This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy Forest Ecology and Management 258 (2009) 415–424
Contents lists available at ScienceDirect
Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco
Assessment of ecosystem services as an opportunity for the conservation and management of native forests in Chile A. Lara a,b,*, C. Little b,c, R. Urrutia b, J. McPhee d, C. A´lvarez-Garreto´n d, C. Oyarzu´n e, D. Soto b,1, P. Donoso a,b, L. Nahuelhual b,f, M. Pino b,e, I. Arismendi b,c a
Instituto de Silvicultura, Facultad de Ciencias Forestales, Universidad Austral de Chile, Casilla 567, Valdivia, Chile Nu´cleo Milenio FORECOS, Iniciativa Cientı´fica Milenio and Fundacio´n FORECOS, Chile Escuela de Graduados, Facultad de Ciencias Forestales, Universidad Austral de Chile, Casilla 567, Valdivia, Chile d Departamento de Ingenierı´a Civil, Facultad de Ciencias Fı´sicas y Matema´ticas, Universidad de Chile, Blanco Encalada 2002, Santiago, Chile e Instituto de Geociencias, Facultad de Ciencias, Universidad Austral de Chile, Casilla 567, Valdivia, Chile f Instituto de Economı´a Agraria, Facultad de Ciencias Agrarias, Universidad Austral de Chile, Casilla 567, Valdivia, Chile b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 May 2008 Received in revised form 12 November 2008 Accepted 1 January 2009
This paper quantifies two important native forest ecosystem services in southern Chile: water supply and recreational fishing opportunities. We analyzed streamflow in relation to forest cover in six watersheds located in the Valdivian Coastal Range (398500 –408050 S), the effect of forest management on streamflow in two watersheds in the Valdivian Andes (600–650 m of elevation; 398370 S), and fish abundance as a function of forest cover in 17 watersheds located in the Coastal Range and the Central Depression (398500 –428300 S). We found that the annual direct runoff coefficient (quickflow/ precipitation) and total streamflow/precipitation in the dry summer season were positively correlated with native forest cover in the watershed (R2 = 0.67 and 0.76; *P = 0.045 and 0.027, respectively) during four years of observations. Conversely, a negative correlation was found between summer runoff coefficients (total streamflow/precipitation) and cover of Eucalyptus spp. and Pinus radiata plantations (R2 = 0.84; *P = 0.010). We estimated a mean increase of 14.1% in total summer streamflow for every 10% increase in native forest cover in the watershed. The analysis of streamflow changes between two paired watersheds dominated by native secondary Nothofagus stands, one thinned with 35% of basal area removal and a control, showed that the former had a 40% increase during summer (four years of observations). The best correlation between fish abundance and forest cover was found between trout abundance (%) and secondary native forest area in 1000 m ! 60 m stream buffers (R2 = 0.65, *** P < 0.0001). We estimated a 14.6% increase in trout abundance for every 10% increase of native forest cover in these buffers. Similar approaches to quantify forest ecosystem services could be used elsewhere and provide useful information for policy and decision-making regarding forest conservation and management. ! 2009 Elsevier B.V. All rights reserved.
Keywords: Ecosystem services Streamflow Water supply Recreational fishing Forest policy Valdivian Rainforest Ecoregion Climatic change mitigation
1. Introduction Temperate rainforests of southern South America have impressed scientists even before Darwin, due to their diversity of life forms, high degree of endemism and large long-lived trees, including Fitzroya cupressoides which may live for more than 3620 years (Armesto et al., 1998; Smith-Ramı´rez, 2004; Lara and Villalba, 1993). These forests cover 13.4 million ha in Chile and account for more than one half of the temperate rainforests in the
* Corresponding author. Tel.: +56 63 221229; fax: +56 63 221227. E-mail address: antoniolara@uach.cl (A. Lara). 1 Present address: Fisheries and Aquaculture Department, FAO of UN, Via delle Terme di Caracalla 00100, Italy. 0378-1127/$ – see front matter ! 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2009.01.004
southern Hemisphere (Alaback, 1991; Donoso, 1993; CONAF et al., 1999). A major part of South American temperate forests are found within the Valdivian Rainforest Ecoregion (35–488S) in Chile and adjacent areas of Argentina, which has been classified among those with the highest conservation priority worldwide (Olson and Dinerstein, 1998). Most native forests in Chile are privately owned (71% of the total). The remaining is in national parks and reserves. Privately owned forests have been generally valued and used for firewood and timber production (mainly with unsustainable logging schemes) or as land for the expansion of other productive activities: agriculture, pastureland and fast-growing commercial tree plantations of exotic species (Pinus radiata and Eucalyptus spp). Rapid conversion to forest plantations between 1975 and 2000 resulted in deforestation rates of 4.5% per year within an area
Author's personal copy 416
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
of 578,000 ha in the Maule region (358S), facilitated through afforestation incentives (Echeverrı´a et al., 2006). Another important cause of deforestation has been human-set fires, with an annual average of 13,000 ha burned in the period 1995–2005 and a high interannual variability associated to rainfall variation (Lara et al., 2006). Most of the timber harvested from native forests is used for firewood, estimated at 8.5 million m3/year (80% of the annual roundwood consumption). The remaining 20% is represented by industrial use of native timber, mainly for the production of OSB
boards (Oriented Strand Boards) and sawn timber (Lara et al., 2006). The poor conservation status of native forests may be explained by the forest policy followed since 1974 in Chile. This policy has not provided economic incentives for the sustainable management and conservation of native forests, in contrast to the use of public funds to support the establishment of plantations. This, along with the liberalization of exports and privatization of state-owned plantations and pulp mills, explain the fast growth of the forest industry based on plantations, often regarded as an economically
Fig. 1. Regional map showing the location of the watersheds included in the study.
