In several land use models statistical methods are being used to analyse spatial data. Land use drivers that best describe land use patterns quantitatively are often selected through (logistic) regression analysis. A problem using conventional statistical methods, like (logistic) regression, in spatial land use analysis is that these methods assume the data to be statistically independent. But, spatial land use data have the tendency to be dependent, a phenomenon known as spatial autocorrelation. Values over distance are more similar or less similar than expected for randomly associated pairs of observations. In this paper correlograms of the Moran's I are used to describe spatial autocorrelation for a data set of Ecuador. Positive spatial autocorrelation was detected in both dependent and independent variables, and it is shown that the occurrence of spatial autocorrelation is highly dependent on the aggregation level. The residuals of the original regression model also show positive autocorrelation, which indicates that the standard multiple linear regression model cannot capture all spatial dependency in the land use data. To overcome this, mixed regressive-spatial autoregressive models, which incorporate both regression and spatial autocorrelation, were constructed. These models yield residuals without spatial autocorrelation and have a better goodness-of-fit. The mixed regressive-spatial autoregressive model is statistically sound in the presence of spatially dependent data, in contrast with the standard linear model which is not. By using spatial models a part of the variance is explained by neighbouring values. This is a way to incorporate spatial interactions that cannot be captured by the independent variables. These interactions are caused by unknown spatial processes such as social relations and market effects.

No abstract

[1] We studied the changes in soil carbon contents when pastures are converted to either secondary forest or plantation forest in north-western Ecuador. At 40 sites within the region, paired pasture and forest plots were compared. We related the observed soil carbon concentrations, stocks, and changes (in the 0-0.25 m and 0.25-0.5 m layers) to land use history, climate, and soil characteristics. Variation in carbon concentrations over sites in volcanic soils could be well predicted for both pastures (R 2 = 0.96) and forests (R 2 = 0.93) on the basis of soil mineralogy, while for sedimentary soils, clearly less variation could be explained (R 2 = 0.14 for pastures and 0.39 for forests). The dominant factor explaining changes in carbon stocks following pasture to forest conversion was pasture age. Forests, paired with pastures less than 10 years old, had on average 9.3 Mg ha À1 less soil carbon than the pastures, while forests paired with pastures between 20 and 30 years old had on average 18.8 Mg ha À1 more soil carbon and forest paired with pastures older than 30 years had on average 15.8 Mg ha À1 more carbon than the pastures. In this region, reforestation of old pastures will generally lead to an increase of soil carbon stocks. These results can be used for optimal site selection for carbon sequestration projects and for including soil carbon in the estimated benefits of these projects.

No abstract

The influence of soil C stabilization mechanisms is normally not considered in studies on the effects of land use changes. Instead, observed changes are typically explained by differences in litter input. As a result, it is not well known if and how quickly newly incorporated C is stabilized in soils. Our goals were to find out how much soil C was stabilized in two different soil orders (Andisols and Inceptisols) and which are the responsible mechanisms of C stabilization. Furthermore, we looked for evidence that newly incorporated soil C was stabilized in these contrasting soil orders. We selected 25 sites in northwestern Ecuador with two paired plots per site: one plot where pasture was converted to secondary forest and one plot where forest was converted to pasture. In all the plots, soil C content, stocks, and stable isotope (δ13C) signal were measured in the surface soil. The δ13C values were used to estimate the stocks of soil C derived from forest (Cdf) and from pasture (Cdp) in all plots. We calculated correlations between these stocks and soil and environmental characteristics to identify mechanisms of soil C stabilization. Our results show that long‐term stabilization in Andisols was through formation of metal–humus complexes and allophane, while in Inceptisols long‐term stabilization was through sorption to clay minerals. We found evidence that recently incorporated C was not stabilized in Andisols, while in Inceptisols, poorly crystalline (hydr‐) oxides seemed to have stabilized part of this soil C. We conclude that unless soil C stabilizing mechanisms are explicitly considered, we will not be able to predict the direction and magnitude of changes in soil C stocks following land use changes in the tropics.