assembling real terrain

Paysage de Ndoki-Likouala · Congo

drag to orbit · scroll to zoom · tap any label or species for its storydrag to orbit · pinch to zoom · tap a label

Forest loss2001

2001Hansen/UMD/Google/USGS/NASA GFC 2024 v1.122024

Viewpoint 1.8°N · 16.9°E Sangha & Likouala · Republic of the Congo

Map key

The paper publishes densities per survey stratum and no locations. Elephant, gorilla, chimpanzee and the duiker guild are shown on the strata they were counted in — the elephant in three, because it is absent from the Lac Télé block of the paper’s own table. Every other animal is listed for the landscape as a whole. Nothing here places an animal inside a stratum, and none of these marks is a sighting — they are indicative of where a line was counted.

Nine survey stratathe landscape

Sangha Trinational neighbours

Priority speciestap a species ↗

African forest elephant

Western lowland gorilla

Central chimpanzee

Bongo

Sitatunga

Blue duiker

Grey-cheeked mangabey

African grey parrot

Dwarf crocodile

The index carries three lines, two of them guilds: the forest elephant on its own, the great apes combined, and the forest ungulates combined. These nine are the guild members and the candidates the paper weighed — tap one to see which.

Numbered roads (RN 2, RN 10)

Village / town

Altitude bands · 292–661 m across the strata

Highest ground540+

Upper interfluves460–540

Interfluves400–460

Low interfluves360–400

Basin floor330–360

Lowest ground292–330

Altitude, and barely that. 369 m of range across 426 km, measured by us from the elevation model clipped to the nine strata. Half the landscape sits within 70 m of its lowest ground, and the surface is the top of the canopy rather than the forest floor. What organises this landscape is water, not height, and no open dataset maps that boundary.

Live · forest condition

Active fire—24h

Deforestation—30d

NASA FIRMS · Global Forest Watch

10 km

Ndoki-Likouala

The strata · the record

Ndoki-Likouala landscape

Nine management units in the northern Republic of the Congo, surveyed together on one line-transect protocol since 2006. That ratio is high because this is not one boundary: it is nine, and three of them are broken into several parts.

Landscape: 34,000 km² stated, approx.
Strata, measured: 35,111 km² the nine drawn
Outlines: 8,144 km as digitised
Gazetted units: 2 of 9
Strata drawn: 9 18 rings
Elevation: 292–661 m measured

Two of the nine are protected areas — Nouabalé-Ndoki National Park (4,071 km²) and the Lac Télé Community Reserve (4,219 km²). Four are logging concessions, three are swamp and other strata.
Sangha Trinational — the national park is the Congolese leg of a UNESCO World Heritage property inscribed in 2012 with Dzanga-Ndoki (Central African Republic) and Lobéké (Cameroon), under an agreement among the three governments.
Monitored across the line since 2006 — elephant, gorilla and chimpanzee counted inside the national park and in the surrounding concessions simultaneously, in partnership with the Ministry of Forest Economy. Few landscapes anywhere can support that comparison.
The only one of seven that did not rise — the paper describes a slight decline overall and reports no statistically significant trend in the protected area. Inside the park, a negative point estimate with no significant trend is a finding of no detected change, not a measured decline.

What this landscape is for, in the paper

The disaggregation. Because the same protocol ran inside the park and in less protected units, the index can be split by protection status. The paper's own reading of that split is that the confidence intervals overlap, so no consistent difference could be established — the point estimate for the less protected forest is negative. Drawing those two lines without their intervals would manufacture a difference the paper declines to claim.

Three lines, and the paper asks for ten. The index here carries three lines, two of them guilds rather than species: the forest elephant on its own, the two great apes combined, and the forest ungulates combined. The paper's own sensitivity analysis found that indexes with few species are sensitive to the removal of any one of them, and it recommends at least ten. This is the smallest index in the set, and it is the landscape where that recommendation matters most.

Why the guilds exist. Not taxonomy — measurement. In dense forest the track and sign of the forest ungulate species cannot reliably be told apart, and neither can the sign of gorilla and chimpanzee. The paper says so in the caption of the table these figures come from, so this page says so too.

Areas and outlines. The nine strata are drawn from WCS Congo's own survey-strata file (2024-12-10), cleared for public display; each unit's area is that file's own figure and the nine sum to 34,966 km². Our independent measurement of the rings we draw gives 35,111 km², 0.4% above it. The landscape figure of 34,000 km² is the paper's, from its landscape description, and it carries no date, method or gazetting authority — it is stated as approximate for that reason. Perimeter and compactness are computed from the drawn rings and are scale-dependent. Elevation 292–661 m is our own measurement, from Copernicus DEM GLO-30 clipped to the nine strata (5,752,842 cells); it is a canopy surface, not a ground survey. OpenStreetMap relation 12416802 corroborates the national park's outline independently (bounding boxes agree to about 600 m) and is deliberately not drawn on top of it. ⚠ These polygons are SURVEY STRATA, not protected-area boundaries: seven of the nine have no protected status.

