Predicting Ministerial Effectiveness a Review of Empirical Research Pickens

Introduction

Albacore tuna (Thunnus alalunga) is an of import upper tropic-level oceanic predator distributed globally betwixt approximately 50°Northward and 45°S, with relatively lower abundance in equatorial areas. Albacore is also a commercially important species that are considered as one of the best types of tuna available for canning (Gloria et al., 1999) and contributed about 3.36% to the almanac global tuna catch of 7.3 million tons in 2019 (FAO, 2021). One of the largest fisheries for albacore is in the South Pacific Ocean where annual harvest contributed nearly 60% of the grab in the Pacific. In the South Pacific Ocean, most of the albacore was catched by longline fisheries (>ninety%), the distant water longline fishing vessels from Taiwan, Communist china, and the domestic longline fleets of various Pacific Isle countries and territories (PICTs) (Brouwer et al., 2018), that total catch has fluctuated around 80,000 tons in recent 10 years. In add-on, the troll fishery mainly captures juvenile albacores in New Zealand's coastal waters and the central Pacific Ocean since the mid-1980s with the amount of total catch fluctuated effectually two,000 tons.

Agreement and predicting responses to global climate change are important issues for the scientific community to assist in designing effective fishery direction to ensure the sustainability of the south Pacific albacore stock given that albacore fisheries provide significant economic, social, and cultural benefits to PICTs (Bell et al., 2013). A range of environmental variables influence albacore spatial distribution. By studies take suggested that the mixed layer depth tin can limit the vertical distribution of albacore tuna (Childers et al., 2011; Williams et al., 2015). Langley and Hampton (2005) indicated that temperature is a fundamental factor for albacore horizontal distribution and its motion tends to correspond to the seasonal oscillation of the location of the 23–28°C isotherm of sea surface temperature. Brill (1994) suggested that the dissolved oxygen (DO) is a adept index of albacore habitat suitability that the reduced ambient oxygen levels would prolong the time required for albacore to recover from strenuous exercise. Salinity and chlorophyll-a (CHL) concentration were also suggested to be related to the albacore abundance and distribution through affecting the availability of their prey (Xu et al., 2013; Novianto and Susilo, 2016).

Climate modify has been suggested to have a pregnant impact on the marine ecosystems that influence species distribution (due east.k., striped marlin, Su et al., 2013; pelagic squid, Alabia et al., 2016; anchovy, Silva et al., 2016; tunas, Erauskin-Extramiana et al., 2019). Longer-term trends in the oceanography of the Southern Hemisphere have too been detected (Ganachaud et al., 2011), which may have influenced the large-scale distribution of albacore. Lehodey et al. (2015) suggested that the northern boundary of south Pacific albacore distribution may shift southward past roughly 5° latitude in 2080 by using the Spatial Ecosystem and Population Dynamics Model (SEAPODYM), which was a coupled physical-biological interaction model that describes spatial and temporal tuna population dynamics (Lehodey et al., 2008). However, Senina et al. (2018) by using the same model indicated that the albacore distribution may remain stable in the western office of the southern Pacific Ocean in 2100, just may expand in the eastern Pacific Bounding main without the hypothesized oxygen change. Erauskin-Extramiana et al. (2019) used the generalized additive models (GAM; Hastie and Tibshirani, 1990) that split the data into 2 components: the probability of zero occurrence and nonzero observations (i.e., delta-GAM approach, Maunder and Punt, 2004); and indicated that the abundance is expected to subtract in temperate areas and the southern boundary of the distribution may shift southward in 2100.

Although previous studies have evaluated the future spatial and temporal variations of the s Pacific albacore distribution nether climate modify, there is no consensus on the predicted results derived from diverse approaches. The doubtfulness in future climatic conditions has been considered as one of the major sources of doubt in the species distribution projections (Beaumont et al., 2008). In this report, nosotros aim to apply an ensemble modeling approach (Hobday, 2010) that includes the dubiety of various atmosphere-ocean full general circulation models (AOGCMs) and anthropogenic emission scenarios of the Intergovernmental Panel on Climatic change (IPCC) fifth stage of the Coupled Model Intercomparison Project (CMIP5) to explore the resulting range of projections on future albacore distribution under climate change (Araújo and New, 2007; Alabia et al., 2016). In this study, the "south" Pacific albacore stock refers to the spatial coverage of available albacore information from both the Western and Cardinal Pacific Fisheries Commission (WCPFC) and the Inter-American Tropical Tuna Commission (IATTC) in the Due south Pacific Bounding main. Firstly, we quantify the human relationship between albacore relative abundance and potential ecology variables by using a GAM approach. Secondly, the large-scale future distribution changes of albacore is investigated by using ensemble forecasts based on the environmental information projected by the IPCC-class AOGCMs under different degrees of ocean warming. Finally, following the emerging objective of using CPUE is an important proxy for the economic viability of the due south Pacific albacore fisheries (Pilling et al., 2016; WCPFC, 2018), we besides evaluate the future alter in albacore relative abundance within the exclusive economical zones (EEZs) of the countries and territories in the S Pacific Ocean. The findings will be relevant for the south Pacific albacore stock conservation and volition contribute to understanding the potential impacts of climate change on albacore fisheries. Our report was developed and illustrated in the context of s Pacific albacore. However, information technology should be broadly applicable to other pelagic species for which similar information are available.

Materials and Methods

Fishery Information

Catch and attempt data grouped by year (1954–2016), quarter and 5° × 5° grid prison cell for the longline fisheries for the s Pacific albacore was downloaded from the public domain dataset of the WCPFC¹ and IATTC.² In addition, catch and effort information for the Taiwanese distant-water longline fishery for south Pacific albacore (1964-2016) obtained from the Overseas Fisheries Development Council of Taiwan³ was used every bit a substitution of public domain dataset for the fleet as the best bachelor information. Albacore nominal take hold of-per-unit-effort (CPUE, number of fish per 1,000 hooks) calculated as a proxy for the albacore relative abundance (Erauskin-Extramiana et al., 2019) during 1997–2016 was used for the analyses because the satellite-based ecology data are non bachelor before 1997. The CPUE was calculated using the formula beneath:

$$CPU{\mathrm{Eastward}}_{y,q,i,f}=\frac{N_{y,q,i,f}}{{\mathrm{Eastward}}_{y,q,i,f}}(1)$$

where N is the number of fish defenseless; E is the number of hooks (in thousands); y is yr; q is quarter; i is the spatial location of the 5°× 5° grid cell; f is the flag of longline fleet.

The longline fisheries dataset past flags included in this study were summarized in Supplementary Table one. The data grooming was practical to remove records exterior of the temporal or spatial span of the analysis, or with improbable records. Records with missing logdates, 0 hook, more than albacore defenseless than the number of hooks, CPUEs larger than 2 times the interquartile range (i.e., outliers) and fleets that were only present in the dataset for a very short menstruation of fourth dimension or the pocket-size proportion (< 1%) of catch, were all excluded from the dataset (Tremblay-Boyer et al., 2015). In total, in that location are 17,393 records left from 19,799 records (with vii% removal) after the data cleaning.

