Microsoft word - jour-13-f

Tropical Ecology 53(2): 235-240, 2012
International Society for Tropical Ecology www.tropecol.com Differences in soil moisture, nutrients and the microbial community
between forests on the upper Pacific and Caribbean slopes at
Monteverde, Cordillera de Tilaran: implications for
responses to climate change
WILLIAM D. EATON1*, MELANIE ROED2,3, OLIVIER CHASSOT4 & DWIGHT BARRY2.3 1School of Environmental and Life Sciences, Kean University, 1000 Morris Ave., 2Peninsula College, Center of Excellence in Environmental Science and Natural Resources, 1502, E. Lauridsen Blvd. Port Angeles, WA 98362 USA 3Huxley College of the Environment at Peninsula College, Western Washington University, 1502 E. Lauridsen Blvd. Port Angeles, WA 98362 USA 4Centro Científico Tropical (CCT), Apdo 8-3870-1000 San José, Costa Rica Abstract: A study was conducted to identify soil ecosystem metrics for use in monitoring
soil ecosystem health in tropical montane cloud forest habitats, and to establish a baseline of soil community characteristics. This work was performed in six unique forested habitats on the Caribbean and Pacific slopes of the Monteverde Reserve in Costa Rica. Comparisons were made of the soil moisture, nitrogen fixation activity, microbial biomass, fungal and bacterial abundance and diversity, and the abundance of key functional genes laccase (for lignin degradation by basidiomycete fungi) and nifH (for bacterial N- fixation). Differences were found in these metrics between habitats, which were correlated with soil moisture. The results of this work show that these metrics can be used as part of a program to monitor the soil ecosystems for early indicators of shifts in conditions in response to environmental change within this part of a fragile ecosystem.
Resumen: Se realizó un estudio para identificar las métricas del ecosistema edáfico que
pueden usarse en el monitoreo de la salud de este ecosistema en hábitats de bosque nublado montano y para establecer información de línea base de las características de la comunidad del suelo. Este trabajo se llevó a cabo en hábitats forestales únicos en las laderas del Caribe y del Pacífico de la Reserva Monteverde, Costa Rica. Se hicieron comparaciones de la humedad del suelo, la actividad de fijación de nitrógeno, la biomasa microbiana, la abundancia y la diversidad de hongos y bacterias, y la abundancia de los genes funcionales clave de la lacasa (para la degradación de la lignina por hongos basidiomicetos) y nifH (para la fijación bacteriana de N). Se encontraron diferencias para estas métricas entre hábitats, las cuales estuvieron correlacionadas con la humedad del suelo. Los resultados de este trabajo muestran que estas métricas pueden usarse en programas de monitoreo de ecosistemas edáficos como indicadores tempranos de modificaciones en las condiciones como respuesta al cambio ambiental en este componente de un ecosistema frágil.
Resumo: Este estudo foi conduzido para identificar as métricas do ecossistema do solo para
uso na monitorização da saúde do ecossistema em habitats florestais tropicais montanos de nevoeiros, e estabelecer uma linha básica quanto às características da comunidade do solo. Este trabalho foi efectuado em seis habitats florestados únicos nas vertentes das Caraíbas e do * Corresponding Author; e-mail: [email protected] Pacífico na Reserva Monteverde na Costa Rica. Foram efetuadas comparações na humidade do solo, abundância fúngica e bacteriana e diversidade, bem como a abundância de genes funcionais chave de lacase (para a degradação da lenhina pelo basidiomicetas) e nifH (para as bactérias de fixação de N). Encontraram-se diferenças nestas métricas entre habitats, os quais estavam correlacionadas com a humidade do solo. Os resultados deste trabalho mostraram que estas métricas podem ser utilizadas como parte de um programa para monitorizar os ecossistemas do solo como indicadores precoces de mudanças nas suas condições em resposta à mudança climática no seio desta parte de um ecossistema frágil. Key words: Climate change, microbial biomass, microbial community structure,
