📈 Composition, stock status and trends

📦 Data: FAO Global Capture Production, release 2026.1.0 (published by FAO on 2026-03-30; catches 1950–2024), downloaded from Capture_2026.1.0.zip. Licence CC BY 4.0. fao_data_info() shows the release you use and how to cite it.

The summary analyses take any catch data: a fishing area from fishing_area_data(), or any subset of load_fao_data() (see 💾 FAO data). Their charts are saved in figures/png/summary/ and their tables in tables/, with the name of the area at the start of the file names (or label). Each function also returns its results invisibly, so they can be assigned and used in R.

FAO27 <- fishing_area_data("Atlantic, Northeast")

⚙️ Common arguments

Most functions of this article share these arguments; the tables of each function list only its own arguments.

Argument	Meaning	Default
area_data	FAO catch data: fishing_area_data()$area_data, or any subset of load_fao_data() (e.g. one country, or several fishing areas)	required
plot	Save the charts in figures/png/summary/	TRUE
table	Save the tables in tables/	TRUE
lang	Language of the chart text and names: "en", "zh" or "both"	NULL: the language of fao_settings() ("en" unless changed)
pptx	Also save the charts as editable PowerPoint files in figures/pptx/summary/	FALSE; fao_settings(pptx = TRUE) changes it for all functions
plot_show	Also draw each chart in the plot window (e.g. the Plots pane of RStudio) as it is saved	FALSE, or plot_show of fao_settings()
label	Name at the start of the output file names	NULL: the fishing area of the data (or its country; "selected_data" for several areas and countries)

🧺 Species group composition

species_group_composition() sums the catch by ISSCAAP species group (the FAO classification of species into about 50 groups, such as “Herrings, sardines, anchovies”) or by major group (fishes, crustaceans, molluscs, …). FAO gives no Chinese names for the groups; the Chinese names used are unofficial translations.

groups <- species_group_composition(FAO27$area_data)

The first chart shows the catch of each group, the second its share of the yearly catch. The top_n largest groups are shown separately, the others combined. The table gives the mean yearly catch of each decade, the catch of each of the last five years and the share of each group in the last year (tonnes and %):

show_table(groups$table, digits = 0)

Species group (ISSCAAP)	1950s	1960s	1970s	1980s	1990s	2000s	2010s	2020s	2020	2021	2022	2023	2024	share_2024_pct
Cods, hakes, haddocks	2,507,170	3,022,010	4,169,322	3,694,448	2,971,859	3,518,007	3,122,258	3,280,509	3,463,440	3,149,488	2,932,054	3,502,551	3,355,012	43
Herrings, sardines, anchovies	2,779,324	3,202,760	1,865,383	1,621,417	2,425,069	2,688,692	2,175,941	1,988,596	2,089,755	2,049,083	2,079,641	1,928,844	1,795,658	23
Miscellaneous pelagic fishes	269,410	890,080	3,173,777	2,661,330	2,371,030	1,735,251	1,761,012	1,454,800	1,176,739	1,441,312	1,857,547	1,654,906	1,143,495	15
Miscellaneous coastal fishes	77,690	186,400	546,076	833,692	989,272	505,487	337,299	252,807	483,374	266,377	197,125	193,064	124,098	2
Miscellaneous demersal fishes	285,220	263,060	345,996	424,319	405,580	382,355	358,843	328,769	329,019	333,933	315,636	314,796	350,461	4
Flounders, halibuts, soles	203,200	306,580	346,399	351,725	340,789	260,697	250,601	182,052	213,784	203,179	172,402	166,824	154,073	2
Brown seaweeds	141,882	249,340	282,269	279,593	316,267	262,672	242,414	335,848	263,397	272,022	518,245	349,189	276,387	4
Marine fishes not identified	252,710	246,840	248,921	153,214	87,960	33,651	24,212	4,129	4,912	8,694	4,501	2,200	340	0
Other groups	439,560	563,979	564,321	716,972	822,271	696,854	635,533	631,309	587,282	634,639	654,495	637,457	642,670	8

species_group_composition(FAO27$area_data, group = "major", lang = "zh")

Argument	Meaning	Default
group	"isscaap" (ISSCAAP species groups) or "major" (major groups)	"isscaap"
top_n	Number of groups shown separately; the others are combined	8
area_data, plot, table, lang, pptx, label	See Common arguments

Output: <label>_group_isscaap.png (or _group_major) and ..._share.png, and the table <label>_group_isscaap.csv; the function returns yearly (catch by group and year) and table.

