ANT 154BN–19 Concluding remarks

ANT 154BN–19 Concluding remarks - ANT 154BN...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ANT 154BN Lecture #19: Concluding remarks: Conservation 10 March 2011 Final exam Tuesday, March 15th; 1–3 pm Olson Hall 207 (this room) Review session Sunday, March 13th 4pm-6:10pm 1060 Bainer Hall www.timlaman.com www.timlaman.com Three simple questions 1. Should we focus solely on pristine forests? 2. Are all forest species affected equally? 3. Are all forests created equal? Three simple questions >1. Should we focus solely on pristine forests? Orangutan Distribution Borneo Sumatra “At some future period, not very distant as measured by centuries,... the anthropomorphous apes... will no doubt be exterminated” C. Darwin, 1871 The Descent of Man, and Selection in Relation to Sex Tim Laman www.timlaman.com Habitat loss and fragmentation Kalimantan Red indicates forest lost between 1996-2002. Fuller, Jessup, and Salim 2003 N0 = 50 N0 = 100 N0 = 250 N0 = 500 Marshall et al. 2009 Husson et al. 2009 250 km2 Less than 25% of the orangutans found on Borneo live inside formally protected areas. (Singleton et al. 2004; Wich et al. 2009) Even if all orangutans living in protected areas were completely safe, these populations are insufficient to ensure long term viability. (Marshall et al. 2006, 2009) Can orangutans survive and reproduce in logged forests? Results are equivocal. Higher densities in logged forests: Lower densities in logged forests: Russon et al. 2001; Marshall et al. 2007 Felton et al. 2003; Johnson et al. 2005 No density change in logged forests: Knop et al. 2004; Meijaard et al. 2007 Multiple factors are probably involved East Kalimantan as a test case 568 A 5˚00 N 1000 0 B I O L O G I C A L C O N S E RVAT I O N 1 2 9 ( 2 0 0 6 ) 5 6 6 –5 7 8 Orangutan distribution range Province/State boundaries 100 200 Kilometers Malaysia 0˚00 Indonesia East Kalimantan N B East Kalimantan as a test case Methods 2001-2004, n=762 transects (381 km) at 22 sites Recorded nest density, habitat disturbance, assessed habitat quality Village surveys to determine hunting pressure East Kalimantan as a test case Results: hunting is the real threat Marshall et al. 2006 Biol. Cons. Conclusion: orangutans are extremely susceptible to population declines due to hunting but tolerant of moderate levels of logging. 100 Population simulations (b) Hunting pressure (% population lost/yr) popu lose h ext i r p sible to lea Population size 100 100 Population size 200 200 0% 1% 3% 2% 200 400 600 800 1000 22. 2. 3 Six o Saraw capab and habita a l so a ished (c) 200 N0= 250; lines mean of 500 simulations Time (years) Marshall et al. 2009 100 22. 2 . 3 The provi Marshall et al. 2009 aged Found Hunting & pet trade hundreds of infants per year reach rescue centers 4-5 mother-infant pairs lost for each infant that comes into a rescue center Structured, validated surveys of 6,972 people in 699 villages Conservation implications 1. Light to moderate logging is compatible with orangutan conservation. Marshall et al. 2006 Biol. Cons. Conservation implications 1. Light to moderate logging is compatible with orangutan conservation. 2. Proximity to humans per se does not have negative effects. Marshall et al. 2006 Biol. Cons. Conservation implications 1. Light to moderate logging is compatible with orangutan conservation. 2. Proximity to humans per se does not have negative effects. 3. Conservation efforts should focus on eliminating hunting. Marshall et al. 2006 Biol. Cons. Conservation implications 1. Light to moderate logging is compatible with orangutan conservation. 2. Proximity to humans per se does not have negative effects. 3. Conservation efforts should focus on eliminating hunting. 