Explore the vast range of titles available.

Riparian zones consist of important environmental regions, specifically to maintain the quality of water resources. Accurately mapping forest vegetation in riparian zones is an important issue, since it may provide information about numerous surface processes that occur in these areas. Recently, machine learning algorithms have gained attention as an innovative approach to extract information from remote sensing imagery, including to support the mapping task of vegetation areas. Nonetheless, studies related to machine learning application for forest vegetation mapping in the riparian zones exclusively is still limited. Therefore, this paper presents a framework for forest vegetation mapping in riparian zones based on machine learning models using orbital multispectral images. A total of 14 Sentinel-2 images registered throughout the year, covering a large riparian zone of a portion of a wide river in the Pontal do Paranapanema region, São Paulo state, Brazil, was adopted as the dataset. This area is mainly composed of the Atlantic Biome vegetation, and it is near to the last primary fragment of its biome, being an important region from the environmental planning point of view. We compared the performance of multiple machine learning algorithms like decision tree (DT), random forest (RF), support vector machine (SVM), and normal Bayes (NB). We evaluated different dates and locations with all models. Our results demonstrated that the DT learner has, overall, the highest accuracy in this task. The DT algorithm also showed high accuracy when applied on different dates and in the riparian zone of another river. We conclude that the proposed approach is appropriated to accurately map forest vegetation in riparian zones, including temporal context.

High thematic resolution vegetation mapping is essential for monitoring wetland ecosystems, supporting conservation, and guiding water management. However, producing accurate, fine-scale vegetation maps in large, heterogeneous floodplain wetlands remains challenging due to complex hydrology, spectral similarity among vegetation types, and the high cost of extensive field surveys. This study addresses these challenges by developing a scalable vegetation classification framework that integrates cluster-guided sample selection, Random Forest modelling, and multi-source remote-sensing data. The approach combines multi-temporal Sentinel-1 SAR, Sentinel-2 optical imagery, and hydro-morphological predictors derived from LiDAR and hydrologically enforced SRTM DEMs. Applied to the Great Cumbung Swamp, a structurally and hydrologically complex terminal wetland in the lower Lachlan River floodplain of Australia, the framework produced vegetation maps at three hierarchical levels: formations (9 classes), functional groups (14 classes), and plant community types (PCTs; 23 classes). The PCT-level classification achieved an overall accuracy of 93.2%, a kappa coefficient of 0.91, and a Matthews correlation coefficient (MCC) of 0.89, with broader classification levels exceeding 95% accuracy. These results demonstrate that, through targeted sample selection and integration of spectral, structural, and terrain-derived data, high-accuracy, high-resolution wetland vegetation mapping is achievable with reduced field data requirements. The hierarchical structure further enables broader vegetation categories to be efficiently derived from detailed PCT outputs, providing a practical, transferable tool for wetland monitoring, habitat assessment, and conservation planning.

Achanakmar Tiger Reserve (ATR), endowed with rich biological diversity and lush green vegetation in and around, makes it more unique. It is also an integral part of the Achanakmar Amarkantak Biosphere Reserve (AABR) and has been identified as one of the important tiger reserves of the Central Indian landscape due to its connectivity with other protected areas and tiger reserves in neighboring landscapes. Vegetation mapping and monitoring are important to understand changes in ecosystem processes and associated temporal and spatial impacts. Pre- and post-monsoon IRS, LISS III, and AWiFS satellite data from 2000, 2004, 2008, 2010, and 2013 were used for the present study. This paper is an attempt to examine the variation in the normalized difference vegetation index (NDVI) of ATR and its buffer zone on a seasonal and temporal basis. Climate conditions such as temperature, precipitation, relative humidity, etc. play an important role in the growth and development of healthy vegetation. The NDVI value of ATR has shown fluctuation and recorded positive growth over the past 14 years with few exceptions. The post-monsoon season recorded a higher NDVI value as compared to the pre-monsoon months. The maximum NDVI value was recorded in 2004 (+0.539) for the entire ATR and its buffer zone.

