Yates et al. 2007

Yates TT, Si BC, Farrell RE, Pennock DJ. 2007. Time, location, and scale dependence of soil nitrous oxide emissions, soil water, and temperature using wavelets, cross-wavelets, and wavelet coherency analysis. Journal of Geophysical Research 112, D09104.

These authors analyzed a dataset of soil parameters and N2O emission using three subtly-different wavelet-based statistical techniques. There were two main purposes to this study; first, to examine the predictive relationships (if any) between soil parameters such as water filled pore space (WFPS) or temperature and N2O emissions; second, to evaluate the utility of these 3 wavelet techniques in analyzing this type of data.

N2O emission data is characterized by high variance in space and time, and frequent extreme values. These characteristics make many sophisticated geospatial statistical techniques not suitable, and the high spatial and temporal autocorrelation of many soil parameters eliminates many other techniques. These authors describe these limitations and some of the techniques that have been employed, and settle on 3 varieties of wavelet analysis.

Wavelet techniques are related to Fourier-transforms, and they appear to be highly complex and sophisticated methods to transform data for analysis, rather than being analytical methods per se. A large fraction of this paper is concerned with detailed description of the parameters of the transformation, and the interpretation of the results. One of the key advantages of these techniques is they usually allow examination of data across a broad range of spatial scales, thus permitting identification of the spatial scale at which important soil processes occur. Beyond that, I did not understand much of this paper.

Besides the interpretation of the differences between the 3 wavelet techniques, which was quite frankly beyond my understanding, the main result of this study was that the soil parameters that can predict N2O emissions in this landscape vary through the season. Early, around snowmelt and soil thawing, soil temperature is predictive of emissions. Later in the season, temperature loses its usefulness, and individual landscape features may present WFPS as predictive, but not in a global sense. By mid-summer, the soil parameters measured in this study no longer bore any relationship to N2O emissions. This loss of predictive value shows how complex this system is, and shows how some modeling efforts need to change in order to improve estimates of landscape-scale N2O processes.

Besides demonstrating my ignorance of advanced geospatial statistical techniques, this paper is primarily useful to me for its clear introduction describing the basic controls on and processes of N2O production in soils. My previous understanding centred on the role of water in restricting O2 availability in soils leading to changes at both the community and cell-physiology levels and consequently N2O production patterns in space and time appears to be essentially correct, and is reinforced by the early introduction section of this paper and the references therein.