Many of the microscopic bacteria and fungi that live in aquatic environments consume oxygen as they break down organic matter. This organic matter includes things such as tree leaves and wood, and the tea-like dissolved organic matter that gives our southern coastal plain rivers their dark color. Logically, one might assume that as more organic matter enters a river, more food and habitat are provided for microorganisms, and greater quantities of oxygen are depleted. Estimates of leaf litter inputs and the quantities of oxygen consumed by them are even used by policy makers who monitor oxygen concentrations in our rivers. However, these dynamics are rarely studied in blackwater rivers, and can be difficult to determine. Leaf inputs are not often measured directly, and values are often obtained from studies taking place in other rivers that may be quite different as far as the amounts and types of leaf litter entering them.
The relationship between leaf litter inputs and oxygen demand is not as simple as ____ amount of leaf litter = ____ amount of oxygen consumed. After a leaf is shed from a tree and has entered the aquatic environment, it is gradually colonized by bacteria and fungi. Their numbers may start out low and increase gradually, so the amount of oxygen consumed per leaf may also start out low and increase over time, and microbial activity also depends on temperature.
At the same time the populations of microorganisms are growing, the leaf is actually “getting smaller” in a sense. This is due to a combination of the “feeding” activity of bacteria and fungi, as well as larger macroinvertebrates (i.e. insects and crustaceans, Plate 1), and physical breakdown caused by flowing water. Some of the leaf material is simply converted to smaller fragments, some becomes a part of the organisms that are consuming it, and the remainder exits the river entirely as various gaseous forms of carbon. Therefore, the amount of oxygen consumption caused by leaves that enter the river in autumn and winter will change throughout the year as the leaves break down.
An additional level of complexity lies in the fact that leaves from different species of trees (Plate 2) differ in quality, from the perspective of microorganisms and macroinvertebrates.
Some tree leaves contain proportionately large amounts of tough structural compounds that make them difficult for microorganisms to break down, and less likely to be eaten by macroinvertebrates. For example, slash pine (Pinus elloittii) needles and water oak (Quercus nigra) leaves may not support as many bacteria and fungi as a more “nutritious” red maple (Acer rubrum) leaf, and are not consumed as rapidly by macroinvertebrates. Therefore, although red maple leaves may deplete more oxygen per gram of leaf, they do not last as long in the river, and while pine and oak leaves consume very little oxygen per gram, they may remain in the river for a very long time and may have more long-lasting effects.
My research addresses the role of leaf litter in blackwater river oxygen dynamics and takes the relationships described above into account. My research can be broken down into the following parts:
Part I. Forest Composition and Oxygen Uptake Rates:
We selected two different sections of a blackwater river known to experience low oxygen events within the Little River Experimental Watershed (LREW).
The “stream” section (Plate 3) is closer to the headwaters (3rd-order), fairly narrow (roughly 5-10 meters wide), has higher water velocity, and only occasionally drops below government standards for dissolved oxygen. The in-stream “swamp” section (Plate 4) (5th-order) is much wider (roughly 400 meters), flows very slowly, and often violates government standards for dissolved oxygen. The two rivers are similar in all other physiochemical parameters such as nutrient levels and temperature.
First, I studied the forest composition in the two sites to determine which trees were most common (Cottam and Curtis 1956).
In the autumn, dry leaves of the five most-common tree species were collected immediately after they were shed from the trees, and known amounts of single tree species’ leaves were placed into mesh bags (Benfield 2006) after being given time to air dry.
These bags were then taken to the river and submerged, and allowed to be colonized by microorganisms and macroinvertebrates. Throughout the year, bags would be removed, and pieces of leaf material were placed into small glass chambers (Carter and Suberkropp 2004, Plate 5) that allowed us to measure the amount of oxygen consumed per gram of leaf.
We found significant differences among tree species in the amount of oxygen consumed (Fig 1), as well as significant effects of temperature, and changing uptake rates per gram over time (Fig 2).