Historical and Future Ecology Data

The environmental variables previously investigated for possible effects on the distribution of s Pacific albacore (Briand et al., 2011; Novianto and Susilo, 2016; Erauskin-Extramiana et al., 2019) were included in this study: sea surface temperature (SST, °C), sea surface salinity (SSS, PSU), mixed layer depth (MLD, yard), dissolved oxygen concentration under 100 grand depth (DO100, mmol L^––1), and chlorophyll-a concentration (CHL, kg L^––i). Monthly averaged satellite data [SST, SSS, MLD, DO100, and CHL] from 1997 to 2016 were used to narrate the environmental preferences of s Pacific albacore by using the GAM assay (see department ''Species Distribution Model''). All satellite datasets were downloaded from NOAA coastwatch⁴ except for the DO100 which were computed using the Pelagic Interactions Scheme for Carbon and Ecosystem Studies volume 2 (PISCES-v2) biogeochemical model and were downloaded from the Copernicus-Marine environment monitoring service^v (Supplementary Tabular array ii). All environmental datasets were rescaled to quarterly five°× 5° spatial resolution based on the coarsest scale of longline fisheries information.

Projections of ecology variables for the reference period (1997-2016), electric current (2020), mid (2050), and tardily (2080) catamenia of the twenty-first century were extracted from the IPCC CMIP5 with the Representative Concentration Pathways (RCP) 4.5 and 8.5 (van Vuuren et al., 2011) to generate historical and time to come potential south Pacific albacore habitat predictions in the GAM analysis. Ecology variables were downloaded from the Globe System Grid Federation (ESGF)⁶. The RCP iv.5 and 8.v are characterized past the stabilization without overshoot pathway to 4.5 W m^–ii (650 ppm CO_ii eq) and by rising radiative forcing pathway leading to viii.5 W thousand^–ii (1370 ppm CO_two eq), respectively, by 2100 (Riahi et al., 2011; Thomson et al., 2011). The 2 scenarios were considered as the intermediate and pessimistic scenarios in this written report. The listing of the 5 AOGCMs used in this study was summarized in Table 1. The spatial resolution of AOGCMs-predicted environmental data varied from 0.3 to i°× 1° to 1.5°× 1° (breadth by longitude). In this report, the predicted environmental data were interpolated to a one° × 1° resolution for use in the forecasts of the spatial distribution of the south Pacific albacore. Some variables that showed high variation among the predictions of AOGCMs in the exploratory assay, identified past the coefficient of variation (CV) larger than 100%, were not included in the further analysis (east.g., CHL, CV ranged from 82 to 107%).

Table 1. List of IPCC atmosphere-ocean full general circulation models where the future environmental variables, including the sea surface temperature (SST, °C), body of water surface salinity (SSS, PSU), mixed layer depth (MLD, thou), chlorophyll-a concentration (CHL, kg L^–i), and dissolved oxygen concentration nether 100 m depth (DO100, mmol L^–one), were obtained to generate potential habitat maps of south Pacific albacore nether sea warming scenarios (RCP iv.5 and RCP 8.5).

Species Distribution Model

A diverseness of approaches can exist used to model the relationship between fish relative abundance and ecology variables for the species distribution models (SDMs) (e.g., MaxEnt, random forests, boosted regression trees, etc.) (Pickens et al., 2021). Because the focus of this newspaper for the ensemble analysis is to reduce the dubiousness in predictions of ocean weather condition from AOGCMs rather than the doubtfulness in model construction among various SDMs, therefore nosotros only considered the nigh common method for predicting the relative affluence of tuna species, such as generalized linear modeling approach (Arrizabalaga et al., 2015; Lan et al., 2018; Erauskin-Extramiana et al., 2019). SDM of the south Pacific albacore was constructed by modeling fishery-dependent CPUEs in relation to the considered historical environmental variables using the generalized additive models (GAMs) (Hastie and Tibshirani, 1990; Wood, 2012, 2017). GAMs were selected rather than the GLMs (generalized linear models) considering they can bargain with multiple non-linear relationships between the covariates and the response variable in a semi-parametric manner (Hastie and Tibshirani, 1990). Apart from the continuous environmental variables, the year, quarter, flag were modeled equally the categorical variables for the availability and/or catchability (Arrizabalaga et al., 2015). The raw CPUE data from the tuna longline fisheries includes many zeros. In this context, it was advisable to fit the delta arroyo which models the probability of encounter of a fish population and the non-zero CPUE when fish are encountered (Pennington, 1983, 1996; Lo et al., 1992). Due to the low percentage of zero catches (<10%) of the quarterly 5° × five° aggregated dataset that implying goose egg inflation was not an issue (Ichinokawa and Brodziak, 2010), we simply fit the catch charge per unit model using the log-transformed CPUEs of albacore, with a small constant (x% of the grand mean) added to avoid log-transformation problems (Campbell et al., 1996; Howell and Kobayashi, 2006; Mugo et al., 2010). GAMs were built using the gam function of the "mgcv" package in R-language (Wood, 2012). Start, we constructed a full model past including all independent variables, information technology can be written as:

$$lo\mathrm{1000}\left(CPUE+1\right)\sim \mu +Year+Quarter+Flag+\mathrm{south}\left(S\mathrm{Due\; south}T\right)+$$

$$\mathrm{southward}\left(SSS\right)+s\left(\mathrm{Grand}\mathrm{Fifty}D\right)+s\left(DO100\right)(2)$$

where μ is the intercept value; Year, Quarter, Flag are the stock-still consequence for year, quarter, and catchability of longline-armada flag, respectively, and s(x) is a smoothing function for the independent covariates. To avoid overfitting, degrees of smoothness ("m" values) were set to less than or equal to 8 (Erauskin-Extramiana et al., 2019).

Variable Choice and Model Validation

Five candidate models were selected by the dredge part of the "MuMIn" R-package (Barton, 2016). This function compares all possible model structures with dissimilar combinations of all independent variables from the full model, and ranks those models according to their Akaike information criterion (AIC) value and the goodness-of-fit measure (percent of deviance explained) (Bruge et al., 2016; Erauskin-Extramiana et al., 2019). Multicollinearity was tested by using the ggpairs part of the "GGally" R-package if whatever highly correlated environmental variables should be removed (Arrizabalaga et al., 2015). A pairwise correlation that exceeds a threshold of 0.5–0.7 is defined as the presence of high collinearity (Dormann et al., 2013). The relative importance of predictor variables was evaluated by the percentage relative changes in the explained deviance and the AIC value of dropping each main effects factor from the total GAM (Kwon et al., 2018). Model performance of the top v candidate models was validated using g-fold cross-validation method. We used grand = 5, data were randomly split into 5 subsets, using 80% of information to validate the remaining 20% (Erauskin-Extramiana et al., 2019; Georgian et al., 2019). For each model, the average R ² value derived from CPUE observations and GAM predictions of grooming data and validation information, respectively, was calculated as a measure for goodness-of-fit. We defined the model of average R ² > 0.6 equally a good prediction performance. A big deviation between the training and testing R ⁱⁱ would indicate overfitting (Villarino et al., 2015; Erauskin-Extramiana et al., 2019). The model with the lowest AIC and highest average R ⁱⁱ value was selected as the best model for the further analyses (Brewer et al., 2016).

Historical and Futurity Albacore Distribution Projection

Historical ecology data projected by the AOGCMs were used as inputs in the GAM to approximate the albacore distribution in 1997–2016. The predicted relative abundance values were and so compared with the observed CPUE from the longline fisheries in 1997–2016 to evaluate the model functioning for forecasting potential habitat. For estimating the time to come impacts of climate change on albacore distribution, the ensemble GAM projections (see Department "Ensemble Forecasting") given the environmental conditions in the current (2020), mid (2050) and late (2080) periods of the 20-get-go century derived from the five AOGCMs were compared to the projections of recent five years of 2012–2016 under ii emission scenarios (RCP 4.v and RCP 8.5). In the future habitat distribution projection (2020–2080) of each AOGCM, model projections were performed with the fixed cistron "Year" kept at the mid-year between 1997 and 2016. Flag factor was gear up to a longline armada (e.g., TW) that simply represents a scaling parameter for the catchability. For some AOGCMs (IPSL and HadGEM) that certain ecology variables were unavailable (MLD and DO100), the quarterly boilerplate value derived from other GCMs was used equally a substitute.