Connectivity · the record

How the strata and the parks connect

How the nine survey strata sit against each other and against the protected areas across the border, measured by us on the survey-strata file the paper's authors supplied and, for the neighbouring parks, on OpenStreetMap outlines. A shared boundary is an administrative fact. It is not evidence that animals cross it, and no corridor is drawn.

Park boundary: 329 km Nouabalé-Ndoki, measured
Shared with concessions: 141 km 43% of it
Along the CAR border: ≈20 km within 250 m of Dzanga-Ndoki
Strata: 9 of 9 one contiguous mosaic

The nine strata

Stratum	Name in the file	Designation (read from the name)	km² in the file	km² measured by us
UFA Mokabi	UFA Mokabi	logging concession (UFA)	2,870	2,870.0
Nouabalé-Ndoki National Park	PNNN	national park	4,071	4,070.9
UFA Loundoungou	UFA Loundoungou	logging concession (UFA)	5,455	5,453.2
UFA Kabo	UFA Kabo	logging concession (UFA)	2,773	2,773.4
UFA Pokola	UFA Pokola	logging concession (UFA)	4,797 ⚠ SM Table 3 gives 4,502 km²	4,797.8
Loundoungou Marécage	Loundoungou Marecage	swamp-forest stratum	3,028	3,026.3
Bailly Marécage	Bailly Marecage	swamp-forest stratum	5,477	5,474.2
Lac Télé Community Reserve	RC Lac Tele	community reserve	4,219 counted as less protected in the paper	4,216.1
Batanga	Batanga	survey stratum	2,276	2,273.6
Nine strata			34,966	34,955.5

NdokiLikouala_SurveyStrata, WCS Congo's survey-strata file dated 2024-12-10, supplied with the paper's manuscript. Its attribute table holds only an id, an area in hectares and km², a name and a link id; the designations above are read from the names, not stated in the file. Batanga carries no stated designation and is drawn neutral. Pokola's area disagrees between the paper's own sources: 4,797 km² in the strata file, 4,502 km² in SM Table 3. The page states both and does not choose; it is a question for the authors. The landscape total is stated four ways: 34,000 km² (SM Appendix A), 34,966 km² (the file's nine areas summed), about 34,955 km² (the file's hectare column) and 35,111 km² (our measurement of the drawn rings). An earlier account of the same landscape gives its management units as NNNP 4,190, Kabo 2,870, Pokola 4,510, Loundoungou 4,230, Toukoulaka 2,080, Mokabi 2,670, Bailly/Bodingo swamps 3,770 and Lac Télé 4,380 km², as of 2006, with logging beginning in 1968 at Kabo and Pokola, 2000 at Mokabi and 2005 at Loundoungou [Stokes et al. 2010].

Where the strata meet

Unit	Unit	km within 250 m from A	from B	Nearest, km
UFA Mokabi	Nouabalé-Ndoki National Park	61.5	61.5	0.00
UFA Mokabi	UFA Loundoungou	15.8	16.3	0.00
UFA Mokabi	Réserve Spéciale de Dzanga-Sangha	0.0	1.3	0.04
Nouabalé-Ndoki National Park	UFA Loundoungou	64.0	64.4	0.00
Nouabalé-Ndoki National Park	UFA Kabo	15.3	16.5	0.01
Nouabalé-Ndoki National Park	Parc national de Dzanga-Ndoki	20.1	0.0	0.11
Nouabalé-Ndoki National Park	Réserve Spéciale de Dzanga-Sangha	15.0	20.4	0.19
UFA Loundoungou	UFA Kabo	50.7	50.5	0.00
UFA Loundoungou	UFA Pokola	145.2	145.3	0.00
UFA Loundoungou	Loundoungou Marécage	1431.2	1440.0	0.00
UFA Loundoungou	Bailly Marécage	713.2	705.3	0.00
UFA Loundoungou	Lac Télé Community Reserve	4.5	4.5	0.00
UFA Kabo	UFA Pokola	68.2	66.3	0.00
UFA Kabo	Parc National de Lobéké	0.0	0.0	0.25
UFA Kabo	Parc national de Dzanga-Ndoki	23.1	0.0	0.11
UFA Pokola	Bailly Marécage	51.0	50.8	0.00
Loundoungou Marécage	Lac Télé Community Reserve	20.7	20.9	0.00
Bailly Marécage	Lac Télé Community Reserve	183.5	186.1	0.00
Lac Télé Community Reserve	Batanga	88.5	88.4	0.00
Parc National de Lobéké	Parc national de Dzanga-Ndoki	10.8	9.4	0.06
Parc national de Dzanga-Ndoki	Réserve Spéciale de Dzanga-Sangha	33.0	33.0	0.00