Author's personal copy A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
successful model in other Latin American countries and elsewhere (Lara and Veblen, 1993; Sedjo et al., 1999; Lara et al., 2006). Nevertheless, a new law including economic incentives for the management and conservation of native forests was finally approved in July 2008, after 16 years of discussion. As a consequence of native forest loss and degradation in Chile, important ecosystem services are being lost to society (Lara et al., 2003). Ongoing research in Chile has started to demonstrate the importance of native forests in the provision of ecosystem services that directly or indirectly benefit society, such as water supply (both quantity and quality), tourism, recreational fishing opportunities and biodiversity conservation (Lara et al., 2003; Nahuelhual et al., 2007). Ecosystem services are crucial to economic development and social well-being (Costanza et al., 1997). The supply of these services depends on the structure and processes of ecosystems and is reduced with ecosystem degradation (Millennium Ecosystem Assessment, 2003). A key explanation for the degradation of ecosystem services in different regions of the world is that their physical and economic values have not been adequately quantified and that, in most cases, they lack a market price that would make them comparable to other goods (e.g. timber). Therefore, ecosystem services have often been ignored or given little importance in policy-making (Costanza et al., 1997; Nahuelhual et al., 2007; Lucke, 2008). It is the interest of the scientific community as well as of the various social actors in Chile and in other countries to better understand the linkages between the function and structure of temperate forest ecosystems and the provision of ecosystem services. This national and international scenario requires the quantification of ecosystem services, so they can be included in policy and decision-making regarding forest conservation and management. Given this challenge, in this paper we address three questions regarding water supply and recreational fishing opportunities as two relevant ecosystem services from native forests: (a) How does streamflow vary in watersheds as native forest cover increases, while exotic plantations decrease? (b) How does the management of native forests for timber production affect streamflow?, and (c) Is trout abundance in streams related to native forest cover in the watersheds? Addressing these complex questions required the integration of different disciplines. 2. Study area The study area (398370 –428300 S) comprised different watersheds located in three main physiographic units, following a longitudinal pattern from west to east (Fig. 1). The Coastal Range is composed by metamorphic rocks, partially overlaid by Tertiary marine sediments, with intermediate slope angles, ranging from 300 to 1000 m in elevation (Servicio Nacional de Geologı´a y Minerı´a, 1982; Le Roux and Elgueta, 2000). The Central Depression is dominated by lahars and pyroclastic fluxes as well as till and outwash deposits, creating smooth slopes or flat areas at a general elevation of 100–300 m (Servicio Nacional de Geologı´a y Minerı´a, 1982). The Andean Range is a geologically complex system dominated by granitic, andesitic and sedimentary rocks, on relatively steep slopes mainly at elevations from 600 to 1700 m a.s.l., with some volcanoes that reach 3000 m a.s.l. (Levi et al., 1966; Servicio Nacional de Geologı´a y Minerı´a, 1982). The regional climate is classified as oceanic wet temperate with Mediterranean influence, expressed in a decrease of summer rainfall (January–March), which represents <10% of the annual value (Direccio´n General de Aguas, unpublished) and is characterized by a high interannual variability (Miller, 1976). Rainfall ranges from 1700 mm/year in the rain shadow eastern slope of the Coastal Range to 4500 mm at the tops of the Coastal Range and at mid and
417
high elevations in the Andes (Oyarzu´n et al., 1998; Pezoa, 2003, from data collected by Direccio´n General de Aguas, unpublished; Little et al., 2008). Precipitation occurs mainly as rain, but snow is common above 700 m of elevation between June and August. Mean annual temperature ranges between 9.9 and 12.1 8C, with mean maxima in January (12.3–17.9 8C) and mean minima in July (5.7– 7.7 8C, Direccio´n Meteorolo´gica de Chile, unpublished). For the analysis of streamflow in relation to forest cover, we studied six watersheds in the Valdivia area (398500 –408050 S). For the study of the effect of forest management on streamflow, we analyzed two watersheds located at 600–650 m of elevation in the Valdivian Andes (398370 S). Fish abundance as a function of forest cover was studied in 17 watersheds located in the Coastal Range and the Central Depression (398500 –428300 S, Fig. 1, Table 1). 3. Methods 3.1. Land use and forest cover A detailed mapping of the watersheds included in this study was performed in order to delineate their boundaries and classify their land use and forest cover. This classification was done from the interpretation of natural color near-vertical 1:8000–1:10,000 scale air photos taken for each watershed in 2004 and digital orthophotos were developed using ERDAS1 software (Environmental Systems Research Institute, Inc., USA). Standard photointerpretation and ground truthing techniques were used (Lillesand and Kiefer, 1994). Classification of land use–vegetation cover followed the protocol described by Lara and Sandoval (2003). This classification included the following categories: (1) Oldgrowth native forest, (2) Second-growth native forest, (3) Young forest exotic plantation (<12 year-old), (4) Adult forest exotic plantation (12–21 year-old), (5) Native shrubland (mainly Chusquea bamboo thickets), (6) Agriculture and pasturelands, (7) Other land uses (include wetlands, urban areas and barren lands). For the analyses and presentation of the results some of the related categories (e.g. young plantation and adult plantation) were grouped together. Final maps were developed and incorporated into a Geographic Information System using Arcview1 3.3 (Environmental Systems Research Institute, Inc., USA). In the watersheds used for the fish abundance study, we also assessed the land use and forest cover for buffer zones of varying width (60, 150, 300 m) and length in 1000 m increments upstream from the sampled stream reach, using the methods and categories mentioned above. 3.2. Streamflow and forest cover We selected six watersheds with a wide range of variation in the area covered by native forests or by fast-growing forest plantations (Pinus radiata, Eucalyptus globulus or Eucalyptus nitens) (Fig. 1, Table 1). In order to make them comparable, the selected catchments had similar topography, geology, elevation, climate and soil type. Rainfall was recorded in the Valdivia weather station (398480 S, 738140 W). Additional daily precipitation amounts were recorded with two tipping bucket rain gauges and from four small pluviometers (surface = 200 cm2) installed in the catchments. Streamflow during four hydrological years (April 2003 to March 2007) was recorded daily by water level meters installed in the streams and from daily observations of staff-gages. Short gaps in the data (<10%) were filled using measurements of streamflow at nearby watersheds having similar vegetation and showing a strong correlation in stream discharge. To derive the stage–discharge relationship, streamflow was measured 20–30 times during the study period, using the velocity-area method (Schulz, 1989). Using