Ensemble Forecasting

This study used ensemble assay to deal with the uncertainty between climate models, information technology has been proved to outperform a single model in past studies (Diniz-Filho et al., 2009; Crimmins et al., 2013). We used two ensemble methods to measure out the distribution of the south Pacific albacore. First, the consensus forecasting (Araújo et al., 2005; Araújo and New, 2007) was made by calculating the cardinal tendency (due east.g., the mean or median) of the GAM predictions based on the ecology information projected by v AOGCMs under the RCP four.5 and 8.5 scenarios, therefore reduce inter-model variances propagated from the AOGCMs. The median prediction is deemed less sensitive to outliers than the hateful; therefore, in the present study, consensus forecasting was employed to GAMs predictions using the median value (Alabia et al., 2016). Furthermore, the spatial distributions of the anomaly of relative affluence of southward Pacific albacore were calculated by subtracting the predictions from the median ensemble GAM projections by the electric current (2020), mid (2050), and belatedly (2080) periods of the xx-kickoff century from the recent boilerplate of 2012–2016. The 2nd ensemble arroyo used in this study was probabilistic forecasting. This involved computing the likelihood of the presence of preferred albacore habitats from each AOGCM and expressing this in the form of a probability map. The set of grid cells that may exist considered preferred habitats were defined as those for which predicted relative abundance from the GAM is in the top 15% (Su et al., 2013). Thus, the probability of preferred habitats at each ane° × 1° cell was based on the percentage of 5 AOGCMs for which the predicted relative abundance at that location is considered preferred, called "agreement map" (Porfirio et al., 2014).

Expected Changes in Exclusive Economic Zones

The potential relative abundance modify averaged per filigree jail cell for the south Pacific albacore under future climate change was estimated within the sectional economic zones (EEZs) for the countries and territories in our study area. EEZ data (downloaded from http://world wide web.marineregions.org) delimit the 200 nautical miles boundary from each coast (Flanders Marine Plant, 2018). This study merely analyzed those countries or territories with more than 30% of the grid cells of fishery data inside the EEZ to ensure it was representative of relative abundance (Erauskin-Extramiana et al., 2019). For each RCP scenario, the change in relative abundance was calculated past subtracting the predicted relative affluence of median ensemble during the primary fishing season within each EEZ in 2020, 2050, and 2080 from that in the recent 5 years of 2012-2016.

Results

Variable Selection and Model Validation

Scatterplots betwixt environmental variables suggested depression collinearity that the accented values of correlation coefficients are less than 0.v (Supplementary Effigy i). Results for each of the top five candidate models (model, predictor variables, constructive degrees of liberty, AIC, P-value, and deviance explained) selected by dredge part were shown in Table ii. In all the models, the shine terms were highly significant (P < 0.001). Overall, the total GAM had the lowest AIC value and the highest deviance explained (62.ane%). Variable rankings based on the per centum relative changes in the explained deviance for the 3 most influential explanatory variables were: (ane) DO100 (22.3%); (2) Flag (21.5%); and (3) SST (half dozen.v%) (Supplementary Table 3). Cross-validation testing suggested that the total GAM had slightly improve performance, with the highest R ² values for the training (0.62) and validation (0.61) datasets, compared to other candidate models (ranged 0.59–0.61 for both training and validation). A negligible departure in the R ^two values indicated the absenteeism of overfitting. Furthermore, the residuals of the full GAM suit well to the supposition of lognormality based on the distribution of residuals and quantile-quantile (QQ) plots (Supplementary Figure ii). Therefore, the full GAM was selected as the best model for the future habitat distribution prediction.

Table two. Elevation v candidate models selected past the dredge function of the "MuMIn" R-package.

Generalized Additive Models-Derived Habitat Characteristics for Albacore Tuna

The response curves for environmental variables derived from the total GAM were interpreted in terms of habitat suitability of the south Pacific albacore. The fitted smoothen curves bear witness the preferred SST range of 13–22°C with a peak at around 15°C and SSS ranges of 34–35 PSU and > 36 PSU. Preferred MLD was betwixt twenty and 60 m and the DO100 betwixt 0.2 and 0.25 mmol Fifty^–i (Figure 1). Approximate confidence interval envelopes were besides plotted for each part. Lower precision was observed in the estimations for lower values of SST (<15°C), SSS (<34 PSU), DO100 (<0.1 mmol L^–1) and higher values of MLD (>100 thou) and DO100 (>0.25 mmol L^–1) because of fewer information. The estimates of the year, quarter, and flag factors from the GAM were shown in Supplementary Figure iii. The year factor showed a significant decline during 1997–2003 and increase during 2004–2010. In that location is a meaning seasonal variation in the quarter factor with the higher estimates in quarters 2 and 3. The flag factor showed college estimates for CK, NC, US, and WS, but lower estimates for FM, AU, JP, KR, and NZ.

Figure 1. Response curves for the environmental variables, including the (A) bounding main surface temperature (SST, °C); (B) bounding main surface salinity (SSS, PSU); (C) mixed layer depth (MLD, m), and (D) dissolved oxygen concentration nether 100 one thousand depth (DO100, mmol Fifty^–ane), derived from the GAM model. The bluish points correspond the partial residuals of observed CPUE data, and the gray polygons represent the 95% confidence interval. The hatch marks at the bottom are a descriptor of the frequency of information points.

Historical and Hereafter Albacore Distribution Projection

The quarterly average distribution of albacore during 1997–2016 was shown in Figure 2. Model prediction has mimicked an apparent north-south seasonal variation of albacore distribution. The predicted spatial distributions in quarters 2–4 generally match well with the CPUE of longline vessels, in which the Pearson correlation coefficient (r) ranged from 0.62 to 0.71. However, the prediction fitted relatively poorly in the beginning quarter for the fishery data located north of 35°S (r = 0.6). In addition, annual aggregated CPUE values superimposed on the projected distributions had the r values ranged from 0.41 to 0.68, which indicated that the ensemble approach could more often than not yield reliable predictions overtime except for 2002.

Figure ii. Quarterly average take hold of-per-unit-attempt (CPUE) (number/i,000 hooks) distributions (open circles) for the south Pacific albacore overlaid on the median-ensemble relative abundance map (color contours) based on the GAM which was developed with data from 1997 to 2016.

Annual time-series of the areal-averaged of the environmental variables from 2020 to 2080 projected by the five AOGCMs were shown in Supplementary Figure four. The predicted levels of environmental changes beyond the five AOGCMs exhibited differences in magnitude, for example, the everyman SST predictions past the IPSL and the highest past the HadGEM. Under the RCP 4.v, the median value was expected to increase from twenty.8 to 21.ane°C for SST, decrease from 81.3 to 73.9 m for MLD, and decrease from 0.22 to 0.218 mmol Fifty^–1 for DO100, respectively, during 2020-2080, while no trend was constitute in the SSS. Nether the RCP 8.5, the more pronounced changes in SST with 2°C warming and in MLD with 12 one thousand shallowed were observed during 2020–2080.