Length of each outline lying within 250 m of the other, measured from both sides, with the minimum distance. Measured by us, tools/measure_strata_adjacency.py, 2026-09-30. Lobéké, Dzanga-Ndoki, Dzanga-Sangha and Ntokou-Pikounda are OpenStreetMap outlines, not the authors'; where one side reads 0.0 the two digitisations do not coincide within 250 m for long, and the figure is asymmetric for that reason. The park's 141 km with concessions: Loundoungou 64.0 km, Mokabi 61.5 km, Kabo 15.3 km.

Unit	Protected area across the line or beyond	Nearest, km
UFA Mokabi	Parc national de Dzanga-Ndoki	44.75
Nouabalé-Ndoki National Park	Parc National de Lobéké	17.77
UFA Loundoungou	Parc national de Dzanga-Ndoki	45.01
UFA Loundoungou	Réserve Spéciale de Dzanga-Sangha	36.04
UFA Loundoungou	Parc national de Ntokou-Pikounda	21.40
UFA Kabo	Réserve Spéciale de Dzanga-Sangha	19.28
UFA Pokola	Parc National de Lobéké	32.51
UFA Pokola	Parc national de Dzanga-Ndoki	43.42
UFA Pokola	Parc national de Ntokou-Pikounda	0.59
Bailly Marécage	Parc national de Ntokou-Pikounda	22.14

Gaps under 50 km between a stratum and a neighbouring park it does not touch, measured by us. Lobéké (Cameroon) is 17.8 km from Nouabalé-Ndoki. UFA Pokola comes within 0.6 km of Ntokou-Pikounda. The Lac Télé Community Reserve is 60.7 km from the national park and reached through the Loundoungou concession and the Loundoungou swamp stratum, which it touches.

What the literature says about movement here

Elephants range widely and avoid roads. GPS-collared forest elephants in the Ndoki wilderness had home ranges of 211–2,402 km² (95% kernel), inside a roadless area of 11,793 km². Across six sites in three countries, only 1 of 27 collared elephants crossed an unprotected road outside a protected area, and it crossed at 14 times its normal speed. Ndoki collars ran October 1998 to November 2001. [Blake et al. 2008]
The roadless area shrank. In the eight years after the collars went on, roads for extractive industry cut the Ndoki roadless area by more than 11,000 km², 88.9%. [Blake et al. 2008] Across the Congo Basin, the chance of finding elephants rose with distance from roads and was higher inside protected areas at every distance. [Blake et al. 2007]
Three regimes side by side. The landscape was designed with integral protection in the park, community management in Lac Télé and wildlife management in the surrounding concessions; the monitoring programme covered about 28,000 km². Units with anti-poaching held gorillas, chimpanzees and elephants at consistently higher abundance than a neighbouring concession with no wildlife management, and logging roads were used by elephant poachers even where there were controls. Where protected from poaching, elephants and gorillas used the herbaceous growth of recently logged forest; chimpanzees preferred mature forest. [Stokes et al. 2010, 2006 survey]
Unlogged forest as a source. In 1.2 million ha of concessions in northern Congo (749 transects, 3,450 km), mammal abundance was often related to distance from unlogged forest, and to distance from roads and villages. [Clark et al. 2009] In one concession surveyed before, during and after logging, chimpanzee nest encounter rates fell as human impact rose, while gorillas avoided active logging and roads but used recently logged areas rich in herbs. [Morgan et al. 2018]
Logging brings hunters. Industrial logging raised the population of the logging towns by 69% (2000–2006) and bushmeat supply by 64%, and immigrants hunted 72% of all bushmeat. [Poulsen et al. 2009; the 69% as given by Stokes et al. 2010]
The Sangha River. The river forming the international border inside the Sangha Trinational is a partial barrier to elephant movement between its parks; its crossing points were surveyed by dung count. [Weinbaum et al. 2007]
Duikers move little. Radio-collared blue duikers under hunting in northern Congo had home ranges averaging 5.5 ha, annual survival of 0.94 and an immature dispersal rate of 0.25; two dispersing young moved about 1.5 km. Blue duiker is one of the six species in the index's ungulate line. [Mockrin 2010]
Across three borders. The Sangha Trinational was established on 7 December 2000 by a cooperation agreement between Cameroon, the Central African Republic and the Republic of the Congo, followed by agreements on anti-poaching (28 June 2002) and free movement of its staff (4 February 2005). [Goué & Kana 2023] UNESCO describes its 1,787,950 ha buffer zone as a response to the ecological links between the property and its surroundings. [UNESCO, property 1380]
Two transboundary landscapes. The paper places Ndoki-Likouala within two of the twelve priority landscapes of the COMIFAC Convergence Plan (2002): the Sangha Tri-National and Lac Télé–Lac Tumba. [SM Appendix A] On the Lac Télé side, the Lac Télé/Likouala-aux-Herbes and Grands Affluents Ramsar sites and Ngiri-Tumba-Maindombe in the DRC have been managed together as one transboundary Ramsar site since 2017, covering more than 129,000 km². [Ramsar Secretariat announcement, as reposted by the Congo Basin Forest Partnership; the month is not confirmed]
The paper itself. It marks the forest elephant, bongo, leopard and red river hog as both connectivity-important and cross-border on its candidate table. [SM Appendix B]
Not shown. A figure of elephants travelling more than 150 km outside the Sangha Trinational and spending 40–60% of their time in the concessions appears only second-hand, citing unpublished reports, and is not used.