Author's personal copy 418
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
Table 1 Total area in hectares and percentage (%) of the different land use/forest cover classes in the 20 watersheds. Code
Watershed
Studya
Area (ha)
Native forest (%)
Forest plantation (%)
Shrubland (%)
Pasture and Agriculture landb (%)
Dominant Native Forest Type or Plantation species (age in 2004)c
SEN SJU MIN JOQ JOA PIL PLA PUN BUT BLA CAI MAU MAC CAN GAT GUI LAH TRA SPT SPC
Senderos San Juan Las Minas Joaquines 1 Joaquines 2 Pillo-pillo La Plata Puntra Butalcura Blanco Cainalhue Maullin Machete ˜ al Can Gatito Guindos ˜ adi Lahuen n Tramaihue San Pablo Thinnned San Pablo Control
SF SF SF S F SF SF F F F F F F F F F F F TH TH
140 1462 884 282 55 1114 419 9352 4853 1764 7763 637 154 3039 75 252 33 4427 12.6 7.4
88.4 92.1 67.8 46.6 6.8 29.1 17.8 59.4 39.6 58.9 85.8 25.8 17.3 27.2 62.8 69.8 92.1 1.0 100 100
3.6 – 13.5 33.1 76.0 54.5 63.4 0.8 – – 13.1 – – – 0.7 22.2 – 0.1 – –
3.6 7.5 14.0 13.6 13.6 9.9 18.2 26.6 52.9 12.0 0.8 72.0 1.7 2.7 1.6 6.0 1.4 – – –
4.4 0.4 4.7 6.7 3.6 5.5 0.6 13.2 7.5 29.1 0.3 2.2 81.0 70.1 34.9 2.0 6.5 98.9 – –
Second-growth/Mixed broadleaved evergreen Second-growth–old-growth/Mixed broadleaved evergreen Second-growth/Mixed broadleaved evergreen Second-growth/Mixed broadleaved evergreen—Eucalyptus nitens (9) Eucalyptus nitens (9) Pinus radiata (21) P. radiata–E. globulus (21 and 25) Old-growth/Mixed broadleaved evergreen Second-growth–old-growth/Mixed broadleaved evergreen Second-growth–old-growth forests/Mixed broadleaved evergreen Old-growth forest/Mixed broadleaved evergreen Second-growth forest/Mixed broadleaved evergreen – Second-growth–old-growth forests/Mixed broadleaved evergreen Second-growth forest/Mixed broadleaved evergreen Second-growth forest/Mixed broadleaved evergreen Second-growth forest/Mixed broadleaved evergreen – Deciduous Nothofagus/Second-growth Deciduous Nothofagus/Second-growth
a S: Streamflow as a function of landuse/forest cover; F: Fish abundance as a function of landuse/forest cover; TH: Streamflow response to forest management (thinning) of native forests. b Includes other uses: wetlands, urban areas and barren lands; c All species in plantations are exotic.
a one-parameter filter (Chapman, 1999; Chapman and Maxwell, 1996), we separated the total streamflow (Qt) in baseflow (Qb, sum of groundwater flow and saturated superficial flow) and quickflow (Qq, total streamflow minus baseflow). Finally, we computed runoff coefficients by dividing both total streamflow volume and quickflow by precipitation (Qt/P and Qq/P, respectively). In order to assess the effect of different forest cover types in streamflow we performed regression analyses. 3.3. Forest management and streamflow We studied two watersheds covered with second-growth forests. One of them was thinned extracting 35% of the total basal area for the production of roundwood, whereas the other one remained unthinned as a control. Both paired watersheds were located on deep loam textured volcanic soils (100–120 cm) over a 90-cm deep pumice layer. Streamflow was recorded in V-notch 908-section gauges following standard procedures (Ward and Trimble, 2004) during four hydrological years (April 2003 to March 2007). In the same notches, water levels were measured with an hourly resolution by DIVER1 pressure sensors and recorded by data loggers. Precipitation was recorded with a HOBO1 tipping bucket rain gauge in a nearby meadow. We studied rainfall storms and analyzed their streamflow response. A one-parameter filter algorithm was used to separate direct runoff or quickflow (Qq) and baseflow (Qb) for the various seasons following the methods already described (Chapman and Maxwell, 1996). 3.4. Fish abundance and forest cover Sampling for fish abundance was done in second or third order streams, where fish capture could be conducted throughout the year. In most cases we selected groups of nearby watersheds dominated by different land use and vegetation cover, which share similar geology, climate and soils. Fish sampling was conducted from January 1999 through May 2004. Each stream was sampled at least once in the dry summerfall season (December–April) and once in the rainy winter–spring
season (May–November). Sampling consisted of wading along a portion of the stream, with ‘‘Smith and Root Model 12B’’ electro fishing equipment, with which we sampled a reach that was at least 10 times as long as the mean width of the stream or river. The area sampled was normally between 300 and 1500 m2 in order to cover all the different habitats of a particular stream. A three-pass electrofishing was used (Lobo´n-Cervia´, 1991) and the first pass accounted for 92% of the captured fish. An estimator of fish abundance was calculated as the number of captured fishes divided by the area electrofished. We used regression analysis to determine the effect of the percentage area of different land uses in the entire watersheds and in stream buffers along the stream corridor on fish abundance, expressed as a percentage. 4. Results 4.1. Land use and forest cover The total area of the watersheds included in the study of streamflow in watersheds with different proportions of land use– forest cover ranged between 140 and 1462 ha, whereas the watersheds used for the effect of forest management (thinning) ranged between 12.6 and 7.4 ha (Table 1). For the fish abundance study, watershed area ranged between 33 and 9352 ha (Table 1). Some watersheds were covered in more than 85% of their area by native forests (e.g. SEN, LAH, SJU), whereas others were dominated by forest plantations (JOA, PLA, Table 1). MAU and BUT were mainly covered by shrublands. Some watersheds exhibited a shared proportion of native forests and forest plantations (e.g. JOQ, PIL). In certain watersheds used only for the fish abundance study, the dominant land use types (>70% of the area) were pasture and agriculture land (MAC, CAN and TRA, Table 1). This set of watersheds with varying amounts of plantations, native forests and other land uses provided the means to empirically analyze the effect of varying proportions of forest cover types on the selected ecosystem services. Most of the watersheds were dominated by second-growth forest stands due to past human disturbance, but some (e.g. PUN, CAI) were dominated by old-growth forest stands (Table 1). Native
Author's personal copy 419
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
Fig. 2. (a) Relationship between mean annual direct runoff coefficient (quickflow/precipitation: Qq/P) and native forest (%) for the six catchments. (b) Same for exotic plantation (%). Water year: April–March of following year, (n = 4). Vertical bars represent standard errors (SE).