The future ensemble projections of albacore distributions in the master angling season [i.e., quarter iii, which represents the highest (36%) historical harvest] derived from the v AOGCMs for the current (2020), mid (2050), and late (2080) periods of the twenty-first century past each RCP scenario were illustrated in Effigy 3. The northern boundary of albacore preferred habitat (defined every bit CPUE > 25 number/one,000 hooks) in the western Pacific was projected to shift southward from 20°S in 2020 to 22°S in 2050 and 25°South in 2080 for the RCP 4.five. A like design of southward shifting was observed for the RCP eight.5, with a wider spatial extent compared to the RCP 4.5. More specifically, the northern boundary has shown a like degree of shifting as RCP 4.v, while the southern boundary in the Tasman Sea has shifted from 40°Due south in 2020 to 43°South in 2050 and 45°S in 2080.

Figure three. The hereafter ensemble projections of south Pacific albacore relative abundance (number/1,000 hooks) distributions in quarter three estimated from the 5 atmosphere-sea general circulation models (AOGCMs) for the current (2020), mid (2050), and late (2080) periods of the twenty-first century under the (A) RCP 4.5 and (B) RCP viii.five (right panel) scenarios. The solid black lines announce the preferred habitat boundaries defined past relative abundance > 25 number/1,000 hooks.

The spatial distributions of the anomaly of relative abundance in quarter 3 past 2020, 2050, and 2080 under two RCP scenarios were shown in Figure 4. The potential albacore habitats increased in the latitudes of thirty–40°S simply decreased in xx–30°S of the western Pacific in 2020, 2050, and 2080 for the RCP 4.v, that the pattern of the gains and losses of potential habitat has become more than apparent overtime. A similar pattern was observed for the RCP 8.5, notwithstanding, the potential albacore habitats increased in the higher latitudes south of 40 °S in 2080 in the western Pacific, specially in the Tasman Ocean, and in the eastern Pacific (east of 110 °W) during 2020–2080 compared to the RCP 4.5.

Figure 4. The spatial distributions of the bibelot of relative abundance for south Pacific albacore in quarter 3 calculated by subtracting the predictions of the median ensemble GAM projections in 2020, 2050, and 2080 from the contempo boilerplate of 2012–2016 under the (A) RCP 4.5 and (B) RCP 8.5 scenarios.

Ensemble understanding maps for time to come albacore distribution in quarter 3 over five AOGCMs nether ii RCP scenarios were shown in Figure 5. For the RCP 4.v, the cadre habitats in western Pacific (crimson areas in Effigy 5A) accept shifted s slightly during 2020–2080. The habitat ranges (blue areas) besides shifted s with an increase of 15% past 2050 and sixteen% by 2080 in the area south of thirty°S relative to 2020, respectively. A similar southward displacement was also observed under the RCP 8.five (Figure 5B), the latitudinal boundaries of core habitats showed an obvious southward shift in the Tasman Body of water from 25 to 32°South in 2020 to 27–36°S in 2050 and to 32–39°S in 2080, respectively, and the habitat ranges in the area south of 30 °S increased xv% in 2050 and 21% in 2080.

Figure 5. Ensemble agreement maps of the south Pacific albacore habitat predictions by v atmosphere-body of water general circulation models (AOGCMs) for the current (2020), mid (2050), and late (2080) periods of the 20-first century under the (A) RCP 4.5 and (B) RCP 8.v (right panel) scenarios. Colors announce the probabilities of habitat map based on the levels of agreement amidst five AOGCMs.

Albacore Relative Abundance Changes in Exclusive Economical Zones

The results of the projected changes in albacore relative affluence during the main fishing flavor within the EEZs in 2020, 2050, and 2080 relative to 2012–2016 under the RCP 4.v and viii.5 scenarios were shown in Effigy 6. Under the RCP four.5, albacore relative abundance is projected to decrease in virtually EEZs overtime (by almost 25% in 2080), but was projected to slightly increase in 2020 and and so decreased afterward 2050 for the EEZs of Republic of kiribati (KI), Tuvalu (TV), and Tokelau (TK) (Figure 6A). The EEZ of New Caledonia (NC) has the greatest projected depletion in relative abundance past 2080 (43%). The relative abundance is projected to increase by 13 and 21% by 2080 in the EEZs of New Zealand (NZ) and Norfolk Island (NK), respectively. Similar patterns of relative abundance decline in most EEZs overtime were plant for the RCP 8.5 (Figure 6B). A more than significant negative and positive bear upon on the alter of albacore relative abundance in 2080 was found for the EEZs of NC and both NZ and NK, respectively. The decline of relative affluence in 2080 for the EEZ of Commonwealth of australia (AUS) was slightly improved in RCP eight.five (–2%) compared to RCP iv.v (–11%).

Figure 6. Changes in albacore relative abundance (number/1,000 hooks) within the exclusive economical zones (EEZs) of the countries or territories in the s Pacific Ocean for the current (2020), mid (2050), and late (2080) periods of the twenty-first century relative to 2012–2016 under the (A) RCP 4.5 and (B) RCP viii.five scenarios. Simply those countries or territories that had more than than 30% of fishery information within their EEZs were analyzed in this written report. Countries or territories were ordered per mean latitude of the EEZ. The vertical dashed lines denote the reference line of ± 5 in anomaly of relative abundance. State codes not shown in Supplementary Table 1 were illustrated beneath. EC, Galapagos Islands; TK, Tokelau; NU, Niue; PC, Pitcairn; NK, Norfolk Island.

Discussion

In this study, we adult an ensemble forecast framework to evaluate the projected future potential distribution of south Pacific albacore under the scenarios of intermediate and high degrees of ocean warming (RCP 4.5 and RCP 8.v). The northern purlieus of albacore preferred habitat is expected to shift due south by about five° latitude and the relative abundance is expected to gradually increase s of 30°S from 2020 to 2080 for both RCP scenarios, especially with a higher caste of change for the RCP 8.5. This study highlights that investigating the future spatio-temporal patterns of the potential albacore habitats could lend significant implications on the availability of tuna resources to the fishery and subsequent evaluation of tuna fishery management options nether climate alter.

The results of this report may be considered every bit reliable predictions of the south Pacific albacore because the information used has contained a substantial amount (>xc%) of the albacore catch and included all fleets covered a broad geographical range of the convention areas of the WCPFC and IATTC. Erauskin-Extramiana et al. (2019) only used the Japanese longline fishery information, constrained in the areas of xx°South northward and the waters off the eastern declension of Australia, which may not be informative to narrate the habitat preferences of albacore. Although the raw CPUE variability among diverse flags has been addressed in the GAM, the results with catchability adjustment can be biased due to the time-variant changes in fishing processes (eastward.g., gear saturation, fishing power, and line-fishing behavior). For instance, the targeting tactic past Taiwanese distant water longline fishery has changed over time from only targeting albacore to both tropical tuna (mainly bigeye and yellowfin) and albacore (Chang et al., 2011). We recommend that the future albacore distribution model should further consider the possible changes in catchability to adjust the potential confounding effects on the estimation of relative affluence based on the detailed set-by-set longline data (including a diverseness of operational variables).

The south Pacific longline fisheries tend to exploit the larger size of albacore (>80 cm fork length) while the juvenile albacore is mainly targeted by the troll fishery near the coastal expanse of New Zealand (Hare et al., 2020; Jordán et al., 2021; Vidal et al., 2021). The projected results of this written report may be considered as a proxy for the young adult and adult population. This highlights that using data from other types of line-fishing gear (i.e., troll fishery) may provide boosted information to better understand the ontogenetic habitat preferences of the south Pacific albacore in future analysis.