References

Blake S, Deem SL, Strindberg S, et al. (2008). Roadless wilderness area determines forest elephant movements in the Congo Basin. PLoS ONE 3(10): e3546. doi:10.1371/journal.pone.0003546
Blake S, Strindberg S, Boudjan P, et al. (2007). Forest elephant crisis in the Congo Basin. PLoS Biology 5(4): e111. doi:10.1371/journal.pbio.0050111
Stokes EJ, Strindberg S, Bakabana PC, et al. (2010). Monitoring great ape and elephant abundance at large spatial scales: measuring effectiveness of a conservation landscape. PLoS ONE 5(4): e10294. doi:10.1371/journal.pone.0010294
Clark CJ, Poulsen JR, Malonga R, Elkan PW (2009). Logging concessions can extend the conservation estate for Central African tropical forests. Conservation Biology 23(5): 1281–1293. doi:10.1111/j.1523-1739.2009.01243.x
Morgan D, Mundry R, Sanz C, et al. (2018). African apes coexisting with logging. Biological Conservation 218: 277–286. doi:10.1016/j.biocon.2017.10.026
Poulsen JR, Clark CJ, Mavah G, Elkan PW (2009). Bushmeat supply and consumption in a tropical logging concession in northern Congo. Conservation Biology 23(6): 1597–1608. doi:10.1111/j.1523-1739.2009.01251.x
Weinbaum KZ, Nzooh Z, Usongo L, Laituri M (2007). Preliminary survey of forest elephant crossings in Sangha Trinational Park, central Africa. Pachyderm 43: 52–62. doi:10.69649/pachyderm.v43i.127
Mockrin MH (2010). Duiker demography and dispersal under hunting in northern Congo. African Journal of Ecology 48(1): 239–247. doi:10.1111/j.1365-2028.2009.01107.x
Goué M, Kana CE (2023). The Sangha Trinational: an example of cross-border biodiversity management in Central Africa. In Managing Transnational UNESCO World Heritage Sites in Africa, 69–81. Springer. doi:10.1007/978-3-030-80910-2_7
UNESCO World Heritage List, property 1380, Sangha Trinational, open-data dataset whc001, read 2026-09-30.
Priority Species Index manuscript and Supplementary Material — the 2026-09-25 draft supplied to Landseed by the paper's authors (revision of 2026-09-29); unpublished and under review: SM Appendices A and B.

Recovery and Translocations · the record

The survey, stratum by stratum

The translocation record has not been suppliedThis chapter is for the per-event record of translocations, reintroductions, recolonisations and releases: date, species, count, source and destination. It draws only records the paper's authors have supplied, and for Ndoki-Likouala none has been. No table is drawn. The received files hold no such record, and the paper's one intervention for this landscape is a trade measure, not a release.

What the received record does hold is the survey itself: the same line-transect protocol run inside the national park and in the less protected units since 2006, stratum by stratum. It is shown here in full.