monteverde, montane forest, soil moisture. Tropical montane cloud forests require both diameter soil cores were collected, pooled by rain and frequent immersion in clouds for their subplot, and the soil sieved while field moist. The maintenance. However, the cloud forests of Monte- soil was composed of highly decomposed materials, verde, Costa Rica, are currently threatened by and all soil cores contained only organic layer documented increases in mist-free periods (Lawton material. All data were adjusted per dry weight of et al. 2001; Pounds & Puschendorf 2004; Ray et al. the soil and bulk density, and results are presen- 2006). Modeling studies there also have provided ted per volume of soil. Each of the four subplot soil significant reason to believe that orographic cloud samples per forest type were analyzed for levels of base height is increasing, that frequency of forest microbial biomass C (Cmic), total mineral nitrogen immersion in cloud is decreasing, and that the (TMN), the PCR-based relative percent contri- area immersed in clouds for significant times is bution (RPC) of bacterial rRNA (primers 27f/ decreasing (Lawton et al. 2001; Ray et al. 2006). In 1492r), fungal rRNA (primers EF4/Fung5), nitrogen response to this concern, the first study was fixation gene (primers nifHf/nifHr), and laccase conducted in these forests to compare soil mois- gene (primers lacIf/lacII), and the restriction frag- ture, nitrogen (N) fixation activity, microbial bio- ment length polymorphism (RFLP)-based relative mass, fungal and bacterial abundance and diver- diversity and abundance of the bacterial and fungal sity, and the abundance of key functional genes community as described by Eaton et al. (2011). laccase (for lignin degradation by basidiomycete fungi) and nifH (for bacterial N- fixation). This differences by 2-tailed t-test and mean effect size study is part of a larger project to monitor these (d) analysis. To account for the small sample sizes, forests for impacts of decreasing precipitation a weight-of-evidence approach was used in which (Salazar et al. 2009). As such, the goals were to mean differences with P values < 0.15, or 0.15 to identify metrics for use in monitoring soil ecosys- < 0.25 and d values > 0.9 were considered to repre- tem health in these habitats over time, and to sent meaningful differences between two mean begin establishing a baseline of soil community values. The RFLP data were used to determine the Shannon-Weiner diversity index (H′ by habitat Six naturally occurring and undisturbed mon- type), and for cluster analysis using a similarity tane cloud forest habitats, representing various matrix and Ward’s minimum variance method of Hartshorn (1983) life zones, were identified in the measuring Euclidean distances in NCSS software. Monteverde Reserve, from various life zones, three Metrics most influenced or influenced by Cmic and on the Caribbean slope (El Valle, Aleman, Eladios) soil moisture were identified as critical corre- and three on the Pacific (Brillantes, Pantanoso, lations for application in future monitoring and Alondra sites) slope. The georeferences, altitude model development if correlation coefficient r values and Hartshorn (1983) life zones are provided in (at P = 0.1) were > 0.2 or < -0.2, considered medium Table 1. In each habitat, a single 50 x 50 m plot was established, divided into four equal subplots, The habitats with the greatest soil moisture on and, following removal of the upper layer of soil both the Pacific and Caribbean slopes (Brillantes debris, 16 randomly located 10 cm deep and 2 cm and El Valle, respectively) had the most fungal- Table 1. Summary of the sample sites, georeferences, altitude, and Hartshorn (1983) life zones.

Fig. 1. Dendrograms comparing the relatedness of the RFLP-based diversity of bacterial and fungal commu-
nity rRNA in soil collected from Pacific slope (ALO, BRI, PAN) and Caribbean slope forests (ALE, ELA, VAL).
(A) Bacterial rRNA diversity, (B) Fungal rRNA diversity.