🚦 Stock status plots

stock_status() classifies the catch series of each species every year, by its catch relative to the peak catch of the series (Kleisner et al. 2013):

Status	Rule
Developing	before the peak year, catch at most 50% of the peak
Exploited	catch above 50% of the peak
Over-exploited	after the peak, catch between 10% and 50% of the peak
Collapsed	after the peak, catch below 10% of the peak
Rebuilding	after a collapse, catch back between 10% and 50% of the peak

status <- stock_status(FAO27$area_data)

The first chart is the share of stocks in each status, the second the share of the catch. A “stock” here is one species item in the data (all countries together), and years after the last reported catch count as zero catch. Only series that meet the inclusion criteria are classified:

Argument	Meaning	Default
taxon_levels	Taxonomic levels of the items included: any of "species", "genus", "family" and "higher" (pooled groups such as “Marine fishes nei” are "higher")	c("species", "genus", "family")
min_years	Minimum number of years from the first to the last year with catch	10
min_consecutive	Minimum number of consecutive years with catch	5
min_catch	Minimum total catch of a series (tonnes)	1000
smooth	Smooth the catch with a 3-year running mean before classifying	FALSE
area_data, plot, table, lang, pptx, label	See Common arguments

The table of stocks gives the peak year and the status of each stock in the last year:

show_table(status$stocks, digits = 0)

species_name	species_en	first_year	peak_year	peak_tonnes	last_year_tonnes	pct_of_peak	status_last_year
Capelin	Capelin	1950	1977	3,773,808	168693.267	4	Collapsed
Atlantic herring	Atlantic herring	1950	1965	3,736,100	1250851.116	33	Over-exploited
Blue whiting(=Poutassou)	Blue whiting(=Poutassou)	1950	2004	2,419,291	1832649.131	76	Exploited
Atlantic cod	Atlantic cod	1950	1969	2,147,600	762408.034	36	Over-exploited
Atlantic mackerel	Atlantic mackerel	1950	2014	1,382,490	904016.963	65	Exploited
Sandeels(=Sandlances) nei	Sandeels(=Sandlances) nei	1954	1997	1,242,671	96630.953	8	Collapsed
European sprat	European sprat	1950	1975	982,776	343221.482	35	Over-exploited
Haddock	Haddock	1950	1969	889,200	293250.936	33	Over-exploited
Norway pout	Norway pout	1959	1974	877,910	5372.615	1	Collapsed
Saithe(=Pollock)	Saithe(=Pollock)	1950	1974	723,652	299652.086	41	Over-exploited

table(status$stocks$status_last_year)
#> 
#>      Collapsed      Exploited Over-exploited     Rebuilding 
#>            122             91             85             38

# Smoothed catches, stricter inclusion
stock_status(FAO27$area_data, smooth = TRUE, min_catch = 10000, label = "FAO27 smoothed")

Output: <label>_stock_status_stocks.png and <label>_stock_status_catch.png, and the tables <label>_stock_status.csv (stocks and catch by status and year) and <label>_stock_status_stocks.csv; the function returns yearly, stocks and status (the status of each stock in each year).

Stock status plots use catch data only. A falling catch can also come from management, markets or reporting, so the results are indicative and do not replace stock assessments.

⛰️ Peak and trend

peak_trend() finds, for the total and for the main species (or countries, or areas), the year and size of the peak catch, the catch in the last year as a percentage of the peak, and the trend over the last recent_years years:

• growth rate (growth_pct_per_year): the average yearly growth between the first and the last of these years, $100\,[(C_n/C_1)^{1/(n-1)} - 1]$, with $C_1$ and $C_n$ the catch of the first and the last of the $n$ years;
• trend: a Mann-Kendall test (Mann 1945; Kendall 1975), Kendall’s τ between year and catch. Catch series are autocorrelated, which makes the usual test find too many trends, so its variance is corrected with the effective sample size of an AR(1) process (Yue and Wang 2004), from the lag-1 autocorrelation of the detrended series corrected for its small-sample bias (Kendall 1954; lag1_autocorrelation). p_value is the corrected p value and p_value_iid that of the usual test. trend is "up" or "down" when p_value < 0.05, otherwise "none". Over ten years the correction is only partial: strongly autocorrelated series still show trends too often.
• many series: among ten or more series, some show a trend by chance. q_value adjusts the p values for the number of series in the result, the total excepted (Benjamini and Hochberg 1995). Read trends with p_value < 0.05 as exploratory and give as significant those with q_value < 0.05; 📝 the report does this for its main species and countries.

