The Conversion of Farmland to Forest (so called Grain for Green) means stopping cultivation in sloping cultivated land which has caused soil erosion and/or land desertification, and restoring vegetation in artificial or natural way. From 1999 to 2008, 2686.67×10⁴ hm² land area has been involved in the Conversion of Farmland to Forest Project (CFP) in China, with about 191.8 billion RMB Yuan (roughly equal to 28 billion US＄) as investment and compensation fund from the Chinese Central Government. As one of the seven largest forest restoration projects in the world, there is a high need to assess the CFP's benefits ecologically, to show its real and potential impacts on improving the regional environment and curbing the local land desertification.
As one of the most serious regions suffering from worsening soil erosion and water loss for years in China or even in the whole world, the Northern Shaanxi, where the Loess Plateau hinterland located, was chosen as one of the pilot zones to carry out the Conversion of Farmland to Forest Project in 1999. More than 10 years after the Project's implementation, the vegetation has changed significantly in Northern Shaanxi, reported and observed by different data sources. Was this change caused by anthropogenic factors such as CFP project? Or how large a proportion of CFP's effect was involved in the influence on the ecologically vegetation change? Monitoring and assessing the vegetation variation could be one of the direct measurements to indicate and depict the ecological benefits brought by the CFP project. However, it is a common sense that there is no easy way to distinguish the vegetation variation caused by the CFP project from among many anthropogenic and natural factors that could influence the vegetation changes. We chose different land cover types and slope gradients in Northern Shaanxi as two main aspects, as the Conversion of Farmland to Forest Project was carried out mainly in sloping farmlands where the slopes were steeper than 15, and compared those areas carrying CFP out with other farmlands without CFP implementation, regardless of the climate factors. Using MODIS/NDVI data which was assumed to be suitable for monitoring large scale vegetation variation, we recognized the vegetation changes resulted from the CFP project through methods of the averaged NDVI variation and the trend analysis. The preliminary results were as follows: (1) a noticeable increase of average NDVI was found in the whole Northern Shaanxi from 2000 to 2008, and particularly the NDVI of sloping farmland and sloping grassland increased most rapidly. (2) Based on the trends analysis, it can be seen that the vegetation was recovering well in most parts of Northern Shaanxi, and the proportion of the study area with significant increase in NDVI was 64.96%, while the proportion with medium increase was 18.58%. (3) The proportions with significant increase in NDVI in sloping farmland and sloping grassland were 45.43% and 17.10% respectively. The proportion of significant increase in sloping farmland took up a major part of the area with significant increase in NDVI. The proportions of areas with significantly increased NDVI for different slope gradients (7 15°, 15 25' and 25 35°) were 39.91%, 25.81% and 2.28% respectively. Among these, the area with 7 25° slope gradient occupied 65.72% of the total area. (4) The mean temperature, drought and annual precipitation did not increase significantly in Northern Shaanxi, which meant that anthropogenic factors such as the CFP project were among the main causes leading to the trend of NDVI increase. The implementation of the CFP project was therefore advantageous in restoring the degraded vegetation in Northern Shaanxi and the CFP project had a favorable benefit in vegetation restoration.