The GAM used in this report was shown to robustly fit the data (deviance explained = 62%) considering the residuals (in log-space) for the lognormal mistake distribution appear normal. Cross-validation testing as well suggested that the GAM performed well without overfitting. However, quarter 1, non represents the master fishing season of albacore, had a relatively poor model functioning compared to other quarters. The albacore may prefer a diversity of habitat weather condition in quarter ane considering some spawning adults in the 2d stage of spawning season are nowadays in the tropical habitat of warmer temperature, but a big portion returns to their favorable feeding zones driven by the local cues in feeding habitat (Senina et al., 2019). A farther extension of the GAM would be to gauge a seasonally variant habitat effect within the model, potentially improving the doubtfulness of seasonal predictions. Withal, this was exterior the scope of this study.

This study used a long-time series dataset (20 years of longline take hold of and try information) with a wide range of observations for quantifying the relationships between the environmental variables and albacore take hold of rate, which improves the reliability of albacore response curve. Although all the ecology variables included in this study showed statistically significant impacts on albacore'south grab rate, the spatial distribution of albacore may exist influenced by other factors in add-on to the ecology variables used in the GAM. A range of observed or satellite-based oceanographic and biological variables has also been used to describe albacore-environment associations, including meridional and zonal geostrophic currents, sea surface height, thermal fronts, eddies, meso-zooplankton biomass over 0–100 m (Zainuddin et al., 2006; Lan et al., 2012; Xu et al., 2013; Arrizabalaga et al., 2015). Nevertheless, one of the major issues to explore a broad range of oceanographic and biological variables for evaluating the time to come distribution shift is the availability of those variables from the future climate models. Furthermore, with a highly uncertain future surround, the predictions from AOGCMs could change in large amounts amongst each other compared to the historical predictions (e.g., CHL in this study). The above issues prevented further improvement for this written report.

This study indicated that DO100 is the predominant environmental predictor of the distribution of due south Pacific albacore because it contributed a substantial amount of explanatory power (22%) to the GAM. The albacore habitat was expected to subtract in the north of fifteen°S under the influence of a subtract in DO100 projected by the AOGCMs. The strong association between DO concentration and albacore distribution was confirmed by Lehodey et al. (2015) and Senina et al. (2018) through the sensitivity analysis in SEAPODYM that the hereafter distribution and abundance of albacore tuna is likely to significantly decrease and remain stable in the Due south Pacific Ocean with and without a projected decrease in Do, respectively. The empirical study also supports the sensitivity of DO that albacore is not tolerant to low DO concentration compared to other tunas (Brill, 1994). On the other hand, studies indicated that albacore is highly sensitive to the temperature changes in both their spawning and feeding environments (Williams et al., 2015; Reglero et al., 2017), which is in agreement with the GAM analysis of this report that SST is considered as the 2d most important environmental factor (6%). Furthermore, the pronounced spatial changes in potential futurity albacore habitats could be also explained past the substantial changes in projected SST under body of water warming scenarios. This emphasizes that accuracy and precision in ensemble forecasts of albacore tuna distribution are fundamentally linked to the performance of the AOGCMs in being able to realistically describe future changes in Practice and SST.

The s distribution shift and changes in habitat suitability of southward Pacific albacore identified in this study are consistent with the previous studies (Lehodey et al., 2015; Erauskin-Extramiana et al., 2019). However, the expected south shift plant past the above studies did non concur well with the report by Senina et al. (2018). This is due to the temperature spawning role implemented in SEAPODYM by Senina et al. (2018) tends to estimate a warmer favorable spawning habitat, resulted in hereafter projected spawning ground similar to the nowadays day. This indicated the importance of understanding the biological mechanism associated with shifts in albacore spawning ground under ocean warming. In addition, the distribution of albacore could exist strongly influenced by the changes in growth, reproduction, and survival rate, and therefore the relevant mechanisms should exist investigated in farther assay.

The uncertainty among the projections from the five AOGCMs and the two body of water warming scenarios that represent the plausible atmospheric forcings and emission pathways was taken into account through the ensemble analysis in this study. In addition to the uncertainty of climate projections, it would be desirable to explore other species distribution modeling approaches (e.k., Random forest, Maximum entropy, and Boosted regression tree models) to reduce the risk that structural inadequacies, inappropriate parameter specifications, or bias in each approach may unduly influence the concluding outputs (Robert et al., 2016; Georgian et al., 2019). Although ensemble forecasting could emphasize the "signal" that one is interested in emerges from the "noise" associated with individual model errors and uncertainties, the overall ensemble accurateness remains dependent on individual predictions (Araújo and New, 2007). Therefore, improve individual forecasts volition provide meliorate combined forecast. Despite these problems, the ensemble forecasting approach could essentially reduce the likelihood of making imitation management decisions based on predictions that are far from the truth.

In this report, the albacore abundance is projected to decrease in most EEZs by 2080 with the greatest depletion for New Caledonia, merely is projected to increase for New Zealand and Norfolk Island. The projections past Senina et al. (2018) indicated that the largest biomass increases occurred in the EEZs of Palau, Papua New Guinea, Federated States of Federated states of micronesia and Republic of nauru at the terminate of the century under the simulation scenario with projected changes in the Exercise concentration, but decreases in the EEZs of Republic of the fiji islands, New Caledonia, and Vanuatu. Erauskin-Extramiana et al. (2019) suggested that the due south Pacific albacore affluence would decrease in Republic of kiribati, Tuvalu, Tokelau, Cook Islands, and Australia, but an increase in New Zealand. The upshot of this written report more often than not supports the published evidence of the expected changes of albacore affluence in EEZs of PICTs except for the Palau, Federated States of Micronesia, and Republic of nauru. The reason why those EEZs was non identified is because the present study merely analyzed the countries or territories with more than 30% of the grid cells of fishery data inside the EEZs. Although the present study showed a subtract of affluence for Australia compared to the identified increase by Erauskin-Extramiana et al. (2019), Australia was establish to have the smallest decrease of affluence amongst all countries and territories that located in the report area under the RCP viii.5. Currently, the conservation management measure out (CMM-2015-2) for south Pacific albacore past the WCPFC has forbidden any increase in fishing vessels south of 20°S above 2005 levels (WCPFC, 2015). Under the current fishing effort constraint by the CMM, the futurity harvest catch of south Pacific albacore is expected to decrease because the potential albacore habitats are likely to increase in the latitudes of 30–40°South simply decreased in 20–30°S of the western Pacific. This suggests that the effectiveness of CMM to accomplish the fishery direction goals could be impacted by the futurity potential distribution shifting. We affirm not only the need merely besides the feasibility of incorporating approaches to address such shifts directly in the analysis of stocks and the management (east.1000., management strategies evaluation, Punt et al., 2016) based thereon, and then that the stock can be managed in a proactive and precautionary manner.

Information Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Writer Contributions

Y-JC, JH, and P-KL contributed to the development of the concept, assay of the data, and writing of this manuscript. Chiliad-WL and West-PT provided the fisheries data and futurity environmental data and contributed to writing those sections of the manuscript related to this. All authors read and agreed to the published version of the manuscript.

Funding

This study was financially supported past the Ministry of Science and Technology and the Fishery Agency of Council of Agriculture of Taiwan through the research grants MOST109-2611-M-002-013- and 110AS-vi.ane.1-FA-F1.

Disharmonize of Interest

The authors declare that the inquiry was conducted in the absence of any commercial or financial relationships that could be construed every bit a potential conflict of interest.