Discriminating marsh vegetation is critical for the rapid assessment and management of wetlands. The study area, Honghe National Nature Reserve (HNNR), a typical freshwater wetland, is located in Northeast China. This study optimized the parameters (mtry and ntrees) of an object-based random forest (RF) algorithm to improve the applicability of marsh vegetation classification. Multidimensional datasets were used as the input variables for model training, then variable selection was performed on the variables to eliminate redundancy, which improved classification efficiency and overall accuracy. Finally, the performance of a new generation of Chinese high-spatial-resolution Gaofen-1 (GF-1) and Ziyuan-3 (ZY-3) satellite images for marsh vegetation classification was evaluated using the improved object-based RF algorithm with accuracy assessment. The specific conclusions of this study are as follows: (1) Optimized object-based RF classifications consistently produced more than 70.26% overall accuracy for all scenarios of GF-1 and ZY-3 at the 95% confidence interval. The performance of ZY-3 imagery applied to marsh vegetation mapping is lower than that of GF-1 imagery due to the coarse spatial resolution. (2) Parameter optimization of the object-based RF algorithm effectively improved the stability and classification accuracy of the algorithm. After parameter adjustment, scenario 3 for GF-1 data had the highest classification accuracy of 84% (ZY-3 is 74.72%) at the 95% confidence interval. (3) The introduction of multidimensional datasets improved the overall accuracy of marsh vegetation mapping, but with many redundant variables. Using three variable selection algorithms to remove redundant variables from the multidimensional datasets effectively improved the classification efficiency and overall accuracy. The recursive feature elimination (RFE)-based variable selection algorithm had the best performance. (4) Optical spectral bands, spectral indices, mean value of green and NIR bands in textural information, DEM, TWI, compactness, max difference, and shape index are valuable variables for marsh vegetation mapping. (5) GF-1 and ZY-3 images had higher classification accuracy for forest, cropland, shrubs, and open water.

Taiga-tundra boundary ecosystems are affected by climate change. Methane (CH 4 ) emissions in taiga-tundra boundary ecosystems have sparsely been evaluated from local to regional scales. We linked in situ CH 4 fluxes (2009-2016) with vegetation cover, and scaled these findings to estimate CH 4 emissions at a local scale (10 × 10 km) using high-resolution satellite images in an ecosystem on permafrost (Indigirka lowland, north-eastern Siberia). We defined nine vegetation classes, containing 71 species, of which 16 were dominant. Distribution patterns were affected by microtopographic height, thaw depth and soil moisture. The Indigirka lowland was covered by willow-dominated dense shrubland and cotton-sedge-dominated wetlands with sparse larch forests. In situ CH 4 emissions were high in wetlands. Lakes and rivers were CH 4 sources, while forest floors were mostly neutral in terms of CH 4 emission. Estimated local CH 4 emissions (37 mg m −2 d −1 ) were higher than those reported in similar studies. Our results indicate that: (i) sedge and emergent wetland ecosystems act as hot spots for CH 4 emissions, and (ii) sparse tree coverage does not regulate local CH 4 emissions and balance. Thus, larch growth and distribution, which are expected to change with climate, do not contribute to decreasing local CH 4 emissions.

A new ecosystem-based climate envelope modeling approach was applied to assess potential climate change impacts on forest communities and tree species. Four orthogonal canonical discriminant functions were used to describe the realized climate space for British Columbia's ecosystems and to model portions of the realized niche space for tree species under current and predicted future climates. This conceptually simple model is capable of predicting species ranges at high spatial resolutions far beyond the study area, including outlying populations and southern range limits for many species. We analyzed how the realized climate space of current ecosystems changes in extent, elevation, and spatial distribution under climate change scenarios and evaluated the implications for potential tree species habitat. Tree species with their northern range limit in British Columbia gain potential habitat at a pace of at least 100 km per decade, common hardwoods appear to be generally unaffected by climate change, and some of the most important conifer species in British Columbia are expected to lose a large portion of their suitable habitat. The extent of spatial redistribution of realized climate space for ecosystems is considerable, with currently important sub-boreal and montane climate regions rapidly disappearing. Local predictions of changes to tree species frequencies were generated as a basis for systematic surveys of biological response to climate change.