What the paper credits

1990: the ivory ban. "Ivory ban — Appendix 1 listing for the African elephant, making commercial trade illegal." This is the paper's only listed intervention for the landscape. [SM Appendix F]
What still presses. "Poaching, commercial hunting, logging and wildlife disease remain the most persistent threats." [SM Appendix A]
The paper's reading. Ndoki-Likouala is the one landscape of seven whose index "declined slightly overall", with "no statistically significant trend in the protected area". Split into the national park and the less protected units, "confidence intervals around the PSI overlapped, so no consistent difference could be established"; the less protected units' point estimate is negative. [manuscript, Results; Figure 3(b): density from line-transect distance sampling at multiple management units, with a 95% confidence interval]

Every stratum, every survey (SM Table 3)

Stratum	km²	2006	2010	2016	2022
Forest Elephant dung per km² ± CV%
Kabo	2,773	1050 ± 20.6	1018 ± 28.4	898 ± 14.3	1441 ± 22.3
NNNP	4,071	662 ± 15.8	680 ± 13.6	1013 ± 11.7	992 ± 16.45
Pokola	4,502	856 ± 27.2	683 ± 32	712 ± 29.6	110 ± 34.95
Sum (standardised)		2568 (1.0)	2381 (0.93)	2623 (1.10)	2543 (0.97)
Ungulates dung per km² ± CV%
Kabo	2,773	1218 ± 20	638 ± 27.4	594 ± 29.1	705 ± 17.05
LTCR	4,219	204 ± 32.8	194 ± 29	485 ± 27.7	340 ± 47.63
NNNP	4,071	1632 ± 12.7	927 ± 15.1	1676 ± 12.5	1005 ± 17.77
Pokola	4,502	1323 ± 23.1	765 ± 19.9	1838 ± 13.7	744 ± 20.19
Sum (standardised)		4377 (1.0)	2524 (0.58)	4593 (1.82)	2794 (0.61)
Great Apes nests per km² ± CV%
Kabo	2,773	264 ± 27.3	260 ± 41.1	274 ± 13.1	253 ± 26.97
LTCR	4,219	157 ± 22.1	257 ± 30.1	309 ± 32	121 ± 31.18
NNNP	4,071	152 ± 19.2	145 ± 17	198 ± 17.1	101 ± 21.04
Pokola	4,502	482 ± 26.6	661 ± 25.7	369 ± 17.7	214 ± 26.15
Sum (standardised)		1055 (1.0)	1323 (1.25)	1150 (0.87)	689 (0.60)
Chimpanzee nests per km²
Kabo	2,773	35	22	47	30
LTCR	4,219	17	27	25	3
NNNP	4,071	98	79	116	25
Pokola	4,502	37	49	33	25
Sum (standardised)		187 (1.0)	177 (0.95)	221 (1.25)	83 (0.38)
Gorilla nests per km²
Kabo	2,773	240	182	234	266
LTCR	4,219	145	229	260	127
NNNP	4,071	60	55	87	130
Pokola	4,502	454	655	317	240
Sum (standardised)		899 (1.0)	1121 (1.25)	898 (0.80)	763 (0.85)

SM Table 3, all five blocks as supplied: density estimates from line-transect distance surveys, dung per km² for elephant and ungulates, nests per km² for the great apes, with the coefficient of variation (%) where given. The value in brackets on each Sum row is the table's own standardised value, as printed. For 9 of the 15 cells after 2006 it is not that year's sum divided by the 2006 sum (ungulates 2016: 4,593 / 4,377 = 1.05, printed 1.82), and the table does not say how it was standardised. Survey reports: Maisels et al. 2012, Brncic et al. 2018, Estienne et al. 2025, and Stokes et al. 2010. Priority Species Index manuscript and Supplementary Material — the 2026-09-25 draft supplied to Landseed by the paper's authors (revision of 2026-09-29); unpublished and under review.

⚠ The caption says "weighted by the size of the reserve"; the Sum rows are not. Each Sum is the plain total of the stratum densities (elephant 2006: 1,050 + 662 + 856 = 2,568). The page shows the rows as printed and does not re-weight them.
⚠ The elephant has no Lac Télé row, where the other lines do. That absence is the table's.
⚠ Pokola's elephant density falls from 712 to 110 between 2016 and 2022, the largest single change in the table, with a CV of 34.95%. Pokola's area is 4,502 km² here and 4,797 km² in the strata file.
⚠ How many survey cycles? Appendix A speaks of "five survey cycles" since 2006 and of "approximately five-year cycles between 2010–2011 and 2022–2023"; the table has four periods, labelled 2006, 2010, 2016 and 2022.

How the index is computed here. The survey years align across the three lines, so the paper computes the index directly for each survey year rather than over standardised intervals; each line is set to 100 at its first survey and the lines are combined by geometric mean. [SM Appendix C; manuscript, Methods] The manuscript's intervention table has no rows for this landscape; Appendix F's 1990 entry is the only one.