Part 2. Microorganisms responsible for oxygen uptake
In order to determine which microorganisms were responsible for oxygen consumption, we collected leaf material from the same mesh bags at the same time that oxygen uptake was being measured.
To measure the amount of fungi present, we extract a component of the fungal cell membrane, known as ergosterol, directly from colonized leaf litter (Gessner and Schmitt 1996). Fungi appear to be responsible for much of the oxygen consumption taking place in the stream site, as their biomass correlates well with oxygen uptake rates per gram of leaf (Figure 3).
However, in the swamp, we could not establish a good relationship between fungal biomass and oxygen uptake rates, although fungal biomass was equal to or higher than that found in the stream (Figure 4)
Aquatic fungi live primarily in the interior of leaves, while bacteria mainly coat the surfaces of objects, forming something referred to as a biofilm. It is possible that in the swamp, where oxygen concentrations are lower and water flows very slowly, surface-living microorganisms such as bacteria (Plate 6) utilize oxygen before it can reach the fungi.
There are several groups of aquatic fungi that can tolerate low-oxygen conditions, and it is possible that the types of fungi living in the swamp are different from those living in the stream.
In order to determine the numbers of bacteria living on our leaves, I use two methods. One involves staining them with a fluorescent dye (Plate 6) and counting them on a microscope (Patel et al 2007). Another graduate student and I are also working out a method to count leaf-associated bacteria on a flow cytometer (modified from Felip et al 2007), which allows us to count thousands of bacteria in a matter of seconds. These results may help to determine why fungal biomass does not match oxygen uptake more closely in the swamp.
Part 3. Leaf breakdown and macroinvertebrate effects
Policy makers also estimate the rates at which leaf litter breaks down after it enters the river, and apply this data to their models for oxygen demand. Again, this is rarely measured directly, and values from studies taking place in other rivers are often applied. From the same mesh bags that we measured oxygen uptake rates of leaves, we also collected all macroinvertebrates present and determined the rate at which the leaves were losing mass. After oxygen uptake was measured and fungal and bacterial samples were collected in the field, bags of leaves were returned to the laboratory where they were washed over nested sieves. Macroinvertebrates were removed and preserved to be indentified and measured later. Rinsed leaf material was dried in an oven at 60º, weighed, combusted at 500ºC, and weighed again to account for any accumulated sand and inorganic sediment that may have been adding weight to the leaves.
The leaves of different tree species break down at different rates in blackwater rivers, such as can be seen for these three species in our stream site (Figure 5). The area in between the vertical dashed lines represents a dry period, during which breakdown is relatively slow.
Leaf breakdown tends to be faster in the swamp site, primarily due to the abundance of macroinvertebrates found there (Figure 6). Notice that our most rapidly-decomposing leaf litter species, Ogeechee tupelo (Nyssa ogeche), is almost completely broken down during the first year in the swamp, while it lasts into the dry season in the small stream.
These results tells us that leaf breakdown rates used to model oxygen dynamics depend on the tree species present in an area, and also on the characteristics of the stream reach that the leaves are falling into. In an area where macroinvertebrates are abundant, leaves will break down more rapidly.
The fact that the leaves of many tree species take more than one year to decay in many blackwater rivers should also be of interest to policy makers. Rather than incorporating only the leaf inputs of the current year into oxygen demand models, it may be of interest to also take into account the additional leaf material remaining from the previous year.
Data from our experiment are supported by what is observable in the natural environment (Plate 7). The photograph above shows three distinct layers of organic matter in the swamp basin during the dry season. The fresh layer of leaf litter from the current year (A), the layer of old leaf litter (blackened from microbial processes under low oxygen conditions, B), and the consolidated layer of sediment and organic matter (C).
Part 4. Leaf inputs and breakdown on ecosystem scales
So far, everything covered here deals with processes on very small scales. In order to understand how these microscopic processes affect entire rivers we must scale these processes up. This involves knowing the oxygen uptake rates per gram of leaf, which has been explained previously, and applying these measurements to the total amount of leaf litter in the river channel.