species_trend <- peak_trend(FAO27$area_data, by = "species", top_n = 10)

show_table(species_trend[, c("name", "peak_year", "pct_of_peak", "growth_pct_per_year", "kendall_tau", "p_value", "q_value", "trend")])

	name	peak_year	pct_of_peak	growth_pct_per_year	kendall_tau	p_value	q_value	trend
p	Total	1976	59.53	-2.0050033	-0.33	0.21	NA	none
p1	Atlantic herring	1965	33.48	-0.4745389	-0.20	0.71	0.95	none
p2	Atlantic cod	1969	35.50	-5.2914437	-0.82	0.03	0.16	down
p3	Capelin	1977	4.47	-13.1229391	-0.02	1.00	1.00	none
p4	Blue whiting(=Poutassou)	2004	75.75	2.9400918	0.16	0.74	0.95	none
p5	Atlantic mackerel	2014	65.39	-3.1263873	-0.38	0.33	0.83	none
p6	Sandeels(=Sandlances) nei	1997	7.78	-12.5836285	-0.47	0.07	0.25	none
p7	European sprat	1975	34.92	-5.7343149	-0.82	0.00	0.01	down
p8	Saithe(=Pollock)	1974	41.41	0.4885031	0.11	0.84	0.95	none
p9	Haddock	1969	32.98	0.3336638	0.07	0.86	0.95	none

peak_trend(FAO27$area_data, by = "country", top_n = 15, recent_years = 20, lang = "zh")

Argument	Meaning	Default
by	What to compare: "species", "country" or "area"	"species"
top_n	Number of series with the largest total catch (Inf for all)	10
recent_years	Number of most recent years used for the trend	10
area_data, plot, table, lang, pptx, label	See Common arguments

Output: <label>_peak_trend_species.png (or _country, _area) and the table with the same name; the function returns a data frame with one row per series, the total first.

🌍 All FAO marine areas

area_overview() compares all FAO marine fishing areas: the catch of each area over time with its peak, and the peak and trend chart of peak_trend(by = "area") for the areas:

overview <- area_overview()

show_table(overview[, c("name", "peak_year", "peak_tonnes", "last_year_tonnes", "pct_of_peak", "trend")], n = 20, digits = 0)

	name	peak_year	peak_tonnes	last_year_tonnes	pct_of_peak	trend
p	Total	1996	87,687,040	80930379.329	92	none
p1	Pacific, Northwest	1997	24,020,927	19067991.408	79	down
p2	Atlantic, Northeast	1976	13,172,602	7842194.162	60	none
p3	Pacific, Southeast	1994	20,422,424	9453525.210	46	none
p4	Pacific, Western Central	2024	14,105,189	14105189.055	100	up
p5	Indian Ocean, Eastern	2017	6,910,165	6401679.921	93	none
p6	Atlantic, Eastern Central	2018	5,520,055	5099475.194	92	none
p7	Indian Ocean, Western	2024	6,223,934	6223934.387	100	none
p8	Atlantic, Northwest	1968	4,612,569	1407262.000	31	down
p9	Pacific, Northeast	1987	3,401,535	2551440.501	75	down
p10	Atlantic, Southeast	1978	3,299,221	1475796.444	45	none
p11	Atlantic, Southwest	1997	2,664,636	1970436.942	74	none
p12	Atlantic, Western Central	1984	2,505,870	1077647.351	43	none
p13	Pacific, Eastern Central	2024	2,349,505	2349505.260	100	up
p14	Mediterranean and Black Sea	1986	2,046,352	1026111.034	50	none
p15	Pacific, Southwest	1991	929,232	381352.879	41	down
p16	Atlantic, Antarctic	1987	510,151	466706.916	91	up
p17	Indian Ocean, Antarctic	1972	189,586	10861.059	6	none
p18	Pacific, Antarctic	1983	10,678	2665.606	25	none
p19	Arctic Sea	2023	20,094	16604.000	83	none

Argument	Meaning	Default
recent_years	Number of most recent years used for the trend	10
data_version	FAO data release to use (see load_fao_data())	"latest"
plot, table, lang, pptx	See Common arguments

area_overview() reads the FAO data itself, so it has no area_data or label. Output: marine_areas_catch.png, marine_areas_peak_trend_area.png and the table marine_areas_peak_trend_area.csv; the function returns the data frame of peak_trend().