The protected / less protected split, as totals

Status	Line	Year	Sum	Strata summed	Index
Protected	Elephant	2006	662	NNNP	100.0
Protected	Great Ape	2006	152	NNNP	100.0
Protected	Ungulate	2006	1632	NNNP	100.0
Unprotected	Elephant	2006	1906	Kabo + Pokola	100.0
Unprotected	Great Ape	2006	903	Kabo + Lac Télé + Pokola	100.0
Unprotected	Ungulate	2006	2745	Kabo + Lac Télé + Pokola	100.0
Protected	Elephant	2010	680	NNNP	102.7
Protected	Great Ape	2010	145	NNNP	95.4
Protected	Ungulate	2010	927	NNNP	56.8
Unprotected	Elephant	2010	1701	Kabo + Pokola	89.2
Unprotected	Great Ape	2010	1178	Kabo + Lac Télé + Pokola	130.5
Unprotected	Ungulate	2010	1597	Kabo + Lac Télé + Pokola	58.2
Protected	Elephant	2016	1013	NNNP	153.0
Protected	Great Ape	2016	198	NNNP	130.3
Protected	Ungulate	2016	1676	NNNP	102.7
Unprotected	Elephant	2016	1610	Kabo + Pokola	84.5
Unprotected	Great Ape	2016	952	Kabo + Lac Télé + Pokola	105.4
Unprotected	Ungulate	2016	2917	Kabo + Lac Télé + Pokola	106.3
Protected	Elephant	2022	992	NNNP	149.8
Protected	Great Ape	2022	101	NNNP	66.4
Protected	Ungulate	2022	1005	NNNP	61.6
Unprotected	Elephant	2022	1551	Kabo + Pokola	81.4
Unprotected	Great Ape	2022	588	Kabo + Lac Télé + Pokola	65.1
Unprotected	Ungulate	2022	1789	Kabo + Lac Télé + Pokola	65.2

The authors' workbook (sheet "Congo"), all 24 status-level rows: three lines × two statuses × four years. The "Number" is our reading of how each is built, checked against SM Table 3: Protected is the national park's density alone; less protected is the plain sum of the other strata that were surveyed. So the Lac Télé Community Reserve, a protected area, is counted as less protected, and the elephant's less protected total has no Lac Télé term. Index = the total against 2006. The index lines with their confidence intervals are in the Priority Species Index chapter.

Year	PSI (stated)	SD	95% interval
2006	100.00	0.00	100.00–100.00
2010	87.97	7.98	73.24–104.61
2016	105.49	6.26	93.72–118.43
2022	74.75	6.39	63.01–87.92

The landscape index as the workbook states it: the mean of the Monte Carlo simulations, with its standard deviation and interval. The paper propagates each species-year's uncertainty with 10,000 log-normal draws, recombines them by geometric mean, and takes the 2.5th and 97.5th percentiles, following Soldaat et al. (2017). By its own account this "does not explicitly account for other potential sources of uncertainty, such as sampling-design uncertainty, model-structure uncertainty, or serial or spatial autocorrelation". [SM Appendix C] The SM's leave-one-species-out results list five landscapes and give no figure for Ndoki-Likouala.

References

Priority Species Index manuscript and Supplementary Material — the 2026-09-25 draft supplied to Landseed by the paper's authors (revision of 2026-09-29); unpublished and under review: Appendices A, C and F; Table 3; Results; Figure 3(b); and the authors' workbook (sheet "Congo").
Maisels F, Nishihara T, Strindberg S, et al. (2012). Great ape and human impact monitoring training, surveys, and protection in the Ndoki-Likouala Landscape, Republic of Congo. WCS, 69 pp.
Brncic T, Madzoke B, Ndzai C, et al. (2018). Results of the 2016–2017 large mammal survey of the Ndoki-Likouala landscape. WCS Congo, 43 pp.
Estienne V, Strindberg S, Kuisma E, Abedine C, et al. (2025). Results of the 2022–2023 large mammal survey of the Ndoki-Likouala landscape. WCS Congo, Brazzaville, 83 pp.
Stokes EJ, Strindberg S, Bakabana PC, et al. (2010). PLoS ONE 5(4): e10294. doi:10.1371/journal.pone.0010294
Soldaat LL, Pannekoek J, Verweij RJT, van Turnhout CAM, van Strien AJ (2017). A Monte Carlo method to account for sampling error in multi-species indicators. Ecological Indicators.

Global Conservation Value

Ndoki-Likouala, measured against the world

What can be said of Ndoki-Likouala against the rest of the world, with a source, a scope and a date. Where a superlative could not be sourced it is not made; the list at the end names what was left out and why.