forests in the watersheds were classified as mixed broadleaved evergreen dominated by Eucryphia cordifolia, Laureliopsis phillipiana, Laurelia sempervirens, Nothofagus dombeyi, Gevuina avellana, Embothrium coccineum, Amomyrtus luma and Aextoxicon punctatum, all evergreen hardwoods. Crown cover of native forests was typically >50% and in most cases >75%. Young forest plantations typically had <50% crown cover and this value increased to >75% in adult plantations. Watersheds used in the thinning experiment were covered by second-growth stands of Nothofagus obliqua–N. nervosa, both deciduous hardwoods. 4.2. Streamflow and forest cover We found a significant positive correlation between native forest cover and the annual runoff coefficient (R2 = 0.67, *P = 0.045; Fig. 2a). The highest average runoff coefficients (0.42 and 0.53) were found at SEN and SJU watersheds, with 89.4% and 91.6% of native forest coverage, respectively (Fig. 2a, Table 1). The inverse relationship was found with exotic plantation cover, with annual runoff coefficients decreasing as the area of plantations increased (R2 = 0.59; P = 0.072). The minimum runoff coefficient (0.30) was estimated for PLA watershed, where 62% of the area was covered by 21-year old Pinus radiata and 12-year old Eucalyptus globulus plantations (Table 1, Fig. 2b). Native forests present in the six watersheds were second-growth forest stands with the exception of SJU where both old-growth and second-growth stands were present (Table 1). Therefore, a specific analysis for old-growth forests was not possible. For the seasonal analysis, the highest correlations between runoff index (total streamflow/precipitation) and native forest cover occurred during summer season (R2 = 0.76, *P = 0.027, Fig. 3), while a negative correlation was found for summer runoff coefficients with exotic plantation cover (R2 = 0.84; *P = 0.010, not shown). Using the linear equations developed in Figs. 2 and 3 and the mean values of
the results, we developed indicators for water supply as an ecosystem service from native forests (Table 2). These indicators are expressed as the increment in annual quickflow (5.8%) and total summer streamflow (14.1%) for a 10% increase in native forest cover for a given precipitation amount (Table 2). The same increase in forest plantation cover would reduce the total summer streamflow by 20.4% (equation not shown). 4.3. Forest Management and streamflow Thinned and control watersheds responded differently to rainstorm events (Fig. 4). Fig. 4a shows the total runoff and baseflow response of both watersheds to a high-intensity and short-duration storm. The thinned watershed total runoff increased more rapidly than the control, and its area-normalized peak flow was also larger. The recession curves after 50 h became very similar in both magnitude and slope. The thinned watershed baseflow curve was always higher than the baseflow curve for the control, but towards the end of the storm both curves tended to converge (Fig. 4a). An example of a response to a lower intensity storm sustained over a longer period of time is shown in Fig. 4b. In this case, the hydrograph separation was influenced by the fact that total streamflow at the beginning of the storm was higher for the thinned watershed. The rate of response was similar for both watersheds for the concentration (increase) phase and for the recession phase, with both hydrographs responding somewhat ‘‘parallel’’ throughout the storm (Fig. 4b). When baseflow was subtracted for storm 1 (Fig. 4c), the direct runoff curves were more similar than in the previous analysis. For storm 2, direct runoff for the thinned watershed became higher compared to the control for the second half of the observed period (Fig. 4d). Similar to the response of streamflow to individual storms, the thinned watershed yielded higher stream discharge values compared to the control during the four years of observation (Fig. 5).
Table 2 Indicators and increase (%) in water supply and recreational fishing opportunities as ecosystem services for a 10% increment in native forest cover. Ecosystem service
Indicator
% Increase for a 10% increment in native forest cover
Water supply (annual) Water supply (summer) Recreational fishing opportunities
Quickflow (l/s)a Total streamflow (l/s)b Trout abundance (%)c
5.8 (5.1–6.7) 14.1 (11.1–19.4) 14.6d (10.4–27.1)
a b c d
Derived from the equation provided in Fig. 2a. Assuming the same precipitation in all watersheds. Derived from the equation provided in Fig. 3. Assuming the same precipitation in all watersheds. Derived from the equation provided in Fig. 7b. Includes only second-growth forests in 1000 m ! 60 m stream buffers.
Author's personal copy 420
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
relative changes in order to be able to compare the watersheds’ response throughout the year, independent of hydrometeorologic conditions. Despite the little differences for streamflow expressed in mm for summer, the largest percentage differences in total streamflow occurred in this season (40%, Fig. 6a). Summer relative difference increased to 50% when calculated for baseflow, which can be attributed to the reduction in canopy interception and evapotranspiration in the thinned watershed compared to the control and a relatively faster depletion of soil and water table reserves in the control watershed through the summer (Fig. 6b). 4.4. Fish abundance as a function of native forest cover Total fish abundance, the abundance of trout (brown trout, Salmo trutta, and rainbow trout, Oncorhynchus mykiss) and native fishes (11 species) along stream reaches in 17 watersheds are shown in Table 3. All watersheds had both trout species and native fish. Native fish were more abundant than trout in 10 watersheds, which were located in the Coastal Range and in the Central Depression. The maximum native fish density (5.07 ind/10 m2) ˜ al watershed and the minimum (0.15 ind/10 m2) was found in Can in Guindos watershed. The maximum trout abundance (13.21 ind/ 10 m2) was found in Gatito and the minimum (0.13 ind/10 m2) in La Plata watershed. Correlation analysis between the non-native trout abundance in streams and the percentage of the watershed covered by different land use–vegetation cover classes indicated a positive correlation between the abundance (%) of exotic trout and native secondgrowth forest cover (R2 = 0.303, *P = 0.022, Fig. 7a). When only the second-growth forests in 60 m-wide and 1000-m long buffers along streams were considered, correlation increased (R2 = 0.65, *** P < 0.0001, Fig. 7b). These significant relationships between trout abundance and second-growth forests at a watershed and buffer level were not found when old-growth forests were used as a predictive variable (R2 = 0.08 and 0.09, respectively). This situation could be mainly explained by the low or null percentage of oldgrowth forests in most of the 17 watersheds. The relative abundance (%) of native fish had a negative correlation with second-growth forest cover within these buffers (R2 = 0.65, ***P < 0.0001, not shown), but the same relationship considering absolute fish Table 3 Abundance (number of individuals/10 m2 and %) of native fishes and introduced trout. Watershed
Fig. 3. Relationship between mean seasonal runoff coefficient (total streamflow of the season/precipitation of the season plus the previous one; Qt/P) and native forest (%) for the six catchments, (n = 4). Fall (April–June); Winter (July–September); Spring (October–December), Summer (January–March). Vertical bars represent standard errors (SE).
