Unexpected consequences of afforestation in degraded drylands: Divergent impacts on soil and vegetation

Forestry has long been considered an effective means of restoring degraded drylands worldwide. Often, afforestation in such lands relies on the establishment of runoff harvesting systems that are formed as contour bench terraces on hillslopes, increasing water availability for the planted trees and shrubs. The construction of terraces requires intensive earthworks by heavy machinery. This study assessed the long-term (>10 yrs) effects of forestry-related land-use change on soil properties and herbaceous vegetation in 16-year-old and 12-year-old afforestation sites (established in 2005 and 2009), and in nearby control ("natural") areas in the semi-arid northern Negev, Israel. Mean herbaceous vegetation height in the 2005 afforestation sites (12.1 cm) was significantly (P = 0.0009) and 23% greater than in the control areas (9.8 cm), whereas in the 2009 afforestation sites (6.2 cm) it was 37% lesser than in the control areas. Mean herbaceous vegetation aboveground biomass was similar in the 2005 afforestation (0.39 Mg ha^-1) and control areas (0.38 Mg ha^-1), and almost significantly (P = 0.0510) and twofold greater than in the 2009 afforestation sites (0.19 Mg ha^-1). The effect of hillslope aspect on these variables was substantial; their mean values were higher in the northern (mesic) hillslopes than in the southern (xeric) hillslopes. Soil samples were obtained from depths of 0-5 and 5-10 cm and physio-chemo-biological properties were assessed in the laboratory. The overall soil quality - as calculated by two soil quality indices (SQIs), including the generalized SQI (SQI_gen) and the minimum dataset SQI (SQI_MDS) - was significantly (P < 0.0001 for both indices) and 13-22% greater in the control areas (0.52 and 0.61, respectively) than that in the afforestation treatments (0.44-0.46 and 0.50-0.51, respectively). These results are generally attributed to the removal of soil's A-horizon during earthworks, and the exposure of the underlying B-horizon. The similar SQI values of both hillslope aspects, as well as of both soil depths, indicate the generally degraded state of the entire region. In conclusion, while contour bench terracing may facilitate the recovery of herbacaeous vegetation to some extent, the effectiveness of this practice for soil restoration is questionable. Overall, insights of this study demonstrate a caveat that converting natural drylands to forestry systems may not yield sufficient ecological benefits, and therefore should be implemented with caution.