Description: Mopane woodland are a source of valuable resources that contribute substantially to rural economies and nutrition across Southern Africa. However, a number of factors have, of late, brought the sustainability of the mopane woodland resources into question. One of such factors is the difficulty in monitoring of defoliation process within the woodland. In this study we set out to discriminate the levels of change in forest canopy cover detectable after insect defoliation using ground based hyperspectral measurements in mopane woodland. Canopy spectral measurements were taken from three levels of defoliation: Undefoliated (UD), Partly defoliated (PD) and Refoliating plants (R) using ASD FieldSpec HandHeld 2. A pre-filtering approach (ANOVA) was compared with random forest independent variable selector in selecting the significant wavelengths for classification. Furthermore, a backward feature elimination method was used to select optimal wavelengths for discriminating the different levels of defoliation in mopane woodland. Results show that optimal wavelengths located at 707 nm, 710 nm, 711 nm, 712 nm, 713 nm, 714 nm, 727 nm, and 1066 nm were able to discriminate between the three levels of defoliation. The results further show that there was no significant difference in the overall accuracy of classification when random forest variable selector was used 82.42% (Kappa = 0.64) and the pre-filtering approach (ANOVA) 81.21% (Kappa = 0.68) used before building the classification. Overall, the study clearly demonstrated that the dynamic process of defoliation in mopane woodland can be assessed and detected using hyperspectral dataset and effective algorithm for discrimination.

Description: The population of subsaharan Africa, and particularly of the countries of the Sahel and western Africa, is one of the most vulnerable to climate change and climate-related extreme events. To provide updated information for targeted climate change adaptation measures, we modeled hotspots of climate change and related extreme events in an integrative manner. This was achieved by constructing a spatial composite indicator of cumulative climate change impact, which integrates four climate- and hazard-related subindicators: seasonal temperature trends, seasonal precipitation trends, drought occurrences, and major flood events. The analysis is based on time-series of freely available continuous, gridded geo-spatial datasets, including remote sensing data. The aggregation of the four subindicators was performed by making use of a regionalization approach, based on segmentation techniques widely used in the remote sensing community in the field of object-based image analysis. Following the approach presented in this paper, 19 hotspots with most severe climatic changes were identified, evaluated, and mapped. The method enables not only the prioritization of intervention areas, but also allows decomposing the identified hotspots into their underlying subindicators, and thus additional information for effective climate change adaptation measures can be provided.

Description: NASA's Soil Moisture Active and Passive (SMAP) mission is planned for launch in October 2014 and will provide global measurements of soil moisture and freeze/thaw state. The project is driven by both basic research and applied science goals. Understanding how application driven end-users will apply SMAP data, prior to the satellite's launch, is an important goal of NASA's applied science program and SMAP mission success. Because SMAP data are unique, there are no direct proxy datasets that can be used in research and operational studies to determine how the data will interact with existing processes. The objective of this study is to solicit data requirements, accuracy needs, and current understanding of the SMAP mission from the potential user community. This study showed that the data to be provided by the SMAP mission did substantially meet the user community needs. Although there was a broad distribution of requirements stated, the SMAP mission fit within these requirements.

Description: Context information plays a critical role in SAR image classification, as high-resolution SAR data provides more information on scene context and visual structures. This paper presents a novel classification method for SAR images based on conditional random fields (CRFs) with integration of low-level features, local label context, and pairwise label compatibility. First, we extract the low-level features used in the SVM-based unary classifier for SAR images. The supertexture is newly introduced as one of the low-level features to model the texture context between image patches. Then, we describe the context information, including local context potential and pairwise potential. Incorporation of the category context helps to resolve the ambiguities of the unary classifier. The performance of our approach in both accuracy and visual appearance for high-resolution SAR image classification is proved in the experiments.