Absolute differences between both watersheds were higher during fall and winter seasons, which concentrate 67% of the annual precipitation (April through September), whereas relative differences increased as total discharge diminished toward the summer (dry season, Fig. 6). Because there is an order-of-magnitude difference between flows in wet and dry seasons, we focused on
Puntra Butalcura Blanco Cainalhue Maullin Machete Minas San Juan ˜ al Can Gatito Joaquines Senderos Guindos Plata Pillo-pillo ˜ adi Lahuen n Tramaihue
Density (ind/10 m2)
Abundance (%)
Trouta
Nativeb
Total
Trouta
Nativeb
0.81 0.75 4.41 0.47 2.85 3.69 0.93 0.49 0.66 13.21 3.29 1.85 0.15 0.13 0.74 2.15 1.77
1.87 1.81 1.47 1.18 1.58 1.60 0.52 1.73 5.07 0.34 4.31 0.35 0.15 2.94 3.06 2.65 4.41
2.68 2.56 5.91 1.65 4.44 5.29 1.44 2.22 5.73 13.55 7.60 2.20 0.29 3.07 3.79 4.79 6.18
30.15 29.44 74.97 28.57 64.33 69.75 64.31 22.09 11.56 97.47 43.26 83.98 50.00 4.16 19.38 44.78 28.66
69.85 70.56 25.03 71.43 35.67 30.25 35.69 77.91 88.44 2.53 56.74 16.02 50.00 95.84 80.62 55.22 71.34
a Introduced trout include brown trout (Salmo trutta) and rainbow trout (Oncorhynchus mykiss). b Native fishes include Geotria australis, Cheirodon australe, Trichomycterus areolatus, Galaxias maculatus, Galaxias platei, Brachygalaxias bullocki, Percichthys trucha and Percilia gillissi.
Author's personal copy A. Lara et al. / Forest Ecology and Management 258 (2009) 415â&#x20AC;&#x201C;424
421
Fig. 4. Streamflow response following two rainstorms for the thinned and control watersheds. (a and b) Hydrographs for total runoff and baseflow for storms 1 and 2. (c and d) Hydrographs for direct runoff.
abundance (ind/10 m2), although negative, was not significant (R2 = 0.22, P = 0.057, not shown). Pearsonâ&#x20AC;&#x2122;s correlation coefficients between trout abundance and the percent area of second-growth forests reached a maximum value for 1000-m long buffers along streams (R2 = 0.76â&#x20AC;&#x201C;0.80, *** P < 0.0004), with decreasing correlation at longer distances and little effect due to buffer width (Fig. 8). Using the linear equation shown in Fig. 7b and the mean values of the results, we developed an indicator for recreational fishing opportunities as an ecosystem service from native forests (Table 2). We estimated a 14.6% increase in trout abundance for each 10% increase of native forest cover in 1000 m ! 60 m buffers along streams (Table 2). A limitation of trout abundance as an indicator of recreational fishing opportunities is that it does not consider the size, vigor and other trout conditions that influence this activity. 5. Discussion and conclusions
Fig. 5. Total monthly runoff for the thinned and control watersheds during four hydrologic years.
This study quantified the ecosystem services from native forests regarding water supply and recreational fishing opportunities in southern Chile. Our results documented a positive correlation
Author's personal copy 422
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
Fig. 6. (a) Total seasonal runoff (mm) for the thinned and control watersheds during four years and difference (%) between both watersheds. (b) Same for baseflow. Vertical bars represent standard errors (SE).
Fig. 8. Pearson correlation coefficients between the percent area of native secondgrowth forests and trout abundance (%) for 60–300 m wide and 1000–5000 m long buffers along streams. Horizontal line indicates significant correlation *P < 0.05.
Fig. 7. Trout abundance as a function of native second-growth forest cover (%) in (a) the entire watershed. (b) 60 m wide and 1000 m long buffers along streams. Each point in (a) and (b) represents one of the 17 watersheds sampled for fish abundance.
between native forest cover and summer (the dry season, when flows decrease) runoff coefficient (quickflow/precipitation, Qq/P), and a negative correlation between this coefficient and percentage of exotic plantations, both statistically significant. In the case of native forests, this significant positive correlation was also found for annual runoff coefficient. Results regarding plantations are consistent with the high transpiration demand described for Eucalyptus spp. and Pinus spp. plantations (Calder et al., 1997; Scott and Lesch, 1997; Farley et al., 2005; Jackson et al., 2005; Huber et al., 2008).
Otero et al. (1994) also indicated that conversion of native forests to fast-growing plantations decrease streamflow especially in summer. In addition, studies of the water balance of young plantations of E. globulus and P. radiata in south-central Chile have revealed an increased depletion of the soil moisture reserves with stand ageing, as well as an increase in the canopy interception and evapotranspiration (Oyarzu´n and Huber, 1999; Huber et al., 2008). Furthermore, conversion to plantations has led to a decrease in water quality due to increased sediment loads associated to clearcuts in plantations managed under 12-year rotations for Eucalyptus spp. and 20 years for Pinus radiata (Lara et al., 2003, ˜ a, 1995). 2006; Oyarzu´n and Pen The analysis of streamflow changes between a thinned and a control watershed, both dominated by second-growth Nothofagus forests, showed that the former had a higher annual and summer streamflow compared to the control. Studies conducted in deciduous hardwood forests in Central Massachusetts showed that a decrease in total basal area by about 34% resulted in an increase in total streamflow, baseflow and groundwater recharge, especially in the dormant season (Bent, 2001). A compilation of experiments including deforestation, afforestation, regrowth and forest conversion resulted, in the majority of cases, in increased streamflow when forest coverage had been reduced (Brown et al., 2005).