Large networks of elevated baskets (Plate 8) were installed in both sites (roughly 50 collectors per site), measuring leaf inputs. Applying our estimated breakdown rates (from mesh bags) to the total inputs of leaves provides another method to estimate the total amount of leaf litter in the stream.
To measure changes in the amount of decaying leaf material remaining over time, we combined methods from Pozo and Elosegi (2005) and Carter and Suberkropp (2004) to determine the standing crop of leaf litter per square meter of stream and swamp bottom. Each month, square-meter plots of leaf litter and wood were removed from the stream and swamp bottom (15 plots per site). While plots were being sampled, leaves from the plots were used for more oxygen uptake measurements, and additional samples were collected for fungal and bacterial biomass.
Leaf samples were returned to the laboratory, where they are separated out according to species, dried, and weighed to determine the total amount of leaf litter in a section of river (Plate 9). By applying microbial biomass estimates, and oxygen uptake to the total mass of leaves, we may estimate the total microbial biomass and associated oxygen consumption that results from leaf decomposition in a section of river. These estimates may be compared to the measurements of sediment oxygen demand collected by other researchers (Jason Todd and Barb Utley) in the same sites, in order to determine how much of the total benthic oxygen demand is accounted for by decaying leaf material.
Through the combination of all our results, we hope to provide tools for policy makers to better estimate contributions of leaf litter to oxygen demand in blackwater rivers, as well as a better scientific understanding of the factors contributing to organic matter processing and oxygen dynamics in these understudied river systems.
Download: Andrew Mehring, R.R. Lowrance, G. Vellidis, A.M. Helton, C.M. Pringle, and D.D. Bosch. 2010. Is Elevated DOC a Result of Low DO Rather than its Cause? Poster presented at the 2010 Land Grant and Sea Grant National Water Conference, Hilton Head, SC, 23 February, 2010.
Andrew Mehring, Ph.D. candidate, Odum School of Ecology, University of Georgia.
George Vellidis, Professor, Biological and Agricultural Engineering, University of Georgia.
Catherine Pringle, Professor, Odum School of Ecology, University of Georgia.
Kevin Kuehn, Assistant Professor, Department of Biological Sciences, University of Southern Mississippi
Richard Lowrance, Ecologist, Southeast Watershed Research Laboratory, USDA-ARS, Tifton, Georgia.
Benfield, E. F. 2006. Decomposition of leaf material. Pages 711-720 in F. R. Hauer and G. A. Lamberti (editors). Methods in stream ecology. Academic Press, San Diego.
Carter, M. D. and K. Suberkropp. 2004. Respiration and annual fungal production associated with decomposing leaf litter in two streams. Freshwater Biology 49:1112 – 1122.
Cottam, G. and J. T. Curtis. 1956. The use of distance measures in phytosociological sampling.
Ecology 37: 451-460.
Felip, M., S. Andreatta, R. Sommaruga, V. Straksrabova, and J. Catalan. 2007. Suitability of flow cytometry for estimating bacterial biovolume in natural plankton samples: Comparison with microscopy data. Applied and Environmental Microbiology 73:4508-4514.
Gessner, M. O., and A. L. Schmitt. 1996. Use of solid-phase extraction to determine ergosterol concentrations in plant tissue colonized by fungi. Applied and Environmental Microbiology
Patel, A., R. T. Noble, J. A. Steele, M. S. Schwalbach, I. Hewson and J. A. Fuhrman. 2007.
Virus and prokaryote enumeration from planktonic aquatic environments by epifluorescence microscopy with SYBR Green I. Nature Protocols 2:269 – 276.
Pozo, J., and A. Elosegi. 2005. Coarse benthic organic matter. Pages 25-31 in M.A.S. Graca, F. Bärlocher and M.O. Gessner (editors). Methods to Study Litter Decomposition: A Practical Guide. Springer, Dordrecht, The Netherlands.