Description: The Geoscience Laser Altimeter System (GLAS) has provided a useful dataset for estimating forest heights in many areas of the globe. Most of the studies on GLAS waveforms have focused on natural forests and only a few were conducted over forest plantations. This work set out to estimate the stand-scale dominant height and aboveground biomass of intensively managed Eucalyptus plantations in Brazil using the most commonly used models developed for natural forests. These forest plantations are valuable case studies, with large and numerous stands that are very uniform, in which field measurements are precise compared to natural forests. The height of planted Eucalyptus forest stands estimated from waveforms acquired by GLAS were compared with in situ measurements in order to determine the model that produced the best forest height estimates. For our slightly sloping study site $({rm slope}〈 {hbox{7}}^{circ})$ , the direct method defined as the difference between the signal begin and the ground peak provided forest height estimates with an accuracy of 2.2 m. The use of statistical models based on waveform metrics and digital elevation models provided slightly better results (1.89 m accuracy) in comparison with the direct method and the most relevant metrics proved to be the trailing edge extent and the waveform extent. Moreover, a power law model was used to fit in situ aboveground biomass to in situ forest height. The results using this model with GLAS-derived heights showed an accuracy for biomass of 16.1 Mg/ha.

Description: Common Ragweed (Ambrosia artemisiifolia) is a plant that constitutes an important and growing public health concern worldwide as it is probably expanding with climate change, which brings forward the need for improved mapping tools. Our final purpose is to operationalize the use of optical remote sensing for the automated mapping and surveillance of Ambrosia artemisiifolia. Analyses considering the probable spectral instability originating from the variability of the urban landscape and from that of sensors characteristics were developed. Worldview 2, Rapid Eye and SPOT 4 HRVIR sensors were used together with geolocalized surveys of Common Ragweed in Montréal and Valleyfield (Quebec, Canada). Images were standardized and various derivatives variables such as multiple vegetation indexes were created. Spectral confusion, statistical analyses, object-oriented technology and Fuzzy-logic functions were used to develop predictive risks maps of Common Ragweed potential presence. The results showed that the green bands (510–590 nm) of higher spatial resolutions sensors had a higher potential to cope with spectral confusions and changing landscape characteristics and to predict the likelihood of Ambrosia artemisiifolia presence with a recurrent stability. The good agreement between observed and predicted ragweed revealed an important potential for the operationalization of this method.

Description: In this study, several major issues associated with forest biomass mapping have been investigated using an integrated dataset, and a preliminary forest biomass map of northeastern China is presented. Three biomass regression models, stepwise regression (SR), partial least-squares regression (PLSR), and support vector regression (SVR), were developed based on field biomass data, Geoscience Laser Altimeter System (GLAS) data, and Moderate Resolution Imaging Spectroradiometer (MODIS) data. The biomass estimates using the SVR model were the most reasonable. The accuracy of the biomass predictions was improved through a combination of bootstrapping and the SVR method. The rich temporal information in MODIS data and the multiple-angle information in Multi-angle Imaging Spectro Radiometer (MISR) data were also explored for forest biomass mapping. Results indicated that a MODIS time series data alone, without MISR data, was capable of mapping forest biomass. A forest biomass map was generated using the optimal biomass regression model and the MODIS time series data. Finally, an uncertainty analysis of the biomass map was carried out and a comparison with published results using other methods was made.

Description: The invasive plant known as bugweed (Solanum mauritianum) is a notorious invader of forestry plantations in the eastern parts of South Africa. Not only is bugweed considered to be one of five most widespread invasive alien plant (IAP) species in the summer rainfall regions of South Africa but it is also one of the worst invasive alien plants in Africa. It forms dense infestations that not only impacts upon commercial forestry activities but also causes significant ecological and environment damage within natural areas. Effective weed management efforts therefore require robust approaches to accurately detect; map and monitor weed distribution in order to mitigate the impact on forestry operations. The main objective of this research was to determine the utility of support vector machines (SVMs) with a 272-waveband AISA Eagle image to detect and map the presence of co-occurring bugweed within mature Pinus patula compartments in KwaZulu Natal. The SVM when utilized with a recursive feature elimination (SVM-RFE) approach required only 17 optimal wavebands from the original image to produce a classification accuracy of 93% and True Skills Statistic of 0.83. Results from this study indicate that (1) there is definite potential for using SVMs for the accurate detection and mapping of bugweed in commercial plantations and (2) it is not necessary to use the entire 272-waveband dataset because the SVM-RFE approach identified an optimal subset of wavebands for weed detection thus enabling improved data processing and analysis.