🔀 Synchronous breakpoints

regime_shift() runs a breakpoint analysis on the catch series of the main species (all species items, including “nei” groups) and counts how many species change their mean catch in each year. Each series is split where its mean changes (strucchange::breakpoints(), Bai and Perron 2003; Zeileis et al. 2002, 2003), with the number of breakpoints chosen by BIC; the year of a breakpoint is the first year of the new period. Years when many species change together may point to an ecosystem-wide regime shift, or to changes in management or reporting.

shifts <- regime_shift(FAO27$area_data, top_n = 20)

show_table(shifts$breakpoints[, c("species", "first_year", "n_breakpoints", "breakpoint_years")], n = 20)

species	first_year	n_breakpoints	breakpoint_years
Atlantic herring	1950	3	1969, 1994, 2011
Atlantic cod	1950	5	1967, 1978, 1989, 2000, 2011
Capelin	1950	3	1970, 1986, 2004
Blue whiting(=Poutassou)	1950	3	1978, 1998, 2009
Atlantic mackerel	1950	2	1966, 2010
Sandeels(=Sandlances) nei	1954	3	1974, 1986, 2003
European sprat	1950	5	1961, 1972, 1983, 1994, 2011
Saithe(=Pollock)	1950	3	1966, 1978, 2011
Haddock	1950	2	1964, 1977
Norway pout	1959	3	1968, 1985, 1998
Brown seaweeds	1950	3	1961, 2000, 2014
European pilchard(=Sardine)	1950	2	1996, 2012
Atlantic redfishes nei	1950	4	1961, 1975, 1987, 2000
Atlantic horse mackerel	1950	4	1970, 1988, 2000, 2014
European plaice	1950	3	1961, 1995, 2006
Whiting	1950	3	1963, 1985, 1998
Finfishes nei	1950	2	1972, 1986
European hake	1950	4	1966, 1977, 1993, 2011
Jack and horse mackerels nei	1950	5	1968, 1979, 1991, 2002, 2013
Northern prawn	1950	3	1971, 1982, 2003

head(shifts$yearly[order(-shifts$yearly$n_species), ], 5)
#>    Year n_species
#> 62 2011         5
#> 12 1961         4
#> 51 2000         4
#> 17 1966         3
#> 29 1978         3

Argument	Meaning	Default
top_n	Number of species with the largest total catch that are analysed	20
h_value	Minimum length of a period between breakpoints: a fraction of the series length or a number of years	0.15
area_data, plot, table, lang, pptx, label	See Common arguments

Species whose series is too short for h_value are skipped with a message. Output: <label>_regime_shift.png (number of species with a breakpoint in each year), <label>_regime_shift_species.png (breakpoints of each species), and the tables <label>_regime_shift.csv and <label>_regime_shift_years.csv; the function returns breakpoints and yearly.

🐟 Pelagic/demersal ratio

pelagic_demersal() divides the catch of pelagic fishes by the catch of demersal fishes (P/D) in each FAO marine area, by year and by decade. Caddy (1993) proposed that nutrient enrichment favours short-lived plankton-feeding pelagic fishes over demersal fishes, and de Leiva Moreno et al. (2000) used P/D as a proxy for the nutrient level of the seas around Europe. The ratio of a decade is the mean of its yearly ratios (as Hondorp et al. 2010), with the lowest and the highest.

