B43A-01
Spatial and Temporal Patterns of Throughfall Amounts and Solutes in a Tropical Montane Forest - Comparisons with Findings From Lowland Rain Forests
The diverse tree species composition, irregular shaped tree crowns and a multi-layered forest structure affect the redistribution of rainfall in lower montane rain forests. In addition, abundant epiphyte biomass and associated canopy humus influence spatial patterns of throughfall. The spatial variability of throughfall amounts controls spatial patterns of solute concentrations and deposition. Moreover, the living and dead biomass interacts with the rainwater during the passage through the canopy and creates a chemical variability of its own. Since spatial and temporal patterns are intimately linked, the analysis of temporal solute concentration dynamics is an important step to understand the emerging spatial patterns. I hypothesized that: (1) the spatial variability of volumes and chemical composition of throughfall is particularly high compared with other forests because of the high biodiversity and epiphytism, (2) the temporal stability of the spatial pattern is high because of stable structures in the canopy (e.g. large epiphytes) that show only minor changes during the short term observation period, and (3) the element concentrations decrease with increasing rainfall because of exhausting element pools in the canopy. The study area at 1950 m above sea level is located in the south Ecuadorian Andes far away from anthropogenic emission sources and marine influences. Rain and throughfall were collected from August to October 2005 on an event and within-event basis for five precipitation periods and analyzed for pH, K, Na, Ca, Mg, NH4+, Cl-, NO3-, PO43-, TN, TP and TOC. Throughfall amounts and most of the solutes showed a high spatial variability, thereby the variability of H+, K, Ca, Mg, Cl- and NO3- exceeded those from a Brazilian tropical rain forest. The temporal persistence of the spatial patterns was high for throughfall amounts and varied depending on the solute. Highly persistent time stability patterns were detected for K, Mg and TOC concentrations. Time stability patterns of solute deposition were somewhat weaker than for concentrations for most of the solutes. Epiphytes strongly affected time stability patterns in that collectors situated below thick moss mats or arboreal bromeliads were in large part responsible for the extreme persistence with low throughfall amounts and high ion concentrations (H+ showed low concentrations). Rainfall solute concentrations were low compared with a variety of other tropical lowland and montane forest sites and showed a small temporal variability during the study period for both between and within-event dynamics, respectively. Throughfall solute concentrations were more within the range when compared with other sites and showed highly variable within-event dynamics. For most of the solutes, within-event concentrations did not reach low, constant concentrations in later event stages, rather concentrations fluctuated (e.g. Cl-) or increased (e.g. K and TOC). The within-event throughfall solute concentration dynamics in this lower montane rain forest contrast to recent observations from lowland tropical rain forests in Panama and Brazil. The observed within-event patterns are attributed (1) to the influence of epiphytes and associated canopy humus, and (2) to low rainfall intensities.
B43A-02
BIOGEOGRAPHY, CLOUD BASE HEIGHTS AND CLOUD IMMERSION IN TROPICAL MONTANE CLOUD FORESTS
Tropical Montane Cloud Forests (TMCFs) are ecosystems characterized by frequent and prolonged immersion within orographic clouds. TMCFs often lie at the core of the biological hotspots, areas of high biodiversity, whose conservation is necessary to ensure the preservation of a significant amount of the plant and animal species in the world. TMCFs support islands of endemism dependent on cloud water interception that are extremely susceptible to environmental and climatic changes at regional or global scales. Due to the ecological and hydrological importance of TMCFs it is important to understand the biogeographical distribution of these ecosystems. The best current list of TMCFs is a global atlas compiled by the United Nations Environmental Program (UNEP). However, this list is incomplete, and it does not provide information on cloud immersion, which is the defining characteristic of TMCFs and sorely needed for ecological and hydrological studies. The present study utilizes MODIS satellite data both to determine orographic cloud base heights and then to quantify cloud immersion statistics over TMCFs. Results are validated from surface measurements over Northern Costa Rica for the month of March 2003. Cloud base heights are retrieved with approximately 80m accuracy, as determined at Monteverde, Costa Rica. Cloud immersion derived from MODIS data is also compared to an independent cloud immersion dataset created using a combination of GOES satellite data and RAMS model simulations. Comparison against known locations of cloud forests in Northern Costa Rica shows that the MODIS-derived cloud immersion maps successfully identify these cloud forest locations, including those not included in the UNEP data set. Results also will be shown for cloud immersion in Hawaii. The procedure appears to be ready for global mapping.