Author's personal copy A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
The positive association between trout abundance and native second-growth forest cover in 1000 m ! 60 m buffers along streams documented by this study may be explained by complex interactions between biotic and abiotic characteristics in the stream habitat associated with the forested riparian zone that favor trout populations (higher abundance of invertebrates as food supply, lower temperature, higher oxygen concentration and lower turbidity). Higher water quality in streams associated to watersheds in the Andean Range, covered extensively with native forests, have been mentioned as an explanation for higher trout abundance compared to that reported for the Coastal Range and the Central Depression (Soto et al., 2006). Studies in the Northern Hemisphere show that forests with riparian buffers, resulted in a different forest invertebrate and aquatic invertebrate communities (Kiffney et al., 2003; Rykken et al., 2007) than in areas without forested buffers. Trout prey on both terrestrial and aquatic invertebrate communities (Elliott, 1973; Klemetsen et al., 2003) and they are most likely dependent on the communities associated to second-growth forests. The negative association between native fish species abundance and second-growth forest cover in buffers is consistent with results by Soto et al. (2006) for entire watersheds. This pattern may be explained by a higher abundance of trout in these streams that likely displaced the native fishes due to competition, as documented for New Zealand (McIntosh et al., 1992; Glova, 2008), and due to interactive segregation, as reported in Chile (Penaluna et al., in press). We were not able to explain individual native fish species abundance, since this is beyond the scope of this study and also because numbers for determined species were in some cases too low. Our results regarding streamflow and trout abundance as indicators of water supply and recreational fishing opportunities as forest ecosystem services have clear policy implications. Given the positive linear relationship between native forest cover and water supply, and the inverse pattern for exotic plantations, increasing the proportion of native forest stands through restoration would increase this ecosystem service, especially in the summer. Nevertheless, pine and eucalypt plantations have advantages for timber production due to their fast growth and high demand for roundwood by the forest industry in Chile and other countries (Sedjo, 2001; Cubbage et al., 2007). Results presented here suggest that ecosystem service outputs from native forests might change societal preferences toward their conservation. The relative quantity and value of water supply from native forest versus the value of industrial wood production from fast growing exotic plantations will determine land use decisions in the various watersheds. The value of goods and services provided by other land uses such as agriculture will also influence these decisions. Water supply is becoming increasingly demanded by society (Lara et al., 2008); nonetheless, the tradeoffs of these benefits from native forests must be balanced against intensive exotic wood plantations that are economically important but ˜ a, 1995; Oyarzu´n supply less, lower quality water (Oyarzu´n and Pen et al., 2005; Huber et al., 2008). A different situation occurs for managed native Nothofagus second-growth stands, since our results indicate the compatibility of timber production and water supply, also reported by Nahuelhual et al. (2007). Conservation and restoration of native forests in the buffers along streams and rivers might be important to increase the abundance of introduced trout and the recreational fishing opportunities (Arismendi and Nahuelhual, 2007). The growing demand and economic value of this ecosystem service would provide support for the conservation of native forests (Arismendi and Nahuelhual, 2007). This paper indicates the importance of ecosystem services from native forests for society, and the need for a policy toward forest
423
conservation and sustainable management. The increase of water supply from native forests is important due to the recent and projected growth in water demand in southern Chile for human consumption, irrigation, tourism, introduced salmon farming and hydropower generation (Lara et al., 2008). The law on native forests approved by the Chilean congress in July 2008, which includes incentives for thinnings, restoration and protection, should be an important instrument to promote the management and conservation of these forests. Results from this study emphasize the role that such incentives might have for the mitigation of precipitation and streamflow decrease related to climatic change. The increase of streamflow in managed native forest stands may counteract the 20–40% decrease in precipitation and streamflow related to global climatic change that has been described for south-central Chile since the early 1900s (Trenberth et al., 2007; Lara et al., 2008). Despite the clear relationship between native forest cover and water supply, as well as recreational fishing opportunities as ecosystem services, the low percentage area of old-growth forests in the different watersheds prevented a better understanding of their role in providing these services. This analysis was not possible since most of the native forests in the Coastal Range and Central Depression are second-growth stands due to past human disturbance. Future studies addressing the assessment of ecosystem services provided by old-growth forests compared to secondgrowth forests in other study areas and how these services change with stand age are necessary. Results from this study show tradeoffs between water supply and wood production in watersheds covered by a combination of native forests and exotic plantations. At the same time, the conservation of native fish and trout-based recreational fishing appear incompatible in the same streams. Therefore, studies on these compatibilities and the interdependences among different ecosystem services are needed. Quantifying the attributes of ecosystems that contribute to service provision is crucial to provide guidelines for land management and policy development. However, such studies are extremely rare in the scientific literature (Luck, 2008). This study quantified and estimated the rates of change of ecosystem services as a function of variations in native forest cover for specific watersheds and environmental conditions in south-central Chile. Similar methods would provide valuable information for other regions in Chile and other countries of the world, promoting new opportunities for forest conservation and sustainable management.
Acknowledgements This research was supported by FORECOS Scientific Nucleus (P04-065-F), Fondecyt grant nos. 1020183 and 1050298, and a grant from the Inter-American Institute for Global Change Research (IAI) CRN II #2047, which is supported by the US National Science Foundation (Grant GEO-0452325). We also thank the valuable comments from two anonymous reviewers. Natalia Carrasco and Aldo Farı´as helped with the figures. References Alaback, P., 1991. Comparative ecology of temperate rainforests of the Americas along analogous climatic gradients. Revista Chilena de Historia Natural 64, 399–412. Arismendi, I., Nahuelhual, L., 2007. Non-native salmon and trout recreational fishing in lake Llanquihue, southern Chile: Economic benefits and management implications. Reviews in Fisheries Sciences 15, 311–325. Armesto, J., Rozzi, R., Smith-Ramirez, C., Arroyo, M.K., 1998. Conservation targets in South American temperate forests. Science 282, 1271–1272. Bent, G.C., 2001. Effects of forest-management activities on runoff components and ground-water recharge to Quabbin Reservoir, central Massachusetts. Forest Ecology and Management 143, 115–129.
