(conf, assets)

Arguments

conf

Country configuration from read_config(). Its fao.surveys names the validated survey file carrying the weekly fishing-days answer; without it the country has no FAO raising.

assets

The Airtable assets snapshot, with its frame_units, gear_groups and vessel_groups tables.

Value

A list, or NULL when the country has no fao.surveys or the raising fails: units, one row per district, fishing unit and month (see estimate_catch_fao()), and species, the same catch by species (see estimate_species_fao(); NULL if that split fails).

Details

  • Fishing unit: the Toolkit's boat type and main gear, as far as the census (Airtable frame_units) counts them: both in Kenya and Zanzibar, boat type alone in Mozambique. Each is grouped as the Toolkit advises, so units get more samples: the boat into its vessel_group (Airtable vessels, set per country; Mozambique's boat types stay apart) and the gear into its FAO ISSCFG main group (FAO_category in Airtable gears).

  • Catch per fishing day: the trip's total catch (or value); one trip is one fishing day. Trips recorded as catching nothing count as zero. A district, unit and month with fewer than five trips takes the catch per trip of all units in that district and month; catch_source and revenue_source say which.

  • Days fished: the survey answer to "days fished last week", scaled to the month. A district, unit and month with fewer than five answers takes the district and unit average over all months; effort_source says which.

  • Too few trips: a district and month with fewer than ten surveyed trips is not raised (totals NA, quality too_few_trips), as in calculate_district_totals().

  • Error: the Toolkit's compound relative error at 90%, flagged pass (up to 15%), warn (up to 20%), fail or unknown.

  • Species: each unit's raised catch is split by the species mix of the trips its catch per trip came from (estimate_species_fao()).

  • Only trips from the survey named in fao.surveys are used, the one that asks the weekly question: KEFS in Kenya, WF in Zanzibar and DINAPA in Mozambique.

Revenue stays in local currency, as in generate_fleet_analysis().