4. The conservation value of lightly degraded forests must be publicized, so they are protected from additional logging, burning, or clear-cutting. Marshall et al. 2006 Biol. Cons. Conservation implications 1. Light to moderate logging is compatible with orangutan conservation. 2. Proximity to humans per se does not have negative effects. 3. Conservation efforts should focus on eliminating hunting. 4. The conservation value of lightly degraded forests must be publicized, so they are protected from additional logging, burning, or clear-cutting. 5. Addition of degraded forest to existing protected areas would substantially increase the amount of suitable habitat available for orangutans. Marshall et al. 2006 Biol. Cons. Three simple questions We should not focus solely on pristine forests. Three simple questions We should not focus solely on pristine forests. >2. Are all forest species affected equally? Meijaard et al. 2005 http://www.cifor.cgiar.org/Publications/ Methods Sensitivity to logging of 41 Bornean mammal species classified as: • Tolerant: Positive effect of logging, i.e., densities increased by more than 20 percent within the first year after logging. • Neutral: No recorded effects from selective logging, i.e., no significant changes (<± 20%) in densities following logging. • Intolerant: Severely impacted by selective logging, i.e., densities decline by more than 20 percent in the first year after logging, and do not recover within 5 years after logging. + 20% change in population density 0% - 20% Prelogging Postlogging Species differ in their responses to habitat degradation. • Tolerant: 14 species (e.g., elephant, banteng, prevost’s squirrel) • Neutral: 12 species (e.g., orangutan, binturong, bearded pig) • Intolerant: 15 species (e.g., sun bear, gibbon, yellow barking deer). Susceptibility to selective logging + 20% change in population density • Tolerant: • Neutral • Intolerant Prelogging Postlogging 0% - 20% n = 41 Bornean mammal species Meijaard, Sheil, Marshall & Nasi 2008 Biotropica What predicts sensitivity to logging? Meijaard, Sheil, Marshall & Nasi 2008 Biotropica Summary of general patterns Tolerant species • use all vegetation strata • generalized diets younger Intolerant species • strict tree canopy dwellers • specialized diets older Two broad groups • Species that evolved during late Pliocene or Pleistocene (0.5-3 mya) Periods of temperature fluctuations and climate instability Similar to logged tropical forest ? Tolerant of logging • Species that evolved during Miocene or early Pliocene (4-13 mya) Periods of relative warmth and climate stability Similar to unlogged tropical forest? Intolerant of logging Conservation implications 1. While most species are tolerant of well-managed, sustainable logging, those that are most negatively affected are those that are also most vulnerable to extinction. 2. Evolutionary history may allow us to predict which species are most likely to be adversely affected by logging. Three simple questions We should not focus solely on pristine forests. Forest species are affected differently. Three simple questions We should not focus solely on pristine forests. Forest species are affected differently. >3. Are all forests created equal? Source-sink population dynamics • Habitats vary & reproductive output (“r”) depends on habitat quality r > 0 = Source r < 0 = Sink • In a habitat mosaics, sink populations can persist because of immigration from nearby sources Gunung Palung National Park, West Kalimantan, Indonesia Borneo Vertebrate population monitoring 50m 200m 400m 800m 1000m Contours (asl) Fourteen 3-4 km transects each walked twice per month Forest types PS Peat swamp FS Freshwater swamp AB Alluvial bench LS Lowland sandstone LG Lowland granite UG Upland granite MO Montane Vertebrate population monitoring 50m 200m 400m 800m 1000m Contours (asl) Fourteen 3-4 km transects each walked twice per month Forest types PS Peat swamp FS Freshwater swamp AB Alluvial bench