Author's personal copy 424
A. Lara et al. / Forest Ecology and Management 258 (2009) 415–424
Brown, A., Zhang, E., McMahon, L., Western, A.W., Vertessy, R.A., 2005. A review of paired catchment studies for determining changes in water yield resulting from alterations in vegetation. Journal of Hydrology 310, 28–61. Calder, I.R., Rosier, P.T.W., Prasanna, K.T., Parameswarappa, S., 1997. Eucalyptus water use greater than rainfall input a possible explanation from southern India. Hydrological & Earth System Science 1, 249–256. Chapman, T., 1999. A comparison of algorithms for stream flow recession and baseflow separation. Hydrological Processes 3, 701–714. Chapman, T., Maxwell, A., 1996. Baseflow separation—comparison of numerical methods with tracers experiments. In: Hydrology and Water Resources Symposium, Institution of Engineers Australia, Hobart, pp. 539–545. CONAF, CONAMA, BIRF (Corporacio´n Nacional Forestal, Comisio´n Nacional del Medio Ambiente and Banco Interamericano de Reconstruccio´n y Fomento), 1999. Proyecto catastro y evaluacio´n de los recursos vegetacionales nativos de Chile. Informe regional De´cima Regio´n. Costanza, R., D’ Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naem, S., O’Neill, R., Paruelo, J., Raskin, R., Sutton, P., Van den Belt, M., 1997. The value of the world’s ecosystem services and natural capital. Nature 387, 253–260. Cubbage, F., Mac Donagh, P., Sawinski, J., Rubilar, R., Donoso, P., Ferreira, A., Hoeflich, V., Morales, V., Ferreira, G., Balmelli, G., Siry, J., Ba´ez, M., Alvarez, J., 2007. Timber investment return for selected plantations and native forest in South America and the southern United States. New Forests 33, 237–255. Donoso, C., 1993. Ecologı´a de los bosques templados de Chile y Argentina. Editorial Universitaria, Santiago, Chile. Elliott, J.M., 1973. The food of brown and rainbow trout (Salmo trutta and S. gairdneri) in relation to the abundance of drifting invertebrates in a mountain stream. Oecologia (Berlin) 12, 329–347. Echeverrı´a, C., Coomes, D., Salas, J., Rey Benayas, J.M., Lara, A., Newton, A., 2006. Rapid deforestation and fragmentation of Chilean temperate forest. Biological Conservation 130, 481–494. Farley, K., Jobba´gy, E., Jackson, R., 2005. Effect of afforestation on water yield: a global synthesis with implications for policy. Global Change Biology 11, 1565– 1576. Glova, G.J., 2008. A test for interaction between brown trout (Salmo trutta) and inanga (Galaxias maculatus) in an artificial stream. Ecology of Freshwater Fish 12, 247–253. Huber, A., Iroume´, A., Bathurst, J., 2008. Effect of Pinus radiata plantation on water balance in Chile. Hydrological Processes 22, 142–148. Jackson, R.B., Jobbagy, E.G., Avissar, R., Roy, S.B., Barrett, D.J., Cook, C.W., Farley, K.A., Le Maitre, D.C., NcCarl, B.A., Murria, B.C., 2005. Trading water for carbon with biological carbon sequestration. Science 210, 1944–1947. Kiffney, P.M., Richardson, J.S., Bull, J.P., 2003. Responses of periphyton and insects to experimental manipulation of riparian buffer width along forest streams. The Journal of Applied Ecology 40, 1060–1076. Klemetsen, A., Amundsen, P.A., Dempson, J.B., Jonsson, B., Jonsson, N., O’Connel, M.F., Mortensen, E., 2003. Atlantic salmon Salmo salar L., brown trout Salmo trutta L. and Artic charr Salvelinus alpinus (L.): a review of aspects of their life histories. Ecology of Freshwater Fish 12, 1–59. Lara, A., Veblen, T.T., 1993. Forest plantations in Chile: a successful model? In: Mather, A. (Ed.), Afforestation: Policies Planning and Progress. Belhaven Press, Londres, pp. 118–139. Lara, A., Villalba, R., 1993. A 3620-year temperature record from Fitzroya cupressoides tree-rings in southern South America. Science 260, 1104–1106. Lara, A., Sandoval, V., 2003. Generacio´n de Cartografı´a y evaluacio´n de recursos vegetacionales. In: Oltremari, J. and Thelen K. D. (Eds.), Planificacio´n de A´reas silvestres protegidas. Un manual para la planificacio´n de a´reas protegidas en Chile con especial referencia a a´reas protegidas privadas. CONAMA & FAO. Santiago, Chile, pp. 48–63. Lara, A., Soto, D., Armesto, J., Donoso, P., Wernli, C., Nahuelhual, L. and Squeo, F., 2003. Componentes cientı´ficos clave para una polı´tica nacional sobre usos, servicios y conservacio´n de los bosques nativos Chilenos. Universidad Austral de Chile, Valdivia. Lara, A., Reyes, R., Urrutia, R., Bosques Nativos, 2006. In: Instituto de Asuntos Pu´blicos, Universidad de Chile (Eds.), Informe Paı´s: Estado del Medio Ambiente en Chile, Santiago, Chile, pp. 107–139. Lara, A., Villalba, R., Urrutia, R., 2008. A 400-year tree-ring record of the Puelo River summer–fall streamflow in the Valdivian rainforest eco-region, Chile. Climatic Change 86, 331–356. Le Roux, J.P., Elgueta, S., 2000. Sedimentologic development of a Late Oligocene– Miocene forearc embayment, Valdivia Basin Complex, southern Chile. Sedimentary Geology 130, 27–44. Levi, B., Aguilar, A., Fuenzalida, R., 1966. Reconocimiento geolo´gico de las Provincias de Llanquihue y Chiloe´, Boletı´n N8 19. Instituto de Investigaciones Geolo´gicas, Santiago, Chile. Little, C., Soto, D., Lara, A., Cuevas, J., 2008. Nitrogen exports at multiple-scales in a southern Chilean watershed (Patagonian Lakes district). Biogeochemistry 87, 297–309. Lillesand, T., Kiefer, R., 1994. Remote sensing and image interpretation, third ed. John Wiley and Sons, New York, USA.
