Soil Carbon Sequestration as a Function of Initial Carbon Content in Different Crop Management Systems of a Long-Term Experiment.
Senthil K. Subramanian, A. N. Kravchenko, and G. P. Robertson. Michigan State Univ, PSSB, Crop and Soil Sciences Dept, East Lansing, MI 48824
The ability to sequester carbon in the soil is influenced by many factors i.e., management practices, topographical factors and inherent soil properties. Among the factors influencing magnitude and direction of long-term management effects are soil C levels prior to management implementation. The net rate of soil carbon built up by any best management practice depends on to what extent the soil is already saturated with the organic matter. We assume that the sequestration rate will be higher in the soil with low initial carbon than in the soil with high initial carbon. In the present study our objective is to study the relationship between initial carbon content and the soil carbon sequestration potential under different crop management practices of a long term experiment. Our first hypothesis is that the initial soil carbon content has a significant influence on the carbon sequestration potential and sequestration rate will be higher in the soil with low initial carbon than in the soil with high initial carbon. Our second hypothesis is that the strength of the relationship between the initial carbon and carbon sequestration is different among different tillage and management systems. For this study, we compared three management systems, namely, Chisel plowed Tillage (CT) and No-Till with conventional chemical inputs (NT), and a certified organic chisel plowed system with a winter leguminous cover crop and zero chemical inputs (CT+cover) from a Long Term Ecological Research (LTER) site located at Kellogg Biological Station, southwest Michigan, established in 1988. Soils are well-drained Typic Hapludalfs of the Kalamazoo (fine-loamy, mixed, mesic) and Oshtemo (coarse-loamy, mixed, mesic) series, developed on glacial outwash. A set of 417 total C measurements (0-20 cm depth) obtained from the experimental site in 1987 prior to the experiment establishment serves as a baseline data in this study. In 2003-2004 approximately 100 soil samples were collected from 0-5 cm depth for total soil C measurements from each of the 15 1-ha experimental plots. We extracted the initial carbon values for all the 15 1-ha experimental plots from the interpolated map of the initial C values generated using inverse distance weighing method. Statistical analyses were conducted using PROC MIXED and PROC REG procedures in SAS. The regression analyses were carried out with initial C as an independent variable and, first, using 2003-2004 C values as a dependant variable, and, second using the change in C values from 1988 to 2003-2004 as a dependant variable. Based on the regression results, we found a significant positive relationship between initial C values and 2003-2004 C values in all the three treatments with the R2 values equal to 0.53, 0.024 and 0.042 in CT, NT and CT+cover systems, respectively, and a significant negative relationship between initial C and the change in C values from 1988 to 2003-2004. These results supported our hypotheses and indicated that initial carbon levels had stronger influence on the carbon pool in CT and CT+ cover crop treatment than in NT system. We observed an increment in carbon levels in most of the plots in NT and decrement in C values in CT management. We found that management system and initial C significantly interacted in affecting 2003-2004 C levels. Carbon levels for NT were found to be significantly higher than those of CT at the whole range of the initial C levels. However, NT C levels were found to be significantly higher than those of CT+cover management only at initial C values less than 1.24 %, whereas at initial C values greater than 1.24 % the CT+cover management was found to be equally effective with NT in sequestering soil carbon. In summary, the initial C content was found to have a stronger influence on carbon processes in CT and CT+cover management systems than in NT system. Comparing treatment responses at different initial C levels the sequestration potential of NT was found to be higher over CT at all the initial carbon levels whereas CT+cover was found to be equally effective with NT in soil carbon build up rate at lower initial carbon levels but less effective at higher initial carbon levels.