Multi-decadal declines in tree density and species richness as alien plants invade a tropical island’s protected wet forests F. B. Vincent FLORENS 1,3 , Claudia BAIDER 2 , Genevieve MARTIN 1 , Nooshruth B. SEEGOOLAM 1 , Zeyn ZMANAY 1 & Dominique STRASBERG 3 1 University of Mauritius 2 The Mauritius Herbarium 3 Université de La Réunion
Introduction Invasive alien species (IAS) cause major environmental damage and represent a main threat to biodiversity.
Introduction Potential for causing species extinction is most obvious and rapid in inter-trophic interactions like predation compared to con-trophic interactions like competition. 1500 1982 Frequency of Pandanus vandermeeschii 1350 1993 1200 Hare and rat eradication 2004 1050 900 750 600 450 Screwpine ( Pandanus vandermeeschii ) 300 150 0 Seedlings and non- Reproductive trees Dead adult trees reprodutive trees Alien mammal eradication from an offshore Mauritius islet: ‘spectacular’ population recovery
Introduction There exists a relative lack of cases demonstrating alien plants’ ability to cause plant extinction ( Davis 2003, Sax and Gaines 2008, Caujapé-Castells et al. 2010, Powell et al. 2013 ).
Introduction Difficulties to incriminate alien plants as drivers of the observed concurrent population decline and extinction of native plants: the coincidence of plant invasion with other threats like predation ( Gurevitch and Padilla 2004 ).
Introduction Another difficulty: competition-driven extinctions possibly take longer to occur than those caused by predation ( Davis 2003 ). Situation exacerbated by the longevity of many tropical trees (e.g. Fichtler et al . 2003 ) A situation of extinction debt ( Kuussaari et al. 2009 ) rather than of extinction per se may thus be favoured.
Introduction Long term studies on the effect of invasive species are rare despite the strong need for ecologists to adopt such a long-term perspective ( Strayer et al. 2006 ).
Aims 1. Measure current invasion by woody alien plants in the best preserved and protected native forests of a tropical oceanic island (Mauritius). 2. Compare tree community changes at some of the most intact native habitats that were studied 20 and 70 years earlier ( with Vaughan and Wiehe 1941; Lorence and Sussman 1986 ). 3. Relate results to more recent experimental approach studies ( comparisons between weeded and non-weeded forests )
Methods Quantifying current invasion level by alien plants • 75 random quadrats of 4 x 25 m distributed in 5 best preserved wet forest sites Quantifying tree community changes through time • Compare tree community with data from 1980’s and 1930’s Experimental approach: weeded v/s non-weeded • Compare community changes between weeded and non-weeded (at 2 sites) Investigating fitness of native plants between weeded and non-weeded forests • Compare survival, growth and reproductive rates between weeded and non-weeded adjacent areas
Study site Myers et al ., 2000, Nature 403: 853-585
Mauritius – basic facts Volcanic oceanic island 7.6 MY old 890 km to the east of Madagascar 1,865 km 2 ; highest peak 828 m First human colonisation: 1638. Uninterrupted since 1722
Native biodiversity Angiosperms: 691 species 39% endemic; 9% extinct; 70% threatened Vertebrates: 50 species 72% endemic; 46% extinct; 85% threatened Molluscs: 125 species 65% endemic; 34% extinct; 80% threatened
Confetti of habitats left Minimum viable populations? Habitat fragmentation ‘Ghost of past deforestation’ (Source: Vaughan and Wiehe 1937; Page and D’Argent 1997)
Study sites: National Park and Mountain reserve The largest area of contiguous native vegetation
Plant invasion level 32000 32000 Mare Longue 30000 Native Macchabe 30000 28000 28000 26000 26000 24000 Density per hectare 24000 Density per hectare 22000 22000 20000 20000 18000 Alien 18000 16000 16000 14000 12000 14000 10000 12000 8000 10000 6000 8000 4000 6000 2000 4000 0 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10-20 > 20 2000 dbh size class (cm) 32000 0 30000 Brise Fer 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10-20 > 20 28000 dbh size class (cm) 26000 Density per hectare 24000 22000 20000 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 32000 32000 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10-20 > 20 30000 Bel Ombre Camisard 30000 dbh size class (cm) 28000 28000 26000 26000 24000 24000 Density per hectare Density per hectare 22000 22000 20000 20000 18000 18000 16000 16000 14000 14000 12000 12000 10000 10000 8000 8000 6000 6000 4000 4000 2000 2000 0 0 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10-20 > 20 0 . 5 1 . 5 2 . 5 3 . 5 4 . 5 5 . 5 6 . 5 7 . 5 8 . 5 9 . 5 1 0 - 2 0 > 2 0 dbh size class (cm) dbh size class (cm)
Invasion Understorey heavily dominated by alien plants 11000 Native Natives Aliens 10000 Alien 9000 (75 plots of 100 m² from 5 sites with ‘best preserved’ forests ) Number of individuals 8000 7000 6000 5000 4000 3000 2000 1000 0 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5 18.5 19.5 >20 Diameter size class (cm)
A perfect nightmare
Native biomass lower when invasion more severe 10000 Basal area of native plants 8000 (cm 2 /100m 2 ) 6000 4000 2000 0 Low Medium High Invasion categories Box and whiskers plots of basal area of native woody plants against relative degree of invasion by alien plants. Each relative category of invasion (low, medium, high) comprises the respective five plots from each site for a total of 25 plots per category. All differences between pairs of categories are significant at P < 0.05 (Post Hoc Tukey Test).