LS Lowland sandstone LG Lowland granite UG Upland granite MO Montane Vertebrate population monitoring Lowland sandstone Segments Lowland granite Upland granite Transect LS LG UG PS Peat swamp FS Freshwater swamp AB Alluvial bench MO Montane FIGURE 3. Group density score (individuals/km2) 10 9 8 7 6 5 4 3 2 1 0 A Gibbon sources and sinks at Gunung Palung? B Density by forest type (# individuals/km2) Group density score (individuals/km2) n = 7 forest types, R2 = 0.72, p = 0.02 Territory quality (# individuals/km2 in territory) 10 9 8 7 n = 33 groups, R2 = 0.81, p < 0.0001 Group size 0 200 400 600 800 (# individuals/group) 1000 Altitude midpoint of territory (m asl) n = 33 groups, R2 = 0.50, p < 0.0001 6 A B C 5 5 4 3 2 1 0 0 200 400 600 800 1000 Altitude midpoint of territory (m asl) Group size 6 4 3 2 0 200 400 600 Altitude (m asl) 800 1000 viduals/km2) 10 9 8 7 5 37 B 6 C Marshall 2009 Biotropica Birth AFR RLS Death RLS = TLS – AFR F1#1 AD Birth offspring #1 Dispersal offspring #1 IBI Birth offspring #2 F1#2 Dispersal offspring #2 Time with no offspring PN=2(F=0) = RLS – (IBI + AD) RLS Time with one offspring F1#1 F1#2 PN=2(F=1) = 2 * IBI RLS Time with two offspring F1#1 & F1#2 PN=2(F=2) = AD – IBI RLS dispersal (mo), data are from H. albibarbis at CPRS (Mitani 1990). P(F = 0), F(F = 1), P(F = 2), and cumulative probability (Cum. P) calculated using equations 1,2,3, and 4, respectively. • indicates combinations of parameters that are logical impossibilities (e.g., IBI + AD > RLS), given the model’s assumption that each group produced two offspring. See text for details. RLS 240.0 IBI 28.8 AD 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 60.0 93.0 120.0 P(F = 0) 0.63 0.49 0.38 0.59 0.45 0.34 0.55 0.41 0.30 0.51 0.32 0.17 0.45 0.27 0.12 0.40 0.22 0.07 0.26 • • 0.18 • • 0.10 • • P(F = 1) 0.24 0.24 0.24 0.32 0.32 0.32 0.40 0.40 0.40 0.32 0.32 0.32 0.43 0.43 0.43 0.53 0.53 0.53 0.48 • P(F = 2) 0.13 0.27 0.38 0.09 0.23 0.34 0.05 0.19 0.30 0.17 0.36 0.51 0.12 0.30 0.45 0.07 0.25 0.40 0.26 • • 0.18 • • 0.10 • • Cum. P 0.411 0.140 0.048 0.500 0.170 0.057 0.581 0.193 0.061 0.267 0.047 0.005 0.318 0.049 0.003 0.340 0.042 0.001 0.058 • • 0.038 • • 0.011 • • bound values as my best- and worst-case scenario values, respectively (Table 2). To my knowledge, there are no published estimates of RLS for wild or captive H. albibarbis. As animals may reasonably be expected to have substantially longer RLS in captivity (both due to earlier AFR and longer TLS), I consider the most appropriate available estimates of the RLS of the gibbon females I studied to be those provided by Palombit (1995) for Hylobates lar at Ketambe, Sumatra. After considering the available data, he estimated female RLS at 10– 20 yr (Palombit 1995). I thus used 240, 180, and 120 mo as my best-case scenario, mean, and worst-case scenario values (Table 2). 240.0 38.4 RESULTS My research assistants and I identified a total of 28 gibbon groups during our line transect surveys, and I identified an additional five groups during systematic reconnaissance surveys of the field site Mean values measured in high orest (Table 3). The majority of the groups detected during my fquality, reconnaissance surveys that were not detected on line transects were found in montane forests (three of five groups; Table 3). This was not a result of different detection probabilities in montane forests (see the following paragraph), but was due to extremely low population density in montane forest. To sample a larger number of gibbon groups in marginal habitat I surveyed montane forests far from the census routes. As described above, I followed all 33 gibbon groups in 2002 to ensure that demographic data and group counts of all groups were complete. Six of these groups inhabited montane forests. 