Weicheng Wu

Different pixel-based, object-based and subpixel-based methods such as time-series analysis, decision-tree, and different supervised approaches have been proposed to conduct land use/cover classification. However, despite their proven advantages in small dataset tests, their performance is variable and less satisfactory while dealing with large datasets, particularly, for regional-scale mapping with high resolution data due to the complexity and diversity in landscapes and land cover patterns, and the unacceptably long processing time. The objective of this paper is to demonstrate the comparatively highest performance of an operational approach based on integration of multisource information ensuring high mapping accuracy in large areas with acceptable processing time. The information used includes phenologically contrasted multiseasonal and multispectral bands, vegetation index, land surface temperature, and topographic features. The performance of different conventional and machine learning classifiers namely Malahanobis Distance (MD), Maximum Likelihood (ML), Artificial Neural Networks (ANNs), Support Vector Machines (SVMs) and Random Forests (RFs) was compared using the same datasets in the same IDL (Interactive Data Language) environment. An Eastern Mediterranean area with complex landscape and steep climate gradients was selected to test and develop the operational approach. The results showed that SVMs and RFs classifiers produced most accurate mapping at local-scale (up to 96.85% in Overall Accuracy), but were very time-consuming in whole-scene classification (more than five days per scene) whereas ML fulfilled the task rapidly (about 10 min per scene) with satisfying accuracy (94.2–96.4%). Thus, the approach composed of integration of seasonally contrasted multisource data and sampling at subclass level followed by a ML classification is a suitable candidate to become an operational and effective regional land cover mapping method.

... Page 8. 226 W. Wu and E. De Pauw Fig. ... Sand control in the sandy land and desert patches, by planting grasses, shrubs and trees in a grid pattern, has also increased the green-ness of land cover. This practice has to some extent restored a number of degraded Page 9. ...

ABSTRACT Desertification, leading to the reduction of land productivity, is one of the major threats in dryland ecosystems. Actions to combat desertification have become one of the important land management practices in the arid regions or countries in the past half century. Fodder shrub plantations have been implemented in the area of Marrakech since 1996 in order to prevent rangeland from further desertification and to rehabilitate rangeland productivity. This paper demonstrates a case study on the assessment of the effectiveness of such interventions in pilot sites in Marrakech by multitemporal remote sensing technology.

ABSTRACT Implementation of land management policies influences land use and hence causes environmental change. Taking the Ordos rangelands in China as a case study, this paper explores the potential of remote sensing to assess in dryland areas the impacts of policies on the environment. Thirteen Landsat images of the period 1978–2010 were acquired and those corresponding to the starting dates of implementation of different policies were selected for land-cover change analysis; others were used to check the detected change and track the normalized difference vegetation index (NDVI) trajectory matched with time series of meteorological data for calibration of natural response of rangelands to rainfall. The results indicate that policy impacts are complex and include both positive and negative aspects depending on the locality in space. On one hand, policies have aroused the enthusiasm of people in agricultural production and sand-control leading to the recovery of about 2618 km2 of desertified rangeland and sandy land, and economic growth, on the other hand, provoked vegetation degradation with an accumulated area of 2439 km2 when policies cannot reconcile the conflict between environmental protection and the interest of rural people. However, degradation is not absolute and can be mitigated by the implementation of rational policies.

This paper presents a study on land degradation monitoring focused in the west part of the MuUs Sandy Land in Ordos, one of the important dry areas in China, aiming to understand land degradation distribution in space and time and the role of anthropogenic action in such land surface processes at local level. Multi-temporal Landsat images (MSS 1978, 1979; TM

Atriplex nummularia has been extensively planted in Northern Africa to combat desertification. However, few studies evaluated the effectiveness of these interventions. This study aimed at assessing the dynamic performance of a number of Atriplex plantations located in the Marrakech province in terms of multitemporal dry biomass production. Three SPOT 5 images (2004, 2008 and 2012) and field biomass measurements were integrated to quantify the dry biomass production dynamics of plantations established from 1996 to 2007. Different plant
ages covered the whole plant life cycle curve. Vegetation indices were derived from the images and those of 2012 were coupled to the measured biomass of 2012 to formulate biomass models. An analysis of shrub biomass production was conducted in plantations and in
adjacent rangelands, covering varying degree of plant development, and an estimate of the economic benefits generated by the plantations in terms of available fodder biomass was performed. The results show that, on average, the plantation sites produced 2·21 to 3·61Mgha
1 of dry biomass more than the surrounding rangelands. The best performing plantations yielded even greater differences, up to more than 7Mg ha
1. It was observed that the most performing plantations, while contributing to mitigating land degradation, have generated economic value and could compensate the economic cost of the intervention even under drought conditions. However, in several cases the plantation performance was far from sustainability, particularly due to poor management (early and/or over grazing), revealing that management is a critical factor for the success of this restoration practice. Copyright © 2014 John Wiley & Sons, Ltd.

Salinization is a common problem for agriculture in dryland environments and it has greatly affected land productivity and even caused cropland abandonment in Central and Southern Iraq. Hence it is of pressing importance to quantify the spatial distribution of salinity and its changing trend in space and time and ascertain the driving forces thereof. This study aims at such a diachronic salinity mapping and analysis using multitemporal remote sensing taking a pilot site, the Dujaila area in Central Iraq, as an example. For this purpose, field survey and soil sampling were conducted in the 2011–2012 period, and a multitemporal remote sensing dataset consisting of satellite imagery dated 1988–1993, 1998–2002, and 2009–2012 was prepared. An innovative processing approach, the multiyear maxima-based modeling approach, was proposed to develop remote sensing salinity models. After evaluation of their suitability, the relevant models were applied to the images for multitemporal salinity mapping, quantification, and change tracking in space and time. The driving causes of salinization in the study area were evaluated. The results reveal that the developed salinity models can reliably predict salinity with an accuracy of 82.57%, indicating that our mapping methodology is relevant and extendable to other similar environments. In addition, salinity has experienced significant changes in the past 30 years in Dujaila, especially, very strongly salinized land got continuously expanded, and all these changes are related to land use practices and management of farmers, which are closely associated with the macroscopic socioeconomic environment of the country.