Total microbial activity and biomass

Microorganisms are the primary agents of the diagenesis of organic matter in marine sediments (Deming and Baross, 1993). Thus, a strong link between carbon fluxes and microbial activity and biomass can be expected. Accordingly, significant correlations between POC fluxes measured with sediment traps and bacterial biomass (Deming and Baross, 1993), oxygen consumption, bacterial activity and total microbial biomass (Boetius and Damm, 1998) have been found in different oceanic regions. One possibility for the calculation of carbon budgets for larger oceanic regions is to exploit the empirical correlations which link the limited data of sediment trap measurements to microbial variables for which a large data base is already available. Such variables could be benthic microbial biomass and activity. However, although many data sets of benthic microbial activity and biomass have been obtained in the last few years, they were never combined on a basin wide scale.

One of the goals of the ADEPD project was to collect and harmonize as many data on benthic microbial biomass and activity as possible - from the project partners and their collaborators as well as by including data from the literature. A number of different variables were compiled as parameters for microbial activity and biomass: bacterial biomass, total adenylates, DNA, phospholipids and the activity of different enzymes (hydrolytic and electron-transporting). For each of these parameters, about 100-200 datapoints entered the databank. These data are now available on the ADEPD home page. The data had to be converted into comparable units (if possible biomass carbon and molar carbon turnover). We also investigated if the data could be linked to other biological (pigment concentrations) or geochemical data (TOC, oxygen consumption, accumulation rates).

The data on microbial activity were highly diverse in terms of the different methods used. Each investigator focussed on different enzyme activities according to the specific scientific questions in each of the different studies. The data of all 13 activity parameters were collected and organized in the data bank with method descriptions and links to the investigators. The potential microbial hydrolysis of organic matter in the sediments can be studied using various model substrates for the different enzymes. This parameter is now used in pelagic as well as benthic environments as a parameter for the potential carbon turnover by the microbes. Good relationships between organic matter availability and the potential activity of the enzymes b-glucosidase and chitobiase were established in a variety of investigations. A compilation on the relationship between some enzymes and e.g. chloroplastic pigment concentrations (CPE) in the sediments showed that linear relationships can be found including data from very different oceanic regions (Fig 5, Lochte et al., 1999).

Figure 5: Correlation of b-glucosidase activity and chloroplastic pigments equivalents (CPE) in surface sediments (data from the Arabian Sea: Boetius and Pfannkuche unpubl. data)

The largest amount of data was available for ATP (200 data points), a variable which can be used for the estimation total microbial biomass. Other parameters of microbial biomass measured in several of the investigations were phospholipid concentrations and bacterial biomass determined by microscopy. One goal of our project was to establish a common conversion factor for each method to obtain comparable estimates of microbial carbon biomass.

Table 1 shows the conversion factors obtained for each method. By applying these empirical relationships based on linear regression analysis of all data for each parameter the different variables for microbial biomass were converted into carbon based total microbial biomass (TMB). Bacterial biomass (det. by microscopy) was also converted to TMB on the basis of linear regression with phospholipid concentrations, to account for other microbial organisms like fungi, yeasts and protozoa which contribute significantly to the total microbial biomass in sediments. DNA data were not converted, because either adenylates or phospholipids concentrations were available from the same samples. A total of 300 data on carbon biomass were obtained by this procedure. Figure 6 shows the distribution of microbial carbon biomass in the Atlantic.

Table 1: Conversion of different parameters of microbial biomass into total microbial biomass (= TMB) in carbon units (µg C cm-3 sediment). TMB was calculated from phospholipid concentrations, based on the finding that 100nmol phospholipids is equivalent to 1 g C (Dobbs and Findlay 1993). The regression analyses are based on pairs of data from the same sample.

Parameter	Number of Data	Conversion into:	Regression	Regression Coefficient
ATP (pmol cm-3)	103	total adenylates (pmol cm-3)	y = 3.3x - 12	R2 = 0.996 p<0.001
total adenylates (pmol cm-3)	226	TMB (µg C cm-3)	y = 0.3x + 35	R2 = 0.465 p<0.001
bacterial biomass (µg C cm-3)	97	TMB (µg C cm-3)	y = 1.2x + 28	R2 = 0.494 p<0.001

A relatively large data set is available for the eastern Arctic basins as well as for the East Atlantic. No data were obtained for the Midatlantic Ridge, the western parts of the Atlantic and Arctic as well as for the Southern Ocean. The data bank could be further improved by including U.S. benthic microbiologists as cooperation partners of future projects.

Bacteria make up the largest fraction of microbial biomass in deep-sea sediments and, hence, their biomass is presumably a good indicator for the trophic supply, i.e. the POC sedimentation to the seafloor (Deming and Baross 1993). It is believed that this relationship between POC input and microbial biomass is caused by the limitation of microbial growth due to the low supply of degradable organic matter to the deep sea. This is also the explanation for the relationship between POC flux to the sediments and oxygen consumption, i.e. carbon turnover. Thus, a correlation between microbial biomass and oxygen demand in the sediments is likely. However, this relationship was rarely tested in abyssal habitats. Our aim was to accumulate a large dataset of total microbial biomass to investigate emperical relationships which could potentially be used as proxies for oxygen consumption.

A total of 300 biomass data are available, however, very few data were from investigations with parallel biogeochemical measurements. Less than 10% of the data were linked to oxygen flux data, and these fell only within 5 grids of 1°x1° degree latitude and longitude. Thus, we were not able to obtain a sufficient data set to test the correlation of microbial biomass and oxygen flux on a basin wide scale. The reason for this missing link is that there are only very few benthic investigations in which geological, biogeochemical research as well as microbiology were carried out at the same stations. Such interdisciplinary studies of deep-sea areas are e.g. SEEP, Eumeli, OMEX, EQPAC, BIO-C-FLUX, BIGSET. However, these investigations mainly focussed on process studies and were carried out repeatedly at a few geographical locations only. Even including the available literature, the current data base for the Atlantic Ocean is not good enough to test the relationship between oxygen consumption and benthic microbial biomass and activity. This can only be improved by further field research, covering large oceanic regions with combined studies of benthic biology and biogeochemistry.