The research team, working with landing records from Palabuhanratu National Fishing Port, assembled monthly catch per unit effort (CPUE) data for skipjack tuna (Katsuwonus pelamis) from 2019 to 2023, focusing on troll-line vessels between 5 and 10 gross tonnage. They paired these fisheries data with seven oceanographic variables drawn from the Copernicus Marine Service: sea surface temperature, salinity, sea surface height, and current speed on the physical side, and chlorophyll-a concentration, pH, and silicate on the biogeochemical side. Using a Generalized Additive Model, a statistical framework designed to capture non-linear relationships without imposing a fixed functional form, the team tested how each variable and each combination of variables related to monthly catch rates.

The results were striking. The integrated model, which combined physical and biogeochemical predictors, explained 70.7 percent of the deviance in CPUE with an adjusted R-squared of 0.623 and the lowest Akaike Information Criterion value of the three candidate schemes (743.26, compared with 762.64 for physical-only and 778.73 for biogeochemical-only models). Five-fold cross-validation yielded a mean root mean square error of 0.485 on the log-transformed CPUE scale, indicating moderate predictive performance without strong evidence of overfitting. In plain terms: knowing the ocean state helped predict the catch, and knowing both the physics and the chemistry of the water helped far more than knowing either alone.

Salinity emerged as the single most influential predictor, showing a strong non-linear relationship with CPUE (P < 0.001). The response curve revealed that higher catch rates tended to occur at lower salinity, a pattern the authors interpret as a signature of water-mass origin and mixing. Lower surface salinity in Palabuhanratu waters may reflect freshwater input from rivers such as the Cimandiri, or the upwelling-driven advection of coastal water masses, both of which are associated with enhanced nutrient availability and productivity. Sea surface height also showed a significant negative association, consistent with the role of negative SSH anomalies as indicators of upwelling, convergence zones, and mesoscale features such as fronts and eddies that concentrate prey and attract pelagic predators.

Chlorophyll-a, the classic satellite-era proxy for phytoplankton biomass and primary productivity, showed a positive relationship with CPUE (P < 0.01), meaning higher catch rates coincided with greener, more productive water. This is exactly what fisheries oceanographers would expect: phytoplankton blooms feed zooplankton, which feed small pelagic fish, which in turn feed skipjack. The variable also carried enormous temporal variability, with a coefficient of variation of 110.48 percent, making it a particularly informative signal of shifting foraging conditions. pH, by contrast, was statistically significant but varied within an extraordinarily narrow range of 7.99 to 8.05, and the authors caution that its apparent effect likely reflects underlying biogeochemical co-processes rather than any direct physiological influence on the fish.

Sea surface temperature told a subtler story. Higher catch rates tended to occur under cooler surface conditions, but the relationship was only marginally significant (P = 0.068). In the southern Java upwelling system, cooler surface water typically signals the advection of nutrient-rich subsurface water and the development of frontal structures that enhance prey aggregation. Rather than acting as a direct physiological constraint on the tuna, the authors suggest, SST in this context functions as an integrative indicator of broader productivity processes. Silicate, included specifically as a tracer of upwelling water-mass intrusion, did not reach statistical significance, likely because chlorophyll-a already captured the integrated productivity signal of upwelling at monthly resolution.

The temporal patterns themselves tell a compelling story. Monthly CPUE ranged from roughly 67 kilograms per trip in July 2019 to a peak of 778 kilograms per trip in July 2022, with higher values generally recorded during 2022 and 2023 and recurring mid-year peaks between June and September. The study period spanned contrasting phases of the Indian Ocean Dipole: 2022 was dominated by a negative IOD phase associated with warmer surface waters and weaker upwelling, while a positive IOD phase developed from August to December 2023, typically bringing enhanced upwelling, cooler SST, and elevated chlorophyll to the eastern tropical Indian Ocean. Notably, high catch rates appeared during both phases, suggesting that CPUE variability at Palabuhanratu cannot be attributed to IOD forcing alone but likely emerges from the interaction of multiple oceanographic processes at regional and local scales, with possible modulation by the El Niño–Southern Oscillation.

The methodological care underlying these findings deserves attention. Because individual fishing coordinates were not consistently available across all months, the team delineated a fishing-ground polygon using participatory field observations from October and November 2022, fishermen’s GPS records, and available logbook positions, then averaged all environmental grid cells within that polygon for each month. This spatial aggregation, while necessary, smooths fine-scale features such as thermal fronts and eddies that may matter to skipjack distribution. The biogeochemical product’s coarser 0.25-degree resolution (roughly 28 kilometers) also meant fewer grid cells contributed to the averages compared with the 0.083-degree physical product, introducing additional uncertainty. Cross-validation against independent MODIS satellite observations confirmed strong temporal agreement for both SST (Pearson r = 0.938) and chlorophyll-a (Pearson r = 0.886), supporting the reliability of the Copernicus products for capturing monthly environmental fluctuations even where absolute values differ.

Perhaps the most important caveat, and one the authors emphasize throughout, is that CPUE is a fishery-dependent index. It reflects not only fish abundance but also fishing behavior, spatial targeting strategies, vessel efficiency, and operational factors that could not be fully standardized across the 5 to 10 GT fleet segment. The relationships identified here are statistical associations between environmental variability and catch rates, not definitive evidence of skipjack habitat suitability or causal ecological mechanisms. Monthly averaging may also obscure short-term oceanographic events relevant to individual fishing trips.

Even with those caveats, the study’s central message carries real weight for coastal fishing communities. Skipjack catch variability in tropical waters is not governed by a single environmental dial but by the simultaneous interplay of temperature, salinity, sea level, currents, productivity, and biogeochemical cycling. For small-scale fishers whose operational range is constrained by vessel size and fuel capacity, and who have traditionally relied on trial-and-error to locate fish schools, environmentally informed forecasting that integrates multiple oceanographic dimensions could reduce search time, lower fuel costs, and improve catch efficiency. As climate-driven variability intensifies across the Indian Ocean, understanding these coupled physical-biogeochemical drivers may become not just scientifically interesting but essential to the adaptive capacity of the communities that depend on these waters.

Subject of Research: Environmental drivers of skipjack tuna catch variability in tropical coastal waters off West Java, Indonesia

Article Title: Environmental drivers of skipjack tuna, Katsuwonus pelamis ( Actinopterygii , Scombriformes , Scombridae ), catch variability in tropical coastal waters: Insights from physical and biogeochemical oceanography

Article References: Environmental drivers of skipjack tuna, Katsuwonus pelamis ( Actinopterygii , Scombriformes , Scombridae ), catch variability in tropical coastal waters: Insights from physical and biogeochemical oceanography. (n.d.). https://doi.org/10.3897/aiep.56.193569

Image Credits: AI Generated

DOI: 10.3897/aiep.56.193569

Keywords: skipjack tuna, CPUE, Palabuhanratu Bay, Generalized Additive Model, chlorophyll-a, sea surface temperature, salinity, upwelling, Indian Ocean Dipole, small-scale fisheries, Copernicus Marine Service, biogeochemical oceanography