Publisher'southward Note

All claims expressed in this article are solely those of the authors and practice not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any production that may be evaluated in this article, or merits that may be fabricated by its manufacturer, is non guaranteed or endorsed by the publisher.

Acknowledgments

We are grateful to the Overseas Fisheries Evolution Quango (OFDC) of Taiwan for providing information from the Taiwanese longline fishery. Nosotros also thank the Western and Central Pacific Fisheries Commission and Inter-American Tropical Tuna Commission for providing public fishery information of South Pacific albacore. Nosotros thank Shui-Kai Chang and Chiee-Young Chen who have both contributed to discussions regarding the species distribution modeling approach and ensemble forecast.

Supplementary Material

The Supplementary Cloth for this article tin can be found online at: https://world wide web.frontiersin.org/manufactures/10.3389/fmars.2021.731950/full#supplementary-fabric

Footnotes

1. ^ https://www.wcpfc.int/public-domain
2. ^ https://world wide web.iattc.org/PublicDomainData/IATTC-Catch-by-species1.htm
3. ^ http://www.ofdc.org.tw/
4. ^ coastwatch.pfeg.noaa.gov/erddap/griddap/index.html
5. ^ http://marine.copernicus.eu/
6. ^ https://esgf-node.llnl.gov/projects/esgf-llnl/

References

Alabia, I. D., Saitoh, S. I., Igarashi, H., Ishikawa, Y., Usui, N., Kamachi, M., et al. (2016). Hereafter projected impacts of ocean warming to potential squid habitat in western and cardinal North Pacific. ICES J. Mar. Sci. 75, 1343–1356. doi: ten.1093/icesjms/fsv203

CrossRef Total Text | Google Scholar

Araújo, K. B., Pearson, R. G., Thuiller, W., and Erhard, Chiliad. (2005). Validation of species–climate impact models under climatic change. Glob. Change Biol. Bioener. xi, 1504–1513.

Google Scholar

Arrizabalaga, H., Dufour, F., Kell, Fifty., Merino, Chiliad., Ibaibarriaga, L., Chust, M., et al. (2015). Global habitat preferences of commercially valuable tuna. Deep Sea Res. Part Two Top. 113, 102–112. doi: ten.1016/j.dsr2.2014.07.001

CrossRef Total Text | Google Scholar

Barton, G. (2016). Bundle "MuMIn": Multi-Model Inference. R package, Version 1.15. 6. Vienna: R Cadre Team.

Google Scholar

Beaumont, L. J., Hughes, L., and Pitman, A. J. (2008). Why is the choice of futurity climate scenarios for species distribution modelling important? Ecol. Lett. 11, 1135–1146.

Google Scholar

Bong, J. D., Reid, C., Batty, Grand. J., Lehodey, P., Rodwell, 50., Hobday, A. J., et al. (2013). Effects of climate change on oceanic fisheries in the tropical Pacific: implications for economic development and food security. Clim. Modify 119, 199–212. doi: x.1007/s10584-012-0606-two

CrossRef Total Text | Google Scholar

Brewer, One thousand. J., Butler, A., and Cooksley, S. Fifty. (2016). The relative operation of AIC, AICC and BIC in the presence of unobserved heterogeneity. Methods Ecol. Evol. 7, 679–692. doi: 10.1111/2041-210X.12541

CrossRef Full Text | Google Scholar

Briand, K., Molony, B., and Lehodey, P. (2011). A report on the variability of albacore (Thunnus alalunga) longline take hold of rates in the southwest Pacific Body of water. Fish. Oceanogr. 20, 517–529. doi: 10.1111/j.1365-2419.2011.00599.x

CrossRef Full Text | Google Scholar

Brill, R. W. (1994). A review of temperature and oxygen tolerance studies of tunas pertinent to fisheries oceanography, motility models and stock assessments. Fish. Oceanogr. iii, 204–216. doi: x.1111/j.1365-2419.1994.tb00098.10

CrossRef Full Text | Google Scholar

Brouwer, S., Pilling, M., and Williams, P. (2018). "Trends in the South Pacific Albacore Longline and Troll Fisheries," in Working paper WCPFC-SC14-2018/SA-IP-08 (Rev. 2) presented to the Fourteenth Meeting of the Scientific Commission of the Western and Cardinal Pacific Fisheries Commission, Majuro, Marshall islands, 6-xiv August 2014, (Kolonia: Western and Primal Pacific Fisheries Commission).

Google Scholar

Bruge, A., Alvarez, P., Fontán, A., Cotano, U., and Chust, K. (2016). Thermal niche tracking and futurity distribution of Atlantic mackerel spawning in response to ocean warming. Front end. Mar. Sci. three:86. doi: x.3389/fmars.2016.00086

CrossRef Full Text | Google Scholar

Campbell, R. A., Tuck, G., Tsuji, Southward., and Nishida, T. (1996). "Indices of abundance for southern bluefin tuna from analysis of fine-calibration take hold of and effort data," in Working Paper SBFWS/96/sixteen Presented at the 2nd CCSBT Scientific Meeting, Hobart, Australia, (Hobart,TAS: CCSBT).

Google Scholar

Chang, Due south. K., Hoyle, South., and Liu, H. I. (2011). Catch rate standardization for yellowfin tuna (Thunnus albacares) in Taiwan's distant-h2o longline fishery in the Western and Fundamental Pacific Ocean, with consideration of target change. Fish. Res. 107, 210–220. doi: 10.1016/j.fishres.2010.11.004

CrossRef Total Text | Google Scholar

Childers, J., Snyder, Southward., and Kohin, S. (2011). Migration and behavior of juvenile North Pacific albacore (Thunnus alalunga). Fish. Oceanogr. twenty, 157–173.

Google Scholar

Crimmins, Due south. M., Dobrowski, S. Z., and Mynsberge, A. R. (2013). Evaluating ensemble forecasts of plant species distributions nether climate change. Ecol. Modell. 266, 126–130. doi: 10.1016/j.ecolmodel.2013.07.006

CrossRef Total Text | Google Scholar

Diniz-Filho, J. A. F., Mauricio Bini, Fifty., Fernando Rangel, T., Loyola, R. D., Hof, C., Nogués-Bravo, D., et al. (2009). Partitioning and mapping uncertainties in ensembles of forecasts of species turnover nether climate alter. Ecography 32, 897–906. doi: 10.1111/j.1600-0587.2009.06196.x

CrossRef Total Text | Google Scholar

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., et al. (2013). Collinearity: a review of methods to bargain with it and a simulation study evaluating their performance. Ecography 36, 27–46. doi: ten.1111/j.1600-0587.2012.07348.x

CrossRef Full Text | Google Scholar

Erauskin-Extramiana, M., Arrizabalaga, H., Hobday, A. J., Cabré, A., Ibaibarriaga, L., Arregui, I., et al. (2019). Big-scale distribution of tuna species in a warming sea. Glob. Alter Biol. 25, 2043–2060. doi: 10.1111/gcb.14630

PubMed Abstruse | CrossRef Full Text | Google Scholar

FAO (2021). FishStatJ Manual. Version 4.01.five. Rome: Food And Agriculture Organization of the United Nations.

Google Scholar

Flanders Marine Constitute (2018). Maritime boundaries geodatabase Maritime boundaries and exclusive economic zones (200NM), version 10. Ostend: Flanders Marine Found.

