科研之友
微信
新浪微博
Facebook
分享链接
摘要
Soil degradation threatens sustainable food production and accelerates global warming. Poorer countries, whose agricultural sectors are highly dependent on their natural resource bases, are hit particularly hard by declining soil productivity. Calls for soil-quality monitoring are therefore, justified and this could inform decision-makers on the preparation of appropriate interventions. However, the provision of monitoring methodologies is not an easy task. Soil degradation affects several soil characteristics that at larger scales cannot be evaluated with models or remote-sensing techniques. Therefore, this study focuses on investigating the use of field-based soil assessment methodologies to differentiate degrees of soil degradation. Specifically, we test the Visual Soil Field Assessment Tool (VS-Fast) for detecting and monitoring soil degradation using a cross-section of 71 sites in Senegal, the soil quality of which were classed by local experts. We found low correlation between VS-Fast classes and expert assessments. By using an ordered logit model to quantify class boundaries, we show that experts categorized areas as degraded for a wider range and higher VS-Fast scores than the corresponding VS-Fast class. Yet, from general linear models and analysis of variance procedures we found that areas classed by experts as degraded had statistically significant lower VS-Fast scores compared to those that were judged as normal and good, while differences of the VS-Fast scores between the latter two were negligible. It is remarkable that the visual assessment, the cheaper component of the VS-Fast score, performs better in differentiating degradation status than its measured counterpart. The results support the need to investigate the applicability of other VSA methodologies that only use field observations and tactile methods.
- 出版日期2012-12