dominated soils, with greater fungal to bacterial
closely related to those from the Caribbean slope rRNA ratios, greater amount of laccase and (gene- soils (Fig. 1A & 1B). Interestingly, the Brillantes rally) fungal rRNA genes, the lowest amount of and Alondra soils also had significantly greater bacterial rRNA gene, with less bacterial and greater Cmic values than the other four soils, suggesting fungal rRNA diversity. These habitats also had the that the diversity and structure of the microbial lowest levels of total mineral N and greatest amount populations might differ in correlation with C cycle of nifH gene, which, along with the generally dynamics. As well, the RFLP data showed that greater biomass, suggests that more active nitrogen there were clearly unique fungal and bacterial fixation and incorporation of mineral N into the population components in the more moist soils biomass is occurring in these forests (Tables 2 & (Fig. 1A & 1B). When examined along with the 3). The diversity of both the bacterial and fungal RPC, N and biomass data, this suggests the possi- rRNA of the Brillantes and Alondra soils from the bility that selection is occurring for N-fixing bacteria Pacific slope were closely related, as were the and basidiomycete fungi (and lignin degradation) Eladios and Aleman soils of the Caribbean slope. in the more moist soils of forests from both slopes. Both types of rRNA from the Pantanoso (Pacific Soil moisture influences microbial biomass slope) and El Valle (Caribbean) soils were more (Arunachalam et al. 1997; Eaton 2001; Eaton et al. Table 2. The mean value, standard deviation (SD), P-value from t-test, and mean effect size (d) are presented
for soil moisture (as % moisture), total mineral nitrogen (TMN as µg g-1), microbial biomass carbon (Cmic as mg
C g-1), and the PCR-based relative percent contribution (RPC) of bacterial rRNA (RPC bact), fungal rRNA (RPC
fung), ratio of fungal to bacterial rRNA (Fung:bact), N- fixation gene (RPC nifH), and laccase gene (RPC lac)
from Monteverde soil samples collected from different forests on the from Pacific slope (Brillantes, Alondra,
Pantanoso) and Caribbean slope forests (Valle, Eladios, Aleman).
Table 3. The mass and diversity index (H′) of the
vated soil moisture levels increase the release of bacterial rDNA and fungal rDNA bands following organic matter from woody debris, stimulating an RFLP analysis of DNA products from PCR increase in microbial activity, nutrient utilization, amplification of soil DNA collected from Pacific slope and microbial biomass development, thus lowering (ALO, BRI, PAN) and Caribbean slope forests (ALE, the pools of inorganic N (Arunachalam et al. 1997; ELA, VAL) in the Monteverde cloud forest reserve. Eaton 2001; Eaton et al. 2011; Schulze 2004; The soil DNA was amplified by PCR methods using Schwendenmann & Veldkamp 2006). These factors the universal bacterial and fungal rRNA primers along with lower soil oxygen levels that occur with increased soil moisture positively regulate the nifH gene activity, which is also associated with biomass development (White et al. 2007). Conver- sely, lower levels of soil moisture decrease N cycle activities and microbial biomass, and result in increased concentrations of pools of inorganic N (Eaton 2001; Eaton et al. 2011; Ewing et al. 2007), which can inhibit both bacterial N-fixation and 1.47 90 1.38 lignin degradation by basidiomycetes, thus decrea- 1.62 92 1.25 sing the microbial biomass and sequestration of 1.54 90 1.31 organic C into the soil biota (Ewing et al. 2007; De Boer et al. 2005). The data from the current study suggest that these relationships have important 1.35 implications for soil ecosystem quality should 1.46 61 1.18 moisture levels decrease in the Monteverde forests. This was the first study to demonstrate that measurable differences existed in the soil nutrients, 2011; Schwendenmann & Veldkamp 2006), which microbial community, and microbial biomass in is also influenced by a wide variety of nutrient and the different habitats, that were correlated with microbial community components (Anderson 2003; soil moisture. Due to the concern of the observed Schulze 2004) in establishing healthy-functioning decreasing amounts of precipitation in the Mont- soils (Doran 2002). Thus, the metrics that most verde cloud forest region (Pounds et al. 1999), the influence or are influenced by soil moisture and metrics suggested by this work should be used as Cmic were considered as target indicators of “soil part of a program to monitor the soil ecosystems quality” in this study. The soil moisture was posi- early indicators of shifts in conditions in response tively correlated (at P = 0.1) with Cmic (r = 0.52), to environmental change within this part of a RPC of fugal rRNA (r = 0.21), fungal to bacterial rRNA ratio (0.22), laccase (0.27) and nifH genes (r = 0.26), and negatively correlated with the TMN Acknowledgements
levels (r = - 0.20) and RPC of bacterial rRNA (r = -0.24). The Cmic was positively correlated with soil This project was supported by the National moisture (r = 0.52), RPC of fungal rRNA (r = 0.21), Science Foundation grants DBI-0452328 and was fungal to bacterial rRNA ratio (0.22), laccase (0.27) conducted under the Costa Rican Government and nifH genes (r = 0.26), and negatively corre- Permit #063-2008-SINAC. The authors especially lated with the TMN levels (r = - 0.20) and RPC of wish to acknowledge and thank all the Tropical bacterial rRNA (r = - 0.24). Thus, it appears that Science Center staff, Brenda Campbell, Emily Giles, these metrics are good indicators of differences in and Robert S. J. Eaton for their incredible effort made in collecting soils samples in the field under Lower concentrations of mineralized N are the most difficult of circumstances. associated with increased nifH gene activity (Schulze References
2004), microbial biomass development (Eaton 2001; Eaton et al. 2011; Schwendenmann & Veldkamp Anderson, T. H. 2003. Microbial eco-physiological 2006), and increased bacterial decomposition which indicators to assess soil quality. Agriculture, Eco- triggers bottom-up processes leading to increases in systems and Environment 98: 285-293.