Google Scholar

Ganachaud, A. S., Sen Gupta, A., Orr, J. C., Wijffels, South. E., Ridgway, K. R., Hemer, M. A., et al. (2011). "Observed and expected changes to the tropical Pacific Ocean," in Vulnerability of tropical pacific fisheries and aquaculture to climate change, eds J. Bong, J. E. Johnson, and A. J. Hobday (Noumea: Secretariat of the Pacific Community), 115–202.

Google Scholar

Georgian, S. E., Anderson, O. F., and Rowden, A. A. (2019). Ensemble habitat suitability modeling of vulnerable marine ecosystem indicator taxa to inform deep-sea fisheries management in the S Pacific Ocean. Fish. Res. 211, 256–274.

Google Scholar

Gloria, M. B. A., Daeschel, 1000. A., Chicken, C., and Hilderbrand, K. Due south. Jr. (1999). Histamine and other biogenic amines in albacore tuna. J. Aquat. Nutrient Prod. Technol. 8, 55–69.

Google Scholar

Hare, S., Williams, P., and Pilling, G. The Wcpfc Secretariat. (2020). "Trends in the South Pacific albacore longline and troll fisheries," in Working Paper WCPFC-SC16-2020/SA-IP-11 (Rev. 01) presented to the Sixteenth Coming together of the Scientific Committee of the Western and Key Pacific Fisheries Committee, Electronic Meeting, eleven-20 August 2020, (Kolonia: Western and Central Pacific Fisheries Commission).

Google Scholar

Hastie, T. J., and Tibshirani, R. J. (1990). Generalized additive models, Vol. 43. Florida, FL: CRC press.

Google Scholar

Hobday, A. J. (2010). Ensemble analysis of the future distribution of large pelagic fishes off Commonwealth of australia. Prog. Oceanogr. 86, 291–301. doi: x.1016/j.pocean.2010.04.023

CrossRef Full Text | Google Scholar

Howell, Due east. A., and Kobayashi, D. R. (2006). El Nino furnishings in the Palmyra Atoll region: oceanographic changes and bigeye tuna (Thunnus obesus) grab rate variability. Fish. Oceanogr. 15, 477–489. doi: x.1111/j.1365-2419.2005.00397.x

CrossRef Total Text | Google Scholar

Ichinokawa, G., and Brodziak, J. (2010). Using adaptive area stratification to standardize catch rates with application to North Pacific swordfish (Xiphias gladius). Fish. Res. 106, 249–260. doi: 10.1016/j.fishres.2010.08.001

CrossRef Total Text | Google Scholar

Jordán, C. C., Hampton, J., Ducharme-Barth, N., Xu, H., Vidal, T., Williams, P., et al. (2021). "Stock assessment of Southward Pacific albacore tuna," in Working Paper WCPFC-SC17-2021/SA-WP-02 (Rev. 02) presented to the Seventeenth Meeting of the Scientific Commission of the Western and Cardinal Pacific Fisheries Commission. Online Meeting, 11-19 August 2021, (Kolonia: Western and Central Pacific Fisheries Committee).

Google Scholar

Kwon, Y., Larsen, C. P., and Lee, M. (2018). Tree species richness predicted using a spatial ecology model including forest area and frost frequency, eastern USA. PLoS I xiii:e0203881. doi: x.1371/journal.pone.0203881

PubMed Abstract | CrossRef Total Text | Google Scholar

Lan, K. W., Kawamura, H., Lee, M. A., Lu, H. J., Shimada, T., Hosoda, One thousand., et al. (2012). Relationship betwixt albacore (Thunnus alalunga) fishing grounds in the Indian Ocean and the thermal surround revealed by cloud-gratis microwave sea surface temperature. Fish. Res. 113, 1–seven. doi: 10.1016/j.fishres.2011.08.017

CrossRef Full Text | Google Scholar

Lan, Chiliad. Westward., Lee, Thousand. A., Chou, C. P., and Vayghan, A. H. (2018). Clan betwixt the interannual variation in the oceanic environment and catch rates of bigeye tuna (Thunnus obesus) in the Atlantic Ocean. Fish. Oceanogr. 27, 395–407.

Google Scholar

Langley, A., and Hampton, J. (2005). "Stock assessment of albacore tuna in the due south Pacific Bounding main," in Working Newspaper WCPFC-SC01-2005/SA-WP03 presented to the First Meeting of the Scientific Committee of the Western and Central Pacific Fisheries Commission. Noumea, New Caledonia, 8-xv August 2005, (Kolonia: Western and Cardinal Pacific Fisheries Commission).

Google Scholar

Lehodey, P., Senina, I., and Murtugudde, R. (2008). A spatial ecosystem and populations dynamics model (SEAPODYM)-Modeling of tuna and tuna-like populations. Prog. Oceanogr. 78, 304–318. doi: 10.1111/fog.12259

CrossRef Full Text | Google Scholar

Lehodey, P., Senina, I., Nicol, South., and Hampton, J. (2015). Modelling the impact of climatic change on South Pacific albacore tuna. Deep Ocean Res. Office 2 Top. 113, 246–259. doi: 10.1016/j.dsr2.2014.ten.028

CrossRef Full Text | Google Scholar

Lo, N. C. H., Jacobson, L. D., and Squire, J. L. (1992). Indices of relative abundance from fish spotter information based on delta-lognornial models. Can. J. Fish. Aquatic Sci. 49, 2515–2526. doi: ten.1139/f92-278

PubMed Abstruse | CrossRef Total Text | Google Scholar

Maunder, M. N., and Punt, A. E. (2004). Standardizing catch and try data: a review of contempo approaches. Fish. Res. lxx, 141–159. doi: 10.1016/j.fishres.2004.08.002

CrossRef Full Text | Google Scholar

Mugo, R., Saitoh, S. I., Nihira, A., and Kuroyama, T. (2010). Habitat characteristics of skipjack tuna (Katsuwonus pelamis) in the western North Pacific: a remote sensing perspective. Fish. Oceanogr. 19, 382–396. doi: 10.1111/j.1365-2419.2010.00552.x

CrossRef Total Text | Google Scholar

Novianto, D., and Susilo, E. (2016). Role of sub surface temperature, salinity and chlorophyll to albacore tuna abundance in Indian Ocean. Indones. Fish. Res. 22, 17–26. doi: 10.15578/ifrj.22.1.2016.17-26

CrossRef Full Text | Google Scholar

Pennington, M. (1983). Efficient estimators of affluence, for fish and plankton surveys. Biometrics 1983, 281–286. doi: 10.2307/2530830

CrossRef Total Text | Google Scholar

Pennington, M. (1996). Estimating the mean and variance from highly skewed marinedata. Fish. Balderdash. U Southward. 94, 489–450.