240.0 48.0 180.0 28.8 lowland forests 180.0 38.4 180.0 48.0 Source-sin 120.0 28.8 120.0 38.4 • 0.64 • • 0.80 • • 120.0 48.0 longest RLS), the mean value, and a worst-case scenario value. I calculated probabilities based on all permutations of these values (Table 2). My estimate for IBI was based on a 6-yr study conducted on ten gibbon groups at CPRS. Mitani (1990) reported a mean IBI of 38.4 mo (± 4.8 SD), based on a sample of ve females that gave birth twice during the study. Mitani (1990) stressed that this should be seen as a minimum value as there were censored observations in the sample, i.e., longer IBIs were less likely to be observed. For my best- and worst-ca SD (Table 2). My estimate was no significant lue. Asnce in effective Discussion, (i.e.e the de- in montane forest are va differe noted in the strip width tru , values He reported 12 pr tection probability) among forest types; effective strip width was most likely between the mean and worst-case scenario values. RLS = calculated as 35 m in each of the seven habitats. There was no by subadults. Bas significant differfemalen rencounter rate among the fourestimates are from wild female ence i eproductive life span (mo), observers minimum age at d or based on moonylobates lar (Palombit 1995). TBI = e groups th interval (mo), data H phase, season, or topography. I he fivinter-bir was 93 mo (± 16. that I detected on reconnaissance surveys that were not detected on are from Hylobates albibarbis at CPRS (Mitani 1990). AD = age at line transect surveys inhabited territories that were > 200 m from bound values as m dispersal (mo), d why e from H. albibarbis on the nearest census trail, explainingata arthey were not observedat CPRS (Mitani 1990). (Table 2). censuses. P(F = 0), F(F = 1), P(F = 2), and cumulative probability (Cum. To my knowl P) calculated using equations 1,2,3, and 4, respectively. • indicates wild or captive H. HABITAT-SPECIFIC POPULATION of parameters that are Sogical impossibilities (e.g., IBI + combinations DENSITY AND GROUP l IZE.—Gibbon densities varied substantially across the seven forest types to have substantia AD forRLS), given the model’s densities (0.44 at CPRS: montane > ests supported the lowestassumption that each group produced AFR and longer T individuals/km2 two olowland sandstone for edetails. ted the high) and ffspring. See text for sts suppor 2 estimates of the R est densities (10.3 individuals/km ; Marshall & Leighton 2006). The next lowest gibbon densities, found in upland granite forest provided RLS IBI AD P(F = 0) P(F = 1) P(F = 2) CuMarshall 2009 Biotropica by Palom m. P After considering t 240.0 28.8 60.0 0.63 0.24 0.13 0.411 20 yr (Palombit 1 TABLE 2. Life history parameters used in calculations. For each variable, I used a best-case STRIP WIDTH AND mean v S.—There HABITAT-SPECIFIC EFFECTIVE scenario value, a BIAS TESTalue, and a worst-case scenario Lowland forest loss at Gunung Palung 1988 1994 2002 Curran et al. 2004 Science Gunung Leuser: Last stronghold of Sumatran orangutans 2002 Projected 2010 Some “source habitat” biases • High reproductive rate • High population density • High food availability Conservation implications 1. It is important to preserve the full complement of habitats that a species occupies. 2. Rapid surveys may be misleading. 3. Beware results from studies conducted in high quality habitats. Three simple questions We should not focus solely on pristine forests. Forest species are affected differently. Forests differ in important ways. Three simple results 1. Degraded forests can retain high conservation value. Some, but not all, vertebrate species can persist in degraded habitats. Species occupy habitats that cannot support them in the long term. 2. 3. ...
View Full Document

This note was uploaded on 04/05/2011 for the course ANT 154bn taught by Professor Debello during the Winter '10 term at UC Davis.

Ask a homework question - tutors are online