 |
 |
 |
The carbon amendment treatment lowered nitrogen levels in my treated plots substantially below the levels found in control plots, a finding seen also by McLendon and Redente (1992). The difference in nitrogen between the control and experimental plots decreased for both ammonium (NH4+) and nitrate (NO3-) within a month after treatments ceased (Figure 1). In order to continuously lower soil nitrogen levels, one must continue to add more labile carbon amendments, such as sucrose, at least at monthly intervals. Less easily dissolved carbon amendments, such as sawdust, may give longer nitrogen-lowering effects with less frequent but larger applications. Morgan (1994) only applied carbon amendments once, but his total amount of carbon additions was much higher than mine -- consisting of twice as much sucrose and about fifteen times as much sawdust. He raked these carbon amendments into his study plots and noted no decrease in the effectiveness of carbon amendments over time in the plots where carbon amendments successfully stopped weed growth. The amount of carbon additions he added may have been sufficient to lower nitrogen for the entire growing season. He did not, however, measure soil nitrogen during the growing season, so we cannot be sure that nitrogen amounts stayed reduced.
The effectiveness of my carbon treatments on weed density and on ratio of invasive to non-invasive species was inconclusive. There was no significant increase in native species density in the treated plots and no significant decrease in the density of the invasive species (Table 1). While Seastedt et al. (1996) saw a significant decrease in pepper-grass (Lepidium campestre), this species was present in only seven of the study plots, so no meaningful comparison could be made (note: the pepper-grass was very likely Alyssum minus which was misidentified as Lepidium campestre in the first two years of the study). There was a trend towards a lower ratio of invasive to non-invasive weeds corresponding with the age of disturbance, although the sample size was too small to make it statistically significant (Table 2). This trend corresponds with observations that later seral sites tend to have a lower number of invasive species than early seral sites (McLendon and Redente 1992).
The reduction in knapweed and non-knapweed biomass on treated sites indicated that nitrogen limitation in the carbon-amended treatments affected plant growth (Table 3). This decrease becomes much more understandable in light of the plant census data. Most of the plants in these plots were invasive species or other non-native forbs and grasses. Invasive species tend to need higher soil nitrogen levels for growth and decrease in biomass in response to lower soil nitrogen (Huenneke et al. 1990). Many of the species were early seral species, and Redente et al. (1992) found that early seral species tend to be more dependent on nitrogen as well. In addition, the most prevalent native grass species, Western wheat, is a C3 species that has been shown to grow less in lower nitrogen conditions (Hunt et al. 1988). Native forb biomass in these plots was too small to give insight into whether the native forbs decreased in biomass in conjunction with reduced nitrogen, or whether they increased in biomass as nitrogen level and competition decreased as seen by Huenneke et al. (1990). The overall reduction in biomass found in this study compares with Willems and van Nieuwstadt s (1996) study showing that a decrease in soil nitrogen levels in a grassland correlates with a decrease in plant biomass. The total biomass decreased in my treatment plots because the vast majority of the plants tended to decrease in biomass as soil nitrogen levels decreased.
There was no significant difference in the diversity of plant species in the control and treated plots, but this is not an unexpected result for such a short-term study. Willems and van Nieuwstadt (1996) showed that a decrease in soil nitrogen levels correlates with an increase in species number and diversity. Huenneke et al. (1990) noted a decrease in species richness in conjunction with nitrogen increases via fertilization. Mountford et al. (1996) also observed this decrease in diversity in response to fertilization, and they noted that even after three years without fertilization the study plots were lower in species diversity than plots that had never been fertilized. Willems and van Nieuwstadt noted that, even after 25 years without fertilization, the diversity and species number of the plots still had not stabilized. With these longer-term studies showing such slow changes in plant diversity, it is not surprising that my short-term study had no significant effect on species number.
The practicality of carbon amendment treatment for a particular site must take into account three considerations: the length of effectiveness of carbon amendments, the amount of labor involved in carbon amendment application, and the cost and amount of carbon amendments. First, my results indicate that a short period of monthly carbon additions results in a short length of effectiveness. Longer periods of effectiveness will require monthly applications of sucrose or less frequent applications of sawdust for a much longer period of time. While I knew that sugar would disappear rapidly, I had hoped that the sawdust would generate a long-term effect. It obviously didn't.
Second, the amount of labor involved in application can affect the practicality of carbon amendment treatment. The use of carbon amendments is made more practical by the method of surface application than by Morgan's (1994) technique of raking the carbon sources into the soil -- a process which only disturbs the site more. On the other hand, Morgan only needed to add carbon amendments to the soil once, and surface application would require a regular monthly applications throughout the growing season, thus replacing the smaller labor investment with a larger time investment.
Finally, the amount and cost of carbon amendments needed affects the large-scale practicality of this treatment. The amount of carbon needed to create the desired effect varies widely with the site and initial soil nitrogen levels. I achieved a significant, but temporary, nitrogen reduction with 1000 g/m2 of sucrose and 640 g/m2 of sawdust per year. McLendon and Redente (1992) achieved a significant soil nitrogen reduction using no sawdust and only 160 g/m2 of sucrose per year. Hunt et al. (1988) noted that 150 g/m2 of sucrose per year controlled invasive species in his sucrose-added plots. Seastedt et al. (1996) noted a significant decrease in density of one of the invasive species using 300 g/m2 in 1994 and 200 g/m2 in 1995. Morgan (1994) used a substantially larger amount of carbon than any of the other researchers, 2000 g/m2 of sucrose and 16 liters of sawdust per m2, to achieve the desired control in one of his sites. Sucrose has proven its effectiveness in all studies, but it is the most expensive of the carbon sources. Waste carbon sources such as sawdust or (in Eastern Colorado) sugar beet pulp may prove far more practical for large-scale treatments. Further studies will have to be conducted before guidelines can be set up for effective use of carbon amendments in different types of sites.
Limitations with this experiment included the small size of the plots in relation to sources of seed rain and the inability to keep cows out of the plots. Seed rain is a key factor because these plots are surrounded by invasive species. While the treatment may have reduced seed sources of annual and biannual weeds within the treatment plots, seeds from adjacent areas undoubtedly reseeded the plots. Greatly lowering soil nitrogen in areas surrounded by invasive species may have the side effect of increasing the likelihood of invasion, as starving all plants in an area that is subject to high seed rain can potentially open up space for invasive species to gain or regain a foothold. The experiment was also potentially affected by cattle activities during three days in mid-June. While my soil analyses indicate that I was successful in reducing inorganic nitrogen availability during spring, the subsequent effects of cattle may have minimized the treatment effects later in the growing season. My results, therefore, may represent a conservative assessment of nitrogen limitation.
In conclusion, my study confirms that carbon amendment treatment is an effective treatment for lowering soil nitrogen levels and plant biomass. However, seed rain into my small plots likely negated any possible restoration benefits of the amendments. Better controls on external influences to plots are needed if this procedure is to become a practical tool for land restoration.
|
|
|