Google Scholar

Pickens, B. A., Carroll, R., Schirripa, M. J., Forrestal, F., Friedland, K. D., and Taylor, J. C. (2021). A systematic review of spatial habitat associations and modeling of marine fish distribution: A guide to predictors, methods, and knowledge gaps. PLoS One sixteen:e0251818. doi: 10.1371/journal.pone.0251818

PubMed Abstract | CrossRef Full Text | Google Scholar

Pilling, Thousand. M., Berger, A. One thousand., Reid, C., Harley, Due south. J., and Hampton, J. (2016). Candidate biological and economic target reference points for the Southward Pacific albacore longline fishery. Fish. Res. 174, 167–178. doi: x.1016/j.fishres.2015.09.018

CrossRef Full Text | Google Scholar

Porfirio, Fifty. Fifty., Harris, R. M., Lefroy, E. C., Hugh, S., Gould, S. F., Lee, Thousand., et al. (2014). Improving the use of species distribution models in conservation planning and management under climate change. PLoS Ane 9:e113749. doi: 10.1371/journal.pone.0113749

PubMed Abstruse | CrossRef Full Text | Google Scholar

Punt, A. Due east., Butterworth, D. Due south., de Moor, C. 50., De Oliveira, J. A., and Haddon, M. (2016). Management strategy evaluation: best practices. Fish Fish. 17, 303–334. doi: x.1111/faf.12104

CrossRef Total Text | Google Scholar

Reglero, P., Santos, Thousand., Balbín, R., Laíz-Carrión, R., Alvarez-Berastegui, D., Ciannelli, 50., et al. (2017). Environmental and biological characteristics of Atlantic bluefin tuna and albacore spawning habitats based on their egg distributions. Deep Bounding main Res. Pt. I 140, 105–116. doi: ten.1016/j.dsr2.2017.03.013

CrossRef Full Text | Google Scholar

Riahi, K., Rao, S., Krey, V., Cho, C., Chirkov, Five., Fischer, G., et al. (2011). RCP 8.5 - A scenario of comparatively high greenhouse gas emissions. Clim. Modify 109:33. doi: x.1007/s10584-011-0149-y

CrossRef Full Text | Google Scholar

Robert, K., Jones, D. O., Roberts, J. M., and Huvenne, V. A. (2016). Improving predictive mapping of deep-h2o habitats: Considering multiple model outputs and ensemble techniques. Deep Ocean Res. Pt. I. 113, lxxx–89. doi: ten.1016/j.dsr.2016.04.008

CrossRef Full Text | Google Scholar

Senina, I. N., Lehodey, P., Hampton, J., and Sibert, J. (2019). Quantitative modelling of the spatial dynamics of South Pacific and Atlantic albacore tuna populations. Deep Ocean Res. Pt. I. 2019:104667. doi: 10.1016/j.dsr2.2019.104667

CrossRef Total Text | Google Scholar

Senina, I., Lehodey, P., Calmettes, B., Dessert, G., Hampton, J., Smith, North., et al. (2018). "Impact of climate change on tropical Pacific tuna and their fisheries in Pacific Islands waters and high seas areas," in Working paper WCPFC-SC14-2018/EB-WP-01 presented to the Fourteenth Meeting of the Scientific Committee of the Western and Cardinal Pacific Fisheries Commission. Busan, Republic of Korea, 8-16 Baronial 2018, (Kolonia: Western and Central Pacific Fisheries Commission).

Google Scholar

Silva, C., Andrade, I., Yáñez, Eastward., Hormazabal, S., and Barbieri, M. Á, et al. (2016). Predicting habitat suitability and geographic distribution of anchovy (Engraulis ringens) due to climatic change in the coastal areas off Chile. Prog. Oceanogr. 146, 159–174. doi: 10.1016/j.pocean.2016.06.006

CrossRef Full Text | Google Scholar

Su, N. J., Sun, C. L., Punt, A. E., Yeh, S. Z., DiNardo, Grand., et al. (2013). An ensemble analysis to predict future habitats of striped marlin (Kajikia audax) in the Northward Pacific Body of water. ICES J. Mar. Sci. lxx, 1013–1022. doi: ten.1093/icesjms/fss191

CrossRef Full Text | Google Scholar

Thomson, A. 1000., Calvin, K. V., Smith, S. J., Kyle, Thousand. P., Volke, A., et al. (2011). RCP4.five: a pathway for stabilization of radiative forcing past 2100. Clim. Alter 109:77.

Google Scholar

Tremblay-Boyer, L., McKechnie, S., and Harley, S. J. (2015). "Standardized CPUE for south Pacific albacore tuna (Thunnus alalunga) from operational longline data," in Working Newspaper WCPFC-SC11-2015/SA-IP03 presented to the Eleventh Meeting of the Scientific Commission of the Western and Central Pacific Fisheries Commission. Pohnpei, Federated States of Micronesia, 5-thirteen Baronial 2015, (Kolonia: Western and Central Pacific Fisheries Commission).

Google Scholar

van Vuuren, D. P., Edmonds, J., Kainuma, Thousand., Riahi, K., Thomson, A., Hibbard, Grand., et al. (2011). The representative concentration pathways: an overview. Clim. Change 109:5. doi: 10.1007/s10584-011-0148-z

CrossRef Full Text | Google Scholar

Vidal, T., Jordán, C., Peatman, T., Ducharme-Barth, North., Xu, H., and Williams (2021). "Background assay and data inputs for the 2021 South Pacific albacore tuna stock cess," in Working paper WCPFC-SC17-SA-IP-03 presented to the Seventeenth Meeting of the Scientific Committee of the Western and Fundamental Pacific Fisheries Committee. Online Meeting, 11-19 August 2021, (Kolonia: Western and Cardinal Pacific Fisheries Commission).

Google Scholar

Villarino, Due east., Chust, Grand., Licandro, P., Butenschön, M., Ibaibarriaga, L., Larrañaga, A., et al. (2015). Modelling the time to come biogeography of North Atlantic zooplankton communities in response to climate change. Mar. Ecol. Prog. Ser. 531, 121–142. doi: ten.3354/meps11299

CrossRef Full Text | Google Scholar

WCPFC (2015). Summary Report of the twelfth Regular Session of the Commission for the Conservation and Direction of Highly Migratory Fish Stocks in the Western and Central Pacific Ocean; Kuta, Bali, Indonesia, 3-8. Dec, 2015. Kolonia: Western and Central Pacific Fisheries Committee.

Google Scholar

WCPFC (2018). Summary Written report of the Scientific Commission Fourteenth Regular Session; Busan, South korea, 8-xvi. Baronial, 2018. Kolonia: Western and Central Pacific Fisheries Commission.

Google Scholar

Williams, A. J., Allain, V., Nicol, S. J., Evans, K. J., Hoyle, South. D., et al. (2015). Vertical behavior and diet of albacore tuna (Thunnus alalunga) vary with breadth in the Southward Pacific Bounding main. Deep Sea Res. Pt. II. 113, 154–169. doi: ten.1016/j.dsr2.2014.03.010

CrossRef Full Text | Google Scholar

Forest, Due south. (2012). mgcv: Mixed GAM Computation Vehicle with GCV/AIC/REML smoothness estimation. R package version i.7-17. Vienna: R Core Team.

Google Scholar

Wood, S. (2017). Generalized Additive Models: an introduction with R, 2nd Edn. Florida: CRC press.

Google Scholar

Xu, Y., Teo, S. L., and Holmes, J. (2013). "Environmental Influences on Albacore Tuna (Thunnus alalunga) Distribution in the Coastal and Open up Oceans of the Northeast Pacific: Preliminary Results from Boosted Regression Trees Models," in Working Paper ISC/13/ALBWG-01/01 presented to the Thirteenth Meeting of the International Scientific commission on Tuna and Tuna-similar Species in the N Pacific Ocean. Shanghai, China, 19-26 March 2013, (Kolonia: Western and Central Pacific Fisheries Commission).

Google Scholar

Zainuddin, Grand., Kiyofuji, H., Saitoh, K., and Saitoh, South. I. (2006). Using multi-sensor satellite remote sensing and catch information to detect ocean hot spots for albacore (Thunnus alalunga) in the northwestern Due north Pacific. Deep Sea Res. Pt. 2 53, 419–431. doi: ten.1016/j.dsr2.2006.01.007

CrossRef Total Text | Google Scholar

bryanthavaily.blogspot.com

Source: https://www.frontiersin.org/articles/731950

Share this post