pd <- pelagic_demersal(FAO27$area_data)

show_table(pd$decades[, c("period", "pd_mean", "pd_min", "pd_max", "counted_pct")])

period	pd_mean	pd_min	pd_max	counted_pct
1950s	0.88	0.80	1.02	85.66
1960s	0.85	0.35	1.18	79.60
1970s	0.34	0.26	0.43	64.27
1980s	0.29	0.24	0.34	66.68
1990s	0.48	0.34	0.59	68.98
2000s	0.56	0.44	0.78	75.38
2010s	0.52	0.43	0.67	72.70
2020s	0.46	0.42	0.54	74.33

show_table(pd$groups[, c("group", "counted_as", "catch_pct")], n = 12)

group	counted_as	catch_pct
Cods, hakes, haddocks	demersal	34.17
Herrings, sardines, anchovies	pelagic	24.62
Miscellaneous pelagic fishes	not counted	18.84
Miscellaneous coastal fishes	demersal	5.00
Miscellaneous demersal fishes	demersal	3.65
Flounders, halibuts, soles	demersal	2.98
Brown seaweeds	not counted	2.69
Marine fishes not identified	not counted	1.46
Shrimps, prawns	demersal	1.11
Sharks, rays, chimaeras	not counted	0.87
Mussels	not counted	0.63
Clams, cockles, arkshells	not counted	0.54

The FAO data give each species an ISSCAAP species group, so the function counts groups. fao_pd_groups() shows how each group is counted by the two published definitions (it has no arguments):

definition	Pelagic	Demersal
"de_leiva_moreno" (default)	Herrings, sardines, anchovies (small pelagics)	Flounders, cods and hakes, miscellaneous coastal and demersal fishes, marine crustaceans except krill
"grainger_garcia" (Grainger and Garcia 1996)	Small pelagics, tunas and billfishes, miscellaneous pelagic fishes	Flounders, cods and hakes, miscellaneous coastal and demersal fishes, sharks and rays, unidentified marine fishes

show_table(fao_pd_groups(), n = 15)

isscaap_code	isscaap_en	de_leiva_moreno	grainger_garcia
31	Flounders, halibuts, soles	demersal	demersal
32	Cods, hakes, haddocks	demersal	demersal
33	Miscellaneous coastal fishes	demersal	demersal
34	Miscellaneous demersal fishes	demersal	demersal
35	Herrings, sardines, anchovies	pelagic	pelagic
36	Tunas, bonitos, billfishes	not counted	pelagic
37	Miscellaneous pelagic fishes	not counted	pelagic
38	Sharks, rays, chimaeras	not counted	demersal
39	Marine fishes not identified	not counted	demersal
42	Crabs, sea-spiders	demersal	not counted
43	Lobsters, spiny-rock lobsters	demersal	not counted
44	King crabs, squat-lobsters	demersal	not counted
45	Shrimps, prawns	demersal	not counted
46	Krill, planktonic crustaceans	not counted	not counted
47	Miscellaneous marine crustaceans	demersal	not counted

Your own groups can be given with pelagic = ... and demersal = .... The published studies classed each species rather than ISSCAAP groups, so their values (below 1 in oligotrophic seas, about 1 to 3.5 in mesotrophic and about 6 to 20 in eutrophic seas, in de Leiva Moreno et al. 2000) are only roughly comparable. P/D also changes with fishing and markets (demersal fishes are in higher demand; Caddy et al. 1998), hypoxia (Hondorp et al. 2010), the natural variability of small pelagics and reporting; see ?pelagic_demersal. In area 27, P/D stays below 1 (by decade, between 0.29 and 0.88) although pelagic species are among the main catches: blue whiting, a pelagic gadoid, is counted with the cods, hakes and haddocks, and Atlantic mackerel (miscellaneous pelagic fishes) is not counted by this definition, which shows how much the grouping matters.

Argument	Meaning	Default
definition	Groups counted as pelagic and demersal: "de_leiva_moreno" or "grainger_garcia"	"de_leiva_moreno"
pelagic, demersal	Your own ISSCAAP groups (English names as in the FAO data, e.g. "Herrings, sardines, anchovies") instead of those of definition	NULL
by	"area", or "division" for the catch by division of load_fao_divisions() (🧩 extension)	"area"
area_data, plot, table, lang, pptx, label	See Common arguments

Inland areas are left out with a message. Output: <label>_pelagic_demersal_ratio.png (P/D by year, one panel per area), <label>_pelagic_demersal.png and ..._share.png (pelagic, demersal and other catch), and the tables <label>_pelagic_demersal.csv, ..._decades.csv and ..._groups.csv; the function returns yearly, decades and groups.