B43A-03
Base Metal Cycling in a Tropical Montane Forest in Ecuador
Earlier work had shown that Ca and Mg deposited from the atmosphere were partly taken up by the canopy of an Ecuadorian mountain forest at the western rim of the Amazon basin and that both elements were immobilized during incubation of acid soil organic layer material. We therefore determined all base metal fluxes from atmospheric input through the forest ecosystem to output with surface runoff in three small catchments under tropical montane forest between 1850 and 2200 m above sea level. We found a large interannual variation in the deposition of Ca and Mg from the atmosphere. This variation was related with the ENSO cycle. During La Nina conditions, Saharan dust was transported via the Amazon basin to our study site. Our ecosystem responded by accumulation of base metals likely because of nutrient accretion by increased growth. During strong forest fires in the Amazon basin, increased acid input was observed at our study site. This resulted in the export of base metals and counteracted the base metal accumulation during years of high base metal input. Base metal loss was promoted by fast near-surface flow flushing the base metal-enriched surface layers of the soil (particularly the organic layer) during rainstorms. Base metal loss during baseflow conditions was attributed to the weathering of the Ca- and Mg-poor bedrock and an unavoidable leaching of Ca and Mg with a similar size as weathering. The low availability of Ca and Mg in the studied ecosystem and the positive response of the studied forest to Ca and Mg inputs suggests that at this probably neither N- nor P-limited site base metals played a key role for forest performance. As the Ca and Mg cycles seem to be linked to large-scale events such as the ENSO phenomenon, forest burning in the Amazon basin, and climatic changes (particularly those influencing the storm frequency and intensity and thus the importanc of near-surface flow), future environmental change might have an impact on the stability of the precious, highly biodiverse north Andean mountain forests.
B43A-04
Land use Controls on Water Quality and Aquatic Ecosystems in the Andean Amazon, Peru
Agro-pastoral systems are replacing many of the tropical forests in the world, and much of this deforestation occurs in watersheds where people's livelihoods rely directly on water and aquatic resources in local streams and rivers. We examined relationships between land use and aquatic ecosystems in 34 catchments exhibiting a gradient of land use disturbance and human settlement in the Andean Amazon of Peru. Our research objectives were to 1) classify and characterize watershed land use and physical properties using remotely sensed data, 2) characterize the physical, biological and chemical conditions of streams 3) examine relationships between land use and water quality parameters at 3 scales: watershed, riparian and reach habitat and 4) translate research findings into management strategies that minimize disturbance and maximize ecosystem services. Primary forest was the dominant cover (68%) across the catchments, and the remaining study area was composed of: mixed forest (11.5%), grassland (10.75%) cropland (9.17%) and bare rock (0.02%). Among watersheds, forest cover ranged from 14% to 100%, mixed forest ranged from 0% to 26%, grassland ranged from 0% to 45% and cropland ranged from 0% to 26%. Physical habitat index scores ranged from 12.80 (very impaired) to 29.50 (reference conditions). Although nitrogen, phosphorous, dissolved organic carbon (DOC), pH, electric conductivity, and dissolved oxygen (DO), varied across sites, total concentrations remained within acceptable levels. Simpson's diversity and Family Biotic Index (FBI) were calculated for macroinvertebrates collected at each site; Simpson's diversity ranged from 0.24 to 0.95 and FBI ranged from 1.97 (excellent) to 7.49 (fairly poor). Forest cover at the watershed scale was the best explanatory variable and was positively correlated with inorganic phosphate (0.50), DO (0.90) and Simpson's diversity (0.46) and negatively correlated with organic phosphate (- 0.64), DOC (-0.75), water temperature (-0.89), electric conductivity (-0.62) and the FBI (-0.82). Of the managed land uses, riparian grassland was found to have the most significant negative impact on water quality and biologic communities. Our results demonstrate the benefits of maintaining forest cover and promoting sustainable agriculture over ranching in riparian zones.