fungal abundance and diversity (Currie 1999). Ele- Arunachalam, K., A. Arunachalam, R. S. Tripathi & H. N. Pandey. 1997. Dynamics of microbial population 2001. Climatic impact of tropical lowland defore- during the aggradation phase of a selectively logged station on nearby montane cloud forests. Science subtropical humid forest in north-east India. 294: 584-587.
Tropical Ecology 38: 333-341.
Pounds, J. A. & R. Puschendorf. 2004. Clouded futures. Currie, W. S. 1999. The responsive C and N biogeo- Nature 427: 107-109.
chemistry of the temperate forest floor. Trends in Pounds, J. A., M. P. L. Fogden & J. J. Campbell. 1999. Ecology and Evolution 14: 316-320.
Biological response to climate change on a tropical De Boer, W., L. B. Folman, R. C. Summerbell & L. mountain. Nature 398: 611-615.
Boddy. 2005. Living in a fungal world: impact of Ray, D. K., U. S. Nair, R. O. Lawton, R. M. Welch & R. fungi on soil bacterial niche development. FEMS A. Pielke. 2006. Impact of land use on Costa Rican Microbiology Reviews 29: 795-811.
tropical montane cloud forests: Sensitivity of oro- Doran, J. W. 2002. Soil health and global sustainability: graphic cloud formation to deforestation in the plains. translating science into practice. Agriculture, Eco- Journal of Geophysical Research 111: 1-16.
systems and Environment 88: 119-127.
Salazar, M., O. Chassot, V. Meza, G. Hernández, V. Eaton, W. D. 2001. Microbial and nutrient activity in Jiménez, B. Eaton, C. Hernández, C. Mena & Y. soils from three different subtropical forest habitats Méndez. 2009. Establecimiento de parcelas perma- in Belize, Central America during the transition nentes de monitoreo en el transecto continental del from dry to wet season. Applied Soil Ecology 16:
bosque nuboso de Monteverde, Costa Rica. XIII Congress of the Mesoamerican Society for Biology Eaton, W. D., S. MacDonald, M. Roed, K. L. Vandecar, J. and Conservation, Belize City, Belize 26-30 October B. Hauge & D. Barry. 2011. A comparison of nutrient dynamics and microbial community charac- Schwendenmann, L. & E. Veldkamp. 2006. Long-term teristics across seasons and soil types in two CO2 production from deeply weathered soils of a different old growth forests in Costa Rica. Tropical tropical rain forest: evidence for a potential positive Ecology 52: 35-48.
feedback to climate warming. Global Change Biology Ewing, S. A., G. Michalski, M. Thiemens, R. C. Quinn, J. 12: 1878-1893.
L. Macalady, S. Kohl, S. D. Wankel, C. Kendall, C. Schulze, J. 2004. How are nitrogen fixation rates regu- P. Mckay & R. Amundson. 2007. Rainfall limit of lated in legumes? Journal of Plant Nutrition and the N cycle on Earth. Global Biogeochemical Cycles Soil Science 167: 125-137.
21, GB3009: (1-12).
White, J., J. Prell, E. K. James & P. Poole. 2007. Nu- Hartshorn, G. S. 1983. Plants. pp. 118-157. In: D. H. trient sharing between symbionts. Plant Physiology Janzen (ed.) Costa Rica Natural History. The University of Chicago Press, Chicago & London. 144: 604-614.
Lawton, R. O., U. S. Nair, R. A. Pielke & R. M. Welch. (Received on 09.11.2009 and accepted after revisions, on 15.07.2011)

Source: http://www.cct.or.cr/documentos/publicaciones/articulos_cientificos/differences_in_soil_moisture.pdf