To facilitate the advances in Sentinel-2A products for land cover from Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat imagery, Sentinel-2A MultiSpectral Instrument Level-1C (MSIL1C) images are investigated for large-scale vegetation mapping in an arid land environment that is located in the Ili River delta, Kazakhstan. For accurate classification purposes, multi-resolution segmentation (MRS) based extended object-guided morphological profiles (EOMPs) are proposed and then compared with conventional morphological profiles (MPs), MPs with partial reconstruction (MPPR), object-guided MPs (OMPs), OMPs with mean values (OMPsM), and object-oriented (OO)-based image classification techniques. Popular classifiers, such as C4.5, an extremely randomized decision tree (ERDT), random forest (RaF), rotation forest (RoF), classification via random forest regression (CVRFR), ExtraTrees, and radial basis function (RBF) kernel-based support vector machines (SVMs) are adopted to answer the question of whether nested dichotomies (ND) and ensembles of ND (END) are truly superior to direct and error-correcting output code (ECOC) multiclass classification frameworks. Finally, based on the results, the following conclusions are drawn: 1) the superior performance of OO-based techniques over MPs, MPPR, OMPs, and OMPsM is clear for Sentinel-2A MSIL1C image classification, while the best results are achieved by the proposed EOMPs; 2) the superior performance of ND, ND with class balancing (NDCB), ND with data balancing (NDDB), ND with random-pair selection (NDRPS), and ND with further centroid (NDFC) over direct and ECOC frameworks is not confirmed, especially in the cases of using weak classifiers for low-dimensional datasets; 3) from computationally efficient, high accuracy, redundant to data dimensionality and easy of implementations points of view, END, ENDCB, ENDDB, and ENDRPS are alternative choices to direct and ECOC frameworks; 4) surprisingly, because in the ensemble learning (EL) theorem, “weaker” classifiers (ERDT here) always have a better chance of reaching the trade-off between diversity and accuracy than “stronger” classifies (RaF, ExtraTrees, and SVM here), END with ERDT (END-ERDT) achieves the best performance with less than a 0.5% difference in the overall accuracy (OA) values, but is 100 to 10000 times faster than END with RaF and ExtraTrees, and ECOC with SVM while using different datasets with various dimensions; and, 5) Sentinel-2A MSIL1C is better choice than the land cover products from MODIS and Landsat imagery for vegetation species mapping in an arid land environment, where the vegetation species are critically important, but sparsely distributed.

Accurate and informative vegetation maps are in urgent demand to support the Kissimmee-Okeechobee-Everglades ecosystem restoration project in South Florida. In this study, we evaluated the applicability of fine spatial resolution hyperspectral data collected from the HyMap sensor for both community- and species-level vegetation mapping. Informative and accurate vegetation maps were produced by combining machine learning methods (Support Vector Machines (SVM) and Random Forest (RF)), object-based image analysis techniques, and Minimum Noise Fraction (MNF) data transformation. An overall accuracy of 90% was obtained in discriminating 14 vegetation communities. Classification of a large number of species is also promising. An overall accuracy of 85% was achieved in identifying 55 species using a SVM model. The results indicate that fine spatial resolution hyperspectral data classification using such automated procedure has great potential to replace the manual interpretation of aerial photos for vegetation mapping in heterogeneous wetland ecosystems.

CITATION: Macintyre, P. D., et al. 2018. Impact of ecological redundancy on the performance of machine learning classifiers in vegetation mapping. Ecology and Evolution, 8(13):6728-6737, doi:10.1002/ece3.4176.

This paper presents an alternative approach for high spatial resolution vegetation productivity mapping at a regional scale, using a combination of Normalised Difference Vegetation Index (NDVI) imagery and widely distributed ground-based Above-ground Net Primary Production (ANPP) estimates. Our method searches through all available single-date NDVI imagery to identify the images which give the best NDVI–ANPP relationship. The derived relationships are then used to predict ANPP values outside of field survey plots. This approach enables the use of the high spatial resolution (30 m) Landsat 8 sensor, despite its low revisit frequency that is further reduced by cloud cover. This is one of few studies to investigate the NDVI–ANPP relationship across a wide range of temperate habitats and strong relationships were observed (R2 = 0.706), which increased when only grasslands were considered (R2 = 0.833). The strongest NDVI–ANPP relationships occurred during the spring “green-up” period. A reserved subset of 20% of ground-based ANPP estimates was used for validation and results showed that our method was able to estimate ANPP with a RMSE of 15–21%. This work is important because we demonstrate a general methodological framework for mapping of ANPP from local to regional scales, with the potential to be applied to any temperate ecosystems with a pronounced green up period. Our approach allows spatial extrapolation outside of field survey plots to produce a continuous surface product, useful for capturing spatial patterns and representing small-scale heterogeneity, and well-suited for modelling applications. The data requirements for implementing this approach are also discussed.