1930’s ”it is now impossible to find even a small area free from exotics” (Vaughan & Wiehe 1941, J. Ecol. 29: 127-160)
Comparison over 70 years Macchabé Vaughan & Florens et Wiehe 1941 al. unpubl. (per 1000 m 2 ) Aliens 2 4,303 > 1 cm dbh (0.002 m -2 ) (4.3 m -2 ) 171 ± 24.6* Natives 85 > 10 cm dbh ( P < 0.05 ) 32.5 ± 5 Native spp 30 richness ( NS ) > 10 cm dbh
Alien plant invasion progress over 20 years > 2.5 cm, Sites Lorence & This study P< 0.01 < 10 cm Sussman 1980’s dbh % alien Brise Fer 34.8 60.7 plants Bel Ombre 20.8 25.7 47.6 ± 16.2 Native Brise Fer 54 n.s. species 37.9 ± 7.4 Bel Ombre 46 richness 58 ± 9.1 Native Brise Fer 76.2 density 63 ± 11.2 Bel Ombre 71.5 n.s. (100 m -2 )
Mortality Sideroxylon grandiflorum (Sapotaceae)* ‘Dodo - tree’ Mortality rate Site Forest regime N Deaths * (over 3 ¾ years) 71 2 2.82% Brise Fer Non-weeded 31 8 25.81% Mare Longue 38 3 7.89% Macchabé 133 1 0.75% Brise Fer Weeded 23 0 0% Mare Longue 4 0 0% Macchabé • Cyclone snapped trees not included Total tree mortality compared Non-weeded forest: 9.3% (N = 140) Weeded forest: 0.6% (N= 160)
Reproductive output Canopy tree Sideroxylon grandiflorum – ‘Dodo tree ’ Invasion strongly reduces reproductive output Flowering is more abundant in areas without alien plants* (U 122,78 = 3520.5; P = 0.002) Fruiting is on average 37 times higher in weeded areas* (U 140,135 = 6662.5; P< 0.001) * Baider & Florens (2006) In Laurance & Peres Emerging threats to tropical forests . Chicago Univ Press
Reproductive output Another canopy tree Canarium paniculatum (Burseraceae) Average number of seeds and fruits per tree (log) Average number of seeds and fruits per tree (log) Average number of seeds and fruits per tree (log) 1000 1000 1000 Z adj45,30 =-2.62; p < 0.01 100 100 100 10 10 10 U = 69.5; p < 0.01 1 1 1 Brise Fer Brise Fer Brise Fer Brise Fer non- Brise Fer non- Brise Fer non- Mare Longue Mare Longue Mare Longue Mare Longue Mare Longue Mare Longue weeded weeded weeded weeded weeded weeded weeded weeded weeded non-weeded non-weeded non-weeded C. paniculatum produced significantly more seeds in weeded areas at both sites studied
Reproductive output Same results for other species and guilds
Growth rate Growth rate of a canopy native tree (4 years monitoring) Comparison between adjacent weeded and non-weeded forest Sideroxylon grandiflorum (Sapotaceae) Growth rate (dbh) Mean yearly growth rate Management regime N (mm ± SE) 0.47 ± 0.23 125 Non-weeded area 1.11 ± 0.21 155 Weeded area
Growth rate Mean annual growth rate of the whole woody native species community (Monitored over about 4 years) Non weeded Weeded Growth Growth -95% 95% -95% 95% N (dbh, mm) N ( dbh, mm) 0.58 0.45 795 0.10 0.08 0.12 686 0.72 Brise Fer Mare 1353 0.08 0.06 0.11 995 0.44 0.35 0.53 Longue
Whole community monitored over 4 years Native woody plants > 1 cm dbh Weeded Invaded (8 yrs earlier) Difference Difference Number (%) Number (%) p* Brise Fer 54.0 < 0.001 Recruitment 8 0.9 401 Retrogression 12 1.4 10 1.3 0.791 Deaths 52 6.0 46 4.4 0.762 Mare Longue Recruitment 52 3.6 311 29.6 < 0.001 0.038 Retrogression 27 1.9 17 1.6 Deaths 49 3.4 24 2.3 0.623 * Per plot of 0.01 ha (100 m 2 )
Recommend
More recommend