Lobo´n-Cervia´, 1991. Dina´mica de poblaciones de peces en rı´os, pesca ele´ctrica y me´todos de capturas sucesivas en la estima de abundancia. Centro de Investigacio´n de Agua, Madrid, Spain. Luck, G., Quantifying the contribution of organisms to the provision of ecosystem services: The SPU concept, 2008. In: Grant, F., Young, J., Harrison, P., Sykes, M., Skourtos, M., Rounsevell, M., Kluva´nkova´-Oravska´, T., Settele, J., Musche, M., Anton, C., Watt, A. (Eds.), Ecosystem Services and Drivers of Biodiversity Change. Report of the RUBICODE electronic conference. Lucke, S., 2008. Approaches to ecosystem service assessment in forest ecosystems. In: Grant, F., Young, J., Harrison, P., Sykes, M., Skourtos, M., Rounsevell, M., Kluva´nkova´-Oravska´, T., Settele, J., Musche, M., Anton, C., Watt, A. (Eds.), Ecosystem Services and Drivers of Biodiversity Change. Report of the RUBICODE electronic conference. Millennium Ecosystem Assessment, 2003. Ecosystems and Human Well-being: A Framework for Assessment. World Resources Institute, Washington, DC. McIntosh, A., Towsend, C., Crowl, T., 1992. Competition for space between introduced brown trout (Salmo trutta L.) and native galaxiid (Galaxias vulgaris Stokell) in a New Zealand stream. Journal of Fish Biology 41, 63–81. Miller, A., 1976. The climate of Chile. In: Schwerdtfeger, W. (Ed.), World Survey of Climatology. Climates of Central and South America. Elsevier, Amsterdam, The Netherlands, pp. 113–131. ˜ ez, D., Oyarzu´n, C., Neira, E., 2007. Valuing Nahuelhual, L., Donoso, P., Lara, A., Nu´n ecosystem services of Chilean temperate rainforests. Environment, Development and Sustainability 9, 481–499. Olson, D., Dinerstein, E., 1998. The Global 200: a representation approach to conserving the earth’s most biologically valuable ecoregions. Conservation Biology 12, 502–515. Otero, L., Contreras, A., Barrales, L., 1994. Efectos ambientales del reemplazo de bosque nativo por plantaciones (Estudio en cuatro microcuencas en la provincia de Valdivia). Ciencia e Investigacio´n Forestal 8, 252–276. Oyarzu´n, C., Pen˜a, L., 1995. Soil erosion and overland flow in forested areas with pine plantations at coastal mountain range, central Chile. Hydrological Processes 9, 111–118. Oyarzu´n, C., Huber, A., 1999. Water balance in young plantations of Eucalyptus globulus and Pinus radiata in southern Chile. Terra 17, 35–44. Oyarzu´n, C., Godoy, R., Sepu´lveda, A., 1998. Water and nutrient fluxes in a cool temperate rainforest at the Cordillera de la Costa in southern Chile. Hydrological Processes 12, 1067–1077. ˜ ez, D., 2005. Los servicios ecosiste´micos del bosque Oyarzu´n, C., Nahuelhual, L., Nu´n templado lluvioso: produccio´n de agua y su valoracio´n econo´mica. Ambiente y Desarrollo 20, 88–95. B. Penaluna, I. Arismendi, D. Soto. Evidence of interactive segregation between introduced trout and native fishes in Northern Patagonian Rivers, Chile. Transaction of American Fisheries Society, in press. Pezoa, L.S., 2003. Recopilacio´n y ana´lisis de la variacio´n de las temperatures (perı´odo 1965–2001) y las precipitaciones (perı´odo 1931–2001) a partir de la informacio´n de estaciones meteorolo´gicas de Chile entre los 338 y 538 de latitud sur. Forestry Engineering Thesis. Universidad Austral de Chile, Valdivia, Chile. Rykken, J., Moldenke, A.R., Olson, D., 2007. Headwater riparian forest-floor invertebrate communities associated with alternative forest management practices. Ecological Applications 17, 1168–1183. Schulz, E.F., 1989. Problems in Applied Hydrology. Water Resources Publications, CO, USA. Sedjo, R.A., Goetzl, A., Moffat, S.O., 1999. Sustainability of Temperate Forests. Resources for the Future. Washington, DC, 102 p. Sedjo, R.A., 2001. The role of forest plantations in the word’s future timber supply. The Forestry Chronicle 77 (2), 221–225. Servicio Nacional de Geologı´a y Minerı´a, 1982. Mapa Geolo´gico de Chile escala 1:1,000,000. Santiago, Chile. Soto, D., Arismend, I., Gonzalez, J., Sanzana, J., Jara, F., Jara, C., Guzma´n, E., Lara, A., 2006. southern Chile, trout and salmon country: invasion patterns and threats for native species. Revista Chilena de Historia Natural 79, 97–117. Scott, D., Lesch, W., 1997. Streamflow responses to afforestation with Eucalyptus grandis and Pinus patula and to felling in the Mokobulaan experimental catchments, South Africa. Journal of Hydrology 199, 360–377. Smith-Ramı´rez, C., 2004. The Chilean coastal range: a vanishing center of biodiversity and endemism in South American temperate rainforests. Biodiversity and Conservation 13, 373–393. Trenberth, K.E., Jones, P.D., Ambenje, P., Bojariu, R., Easterling, D., Klein Tank, A., Parker, D., Rahimzadeh, F., Renwick, J.A., Rusticucci, M., Soden, B., Zhai, P., 2007. Observations: Surface and Atmospheric Climate Change. In: Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., Miller, H.L. (Eds.), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Ward, A.D., Trimble, S.W., 2004. Environmental Hydrology. Lewis Publishers/CRC Press Company, London/New York, 472 pp.