A World Heritage property across three countries. The Sangha Trinational was inscribed in 2012 under natural criteria (ix) and (x): 746,309 ha across three contiguous national parks (Lobéké in Cameroon, Dzanga-Ndoki in the Central African Republic and Nouabalé-Ndoki in the Republic of the Congo), with a buffer zone of 1,787,950 ha. UNESCO cites ecological processes on a very large scale in an intact landscape under (ix), and under (x) viable forest elephant populations and significant populations of western lowland gorilla and chimpanzee "both in and around the property". [UNESCO, property 1380]
Wetlands of international importance. Lac Télé/Likouala-aux-Herbes (Ramsar site 950, 438,960 ha, designated 18 June 1998); Grands Affluents (site 1742, 5,908,074 ha, 13 December 2007); Sangha-Nouabalé-Ndoki (site 1858, 1,525,000 ha, 2009). Across the Sangha, Cameroon's site 1739 (6,200 ha, 2008) and the Central African Republic's site 1889 (275,000 ha, 2009). [Ramsar Sites Information Service, annotated lists]
The world's largest tropical peatland lies to the south. The central Congo Basin peatland was mapped at about 145,500 km² holding about 30.6 Pg of carbon, the most extensive peatland complex in the tropics [Dargie et al. 2017]; a later field-based model puts it at 167,600 km² (36% of the world's tropical peatland area) storing 29.0 Pg C, about 28% of tropical peat carbon [Crezee et al. 2022]. These are two estimates of different method and date, not a trend. Peat was sampled along the Likouala-aux-Herbes river in the Republic of the Congo. How much of this landscape is peat is not established by any source read, so no share is given.
Intact forest. In 2000 the Republic of the Congo held 138,700 km² of intact forest landscape, 40.7% of its forest zone, and that area shrank by 17.7% between 2000 and 2013. Scope: the whole country. [Potapov et al. 2017]
Among the highest elephant densities in the region. Forest elephants across Central Africa fell by about 62% between 2002 and 2011 and lost 30% of their range (range defined there by forest cover). The same study places northern Congo among the region's highest elephant densities and names Nouabalé-Ndoki, Dzanga-Sangha and the certified concessions that connect and surround them as nationally important sites. [Maisels et al. 2013] It gives no numeric share for northern Congo, so none is shown.
Three threatened flagship species. African forest elephant, Critically Endangered (2021 assessment); western lowland gorilla, Critically Endangered (2016, amended 2018); central chimpanzee, Endangered (2016). Categories only. [Red List assessments; categories via Wikidata, CC0]
Formal standing. The paper's standing table records the 2002 COMIFAC Convergence Plan and the 2012 World Heritage inscription for this landscape; six of its seven landscapes hold a formal standing under statute, treaty or a government-approved plan. Its mandate table notes that elephant, gorilla and chimpanzee monitoring has run since 2006 across park and concession simultaneously, in partnership with the Ministry of Forest Economy. [SM Appendices A and B]
The only Afrotropical landscape in the paper. The paper's seven test landscapes span five realms, and Ndoki-Likouala is the only one in the Afrotropics. [manuscript, Results]
Monitored across the line since 2006. The same protocol has run in the national park and in the concessions simultaneously since 2006, in partnership with the Ministry of Forest Economy, "which few landscapes anywhere can match". [SM Appendices A and B] The paper argues that a protocol which excluded "great apes and forest elephants from Ndoki-Likouala" would be unacceptable to the agencies that would implement it. [manuscript, Discussion]

People

About one person per km², measured. About 34,615 people live in the grid cells whose centres fall inside the nine strata (35,272 km² of cells), 0.98 per km². Measured by Landseed on the 2020 epoch of the GHS-POP grid (release R2023A, 30 arc-second), 2026-09-30. Population: GHS-POP R2023A (Schiavina et al. 2023, doi:10.2905/2FF68A52-5B5B-4A22-8F40-C41DA8332CFE; Pesaresi et al. 2024, International Journal of Digital Earth 17(1)). © European Union, 1995-2025. Reuse of this data is authorised with proper acknowledgment of the source. The GHSL data are provided under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Clipped and aggregated by us. It is a modelled grid, not a census. It agrees with the paper's own figure, "close to 1 inhabitant/km²" [SM Appendix A].
No density rank. The paper states no human-density rank for this landscape among its candidate landscapes, so none is shown here.