🧩 By division (extension)

De Leiva Moreno et al. (2000) compared the divisions of the Mediterranean and Black Sea. The same units are available from the catch data of regional fishery bodies, loaded with load_fao_divisions() (an extension of the package, not part of the Word report; see 🚢 Trade and employment for the other datasets):

pelagic_demersal(load_fao_divisions("GFCM"), by = "division")

In release 2026.1.0, the mean P/D of the 1980s is below 1 in the Ionian (0.6) and the Levant (0.4), about 4 in the Adriatic and about 18 in the Black Sea, within the ranges de Leiva Moreno et al. (2000) found for oligotrophic, mesotrophic and eutrophic seas.

📝 In the Word report

fao_report() writes these analyses into the Word report of a fishing area (see 📝 Assessment report: Word, LaTeX and PDF). The ISSCAAP groups (groups = TRUE) and the stock status plots (status = TRUE) are sections by default; the pelagic/demersal ratio is added with pelagic = TRUE. The settings of a function are given with the section name as prefix, e.g. status_smooth = TRUE is smooth = TRUE of stock_status():

fao_report(27, pelagic = TRUE, status_smooth = TRUE, pelagic_definition = "grainger_garcia")

peak_trend() gives the trends in the tables of the main species and countries of the report; area_overview() and regime_shift() are not part of it.

📖 References

Bai, J. and Perron, P. (2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics 18: 1–22. doi:10.1002/jae.659

Benjamini, Y. and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B 57: 289–300. doi:10.1111/j.2517-6161.1995.tb02031.x

Caddy, J.F. (1993). Toward a comparative evaluation of human impacts on fishery ecosystems of enclosed and semi-enclosed seas. Reviews in Fisheries Science 1: 57–95. doi:10.1080/10641269309388535

Caddy, J.F., Carocci, F. and Coppola, S. (1998). Have peak fishery production levels been passed in continental shelf area? Some perspectives arising from historical trends in production per shelf area. Journal of Northwest Atlantic Fishery Science 23: 191–219. doi:10.2960/J.v23.a13

de Leiva Moreno, J.I., Agostini, V.N., Caddy, J.F. and Carocci, F. (2000). Is the pelagic-demersal ratio from fishery landings a useful proxy for nutrient availability? A preliminary data exploration for the semi-enclosed seas around Europe. ICES Journal of Marine Science 57: 1091–1102. doi:10.1006/jmsc.2000.0705

FAO (2026). Global Capture Production 1950–2024, release 2026.1.0. In: Fisheries and Aquaculture. Rome: FAO. https://www.fao.org/fishery/collection/capture/en. Licence: CC BY 4.0. Release file: https://www.fao.org/fishery/static/Data/Capture_2026.1.0.zip

Grainger, R.J.R. and Garcia, S.M. (1996). Chronicles of marine fishery landings (1950–1994): trend analysis and fisheries potential. FAO Fisheries Technical Paper 359. Rome: FAO. https://www.fao.org/4/w3244e/w3244e00.htm

Hondorp, D.W., Breitburg, D.L. and Davias, L.A. (2010). Eutrophication and fisheries: separating the effects of nitrogen loads and hypoxia on the pelagic-to-demersal ratio and other measures of landings composition. Marine and Coastal Fisheries 2: 339–361. doi:10.1577/C09-020.1

Kendall, M.G. (1954). Note on bias in the estimation of autocorrelation. Biometrika 41: 403–404. doi:10.1093/biomet/41.3-4.403

Kendall, M.G. (1975). Rank correlation methods, 4th edition. London: Charles Griffin.

Kleisner, K., Zeller, D., Froese, R. and Pauly, D. (2013). Using global catch data for inferences on the world’s marine fisheries. Fish and Fisheries 14: 293–311. doi:10.1111/j.1467-2979.2012.00469.x

Mann, H.B. (1945). Nonparametric tests against trend. Econometrica 13: 245–259. doi:10.2307/1907187

Yue, S. and Wang, C.Y. (2004). The Mann-Kendall test modified by effective sample size to detect trend in serially correlated hydrological series. Water Resources Management 18: 201–218. doi:10.1023/B:WARM.0000043140.61082.60

Zeileis, A., Leisch, F., Hornik, K. and Kleiber, C. (2002). strucchange: an R package for testing for structural change in linear regression models. Journal of Statistical Software 7(2): 1–38. doi:10.18637/jss.v007.i02

Zeileis, A., Kleiber, C., Krämer, W. and Hornik, K. (2003). Testing and dating of structural changes in practice. Computational Statistics & Data Analysis 44: 109–123. doi:10.1016/S0167-9473(03)00030-6