Not claimed, and why

Country shares of the world's gorillas and chimpanzees. A range-wide study reports what share of western lowland gorillas and central chimpanzees the Republic of the Congo holds [Strindberg et al. 2018]. Those shares are not shown until it is confirmed how that study drew the species' ranges; figures in the same paper that rest on Red List population estimates or protected-area categories are not used at all.
Protected shares of the peat and the intact forest. The peat and intact-forest papers each give a share lying inside protected areas. Both are computed against a protected-area database this site does not use, so neither is shown.
Potential landscapes in the Republic of the Congo. The paper counts 7 “potential Priority Species Landscapes” larger than 1,000 km² in the Republic of the Congo (283,586 km² in total), 7 larger than 2,000 km² (283,586 km²) and 5 larger than 5,000 km² (277,282 km²). Their outlines are not drawn. SM-Table 4, derived from the WDPA (UNEP-WCMC), used for this paper by permission.
A Sangha transboundary Ramsar site. The official national lists hold three Sangha sites and no transboundary designation; the Ramsar transboundary list could not be read, so nothing is claimed either way.
The paper's species photographs for this landscape (chimpanzee, gorilla, forest elephant and yellow-backed duiker, credited to WCS) are not served: no image licence came with them.
A total area for the Trinational landscape ("4.4 million ha") appears only in secondary sources; the UNESCO figures are used instead.

References

UNESCO World Heritage List, property 1380, Sangha Trinational, open-data dataset whc001, read 2026-09-30.
Ramsar Sites Information Service, Annotated List of Wetlands of International Importance: Congo; Cameroon; Central African Republic (rsis.ramsar.org), read 2026-09-30.
Dargie GC, Lewis SL, Lawson IT, et al. (2017). Age, extent and carbon storage of the central Congo Basin peatland complex. Nature 542: 86–90. doi:10.1038/nature21048
Crezee B, Dargie GC, Ewango CEN, et al. (2022). Mapping peat thickness and carbon stocks of the central Congo Basin using field data. Nature Geoscience 15: 639–644. doi:10.1038/s41561-022-00966-7
Potapov P, Hansen MC, Laestadius L, et al. (2017). The last frontiers of wilderness: tracking loss of intact forest landscapes from 2000 to 2013. Science Advances 3(1): e1600821. doi:10.1126/sciadv.1600821
Maisels F, Strindberg S, Blake S, et al. (2013). Devastating decline of forest elephants in Central Africa. PLoS ONE 8(3): e59469. doi:10.1371/journal.pone.0059469
Red List categories: Loxodonta cyclotis doi:10.2305/IUCN.UK.2021-1.RLTS.T181007989A204404464.en; Gorilla gorilla gorilla doi:10.2305/iucn.uk.2016-2.rlts.t9406a136251508.en; Pan troglodytes troglodytes doi:10.2305/iucn.uk.2016-2.rlts.t15936a17990042.en (categories only, read via Wikidata P141).
Schiavina M, Freire S, Carioli A, MacManus K (2023). GHS-POP R2023A — GHS population grid multitemporal (1975–2030). European Commission, Joint Research Centre. doi:10.2905/2FF68A52-5B5B-4A22-8F40-C41DA8332CFE.
Pesaresi M, Schiavina M, Politis P, Freire S, Krasnodębska K, Uhl JH, et al. (2024). Advances on the Global Human Settlement Layer by joint assessment of Earth Observation and population survey data. International Journal of Digital Earth 17(1).
Priority Species Index manuscript and Supplementary Material — the 2026-09-25 draft supplied to Landseed by the paper's authors (revision of 2026-09-29); unpublished and under review: Results, Discussion, SM Appendices A, B and E.

Imagery: Contains modified Copernicus Sentinel data 2025–2026 — a median composite of 200 scenes, cloud-masked per pixel with the Sentinel-2 scene classification layer. 5.28% of the Likouala plate was never observed by any scene in the five-year search and is filled rather than imaged; those pixels are not mapped separately, so fine detail on this plate is not a measurement everywhere. · Elevation: Copernicus GLO-30 (COP-DEM_GLO-30-PUBLIC). Produced using Copernicus WorldDEM-30 © DLR e.V. 2010-2014 and © Airbus Defence and Space GmbH 2014-2018 provided under COPERNICUS by the European Union and ESA; all rights reserved. The organisations in charge of the Copernicus programme by law or by delegation do not incur any liability for any use of the Copernicus WorldDEM-30. · Survey strata: WCS Congo (Estienne et al.; Brncic et al.) · Neighbouring protected areas, water & roads: © OpenStreetMap contributors (ODbL) · Priority Species Index, from the paper's Supplementary Material · An independent visualisation by Landseed PBC. Not affiliated with, endorsed by, or produced for WCS, the Ministry of Forest Economy or the Government of the Republic of the Congo.