Monitoring the spatiotemporal dynamics of surface water from remote sensing imagery is essential for understanding water's impact on the global ecosystem and climate change. There is often a tradeoff between the spatial and temporal resolutions of imagery acquired from current satellite sensors and as such various spatiotemporal image fusion methods have been explored to circumvent the challenges this situation presents (e.g., STARFM). However, some challenges persist in mapping surface water at the desired fine spatial and temporal resolution. Principally, the spatiotemporal changes of water bodies are often abrupt and controlled by topographic conditions, which are usually unaddressed in current spatiotemporal image fusion methods. This paper proposes the SpatioTemporal Surface Water Mapping (STSWM) method, which aims to predict Landsat-like, 30 m, surface water maps at an 8-day time step (same as the MODIS 8-day composite product) by integrating topographic information into the analysis. In addition to MODIS imagery acquired on the date of map prediction and a pair of MODIS and Landsat images acquired temporally close to the date of prediction, STSWM also uses the surface water occurrence (SWO, which represents the frequency with which water is present in a pixel) and DEM data to provide, respectively, topographic information below and above the water surface. These data are used to translate the coarse spatial resolution water distribution representation observed by MODIS into a 30 m spatial resolution water distribution map. The STSWM was used to generate an 8-day time series surface water maps of 30 m resolution in six inundation regions globally, and was compared with several other state-of-the-art spatiotemporal methods. The stratified random sampling design was used, and unbiased estimators of the accuracies were provided. The results show that STSWM generated the most accurate surface water map in which the spatial details of surface water were well-represented.
Trees in urban environments hold significant value in providing ecosystem services, which will become increasingly important as urban populations grow. Tree phenology is highly sensitive to climatic variation, and resultant phenological shifts have significant impact on ecosystem function. Data on urban tree phenology is important to collect. Typical remote methods to monitor tree phenological transitions, such as satellite remote sensing and fixed digital camera networks, are limited by financial costs and coarse resolutions, both spatially and temporally and thus there exists a data gap in urban settings. Here, we report on a pilot study to evaluate the potential to estimate phenological metrics from imagery acquired with a conventional dashcam fitted to a car. Dashcam images were acquired daily in spring 2020, March to May, for a 2000 m stretch of road in Melksham, UK. This pilot study indicates that time series imagery of urban trees, from which meaningful phenological data can be extracted, is obtainable from a car-mounted dashcam. The method based on the YOLOv3 deep learning algorithm demonstrated suitability for automating stages of processing towards deriving a greenness metric from which the date of tree green-up was calculated. These dates of green-up are similar to those obtained by visual analyses, with a maximum of a 4-day difference; and differences in green-up between trees (species-dependent) were evident. Further work is required to fully automate such an approach for other remote sensing capture methods, and to scale-up through authoritative and citizen science agencies.
Surface water area estimation is essential for understanding global environmental dynamics, yet it presents significant challenges, particularly when dealing with small water bodies like ponds and narrow width rivers. Surface water areas for these small bodies are often inaccurately represented by existing methods due to the spatial resolution limitations in commonly used remote sensing images. This study introduces DeepWaterFraction (DWF), a deep learning approach, to estimate percent surface water area from Landsat mission imagery. DWF is trained with a self-training method, which creates training data by upscaling remote sensing images and water map labels to a lower resolution, enabling the creation of a large-scale, global coverage training dataset. DWF demonstrates superior accuracy in estimating areas for small water bodies compared to several existing methods for surface water area estimation, with a pixel-wise root mean squared error of 14.3 %. Specifically, it reduces error rates by 54.3 % for water bodies with a minimum area of 0.001 km2 and by 22.6 % for those with a minimum area of 0.01 km2. DWF’s application in global river discharge inversion is also explored, showcasing its capability to capture width variations in narrow rivers (<90 m) better than existing methods, and its robustness across environments including wetland, tree covers, and urban areas. Even for wider rivers (>150 m), DWF’s performance remains superior, as its ability to accurately quantify mixed water pixel areas effectively reflects discharge variations when the variation area is small. We find that self-training is an effective strategy for generating extensive global training datasets for water mapping, with a high upscaling factor being critical for ensuring label accuracy. This study presents a step forward in the accurate global mapping of water resources.
The understanding and management of biodiversity is often limited by a lack of data. Remote sensing has considerable potential as a source of data on biodiversity at spatial and temporal scales appropriate for biodiversity management. To-date, most remote sensing studies have focused on only one aspect of biodiversity, species richness, and have generally used conventional image analysis techniques that may not fully exploit the data's information content. Here, we report on a study that aimed to estimate biodiversity more fully from remotely sensed data with the aid of neural networks. Two neural network models, feedforward networks to estimate basic indices of biodiversity and Kohonen networks to provide information on species composition, were used. Biodiversity indices of species richness and evenness derived from the remotely sensed data were strongly correlated with those derived from field survey. For example, the predicted tree species richness was significantly correlated with that observed in the field (r =0.69, significant at the 95% level of confidence). In addition, there was a high degree of correspondence (∼83%) between the partitioning of the outputs from Kohonen networks applied to tree species and remotely sensed data sets that indicated the potential to map species composition. Combining the outputs of the two sets of neural network based analyses enabled a map of biodiversity to be produced.
No abstract is provided for this article.
No abstract is provided for this article.
Gaps often form in forest canopies due to windthrow and have important management and ecological implications. Remote sensing has considerable potential for the provision of information on gap properties but this has not been fully realized. This is largely due to the use of conventional (hard, one-pixel one-class) image analysis techniques and imagery with a relatively coarse spatial resolution. This article investigates the potential to extract information on gap properties from fine spatial resolution airborne thematic mapper imagery using soft classification techniques that allow image pixels to have multiple and partial class membership. It is shown that a standard hard maximum likelihood classification may be used to derive an accurate map of the land cover of a forested site (95.1%) from which gaps in a canopy of Sitka spruce were accurately identified (94.5%). The maximum likelihood classification was also softened by outputting probabilities of class membership for each pixel. Softening the classification increased the information on gap properties that could be extracted from the data. In particular, the accuracy with which key gap properties, such as gap area, perimeter length and shape, were estimated was higher in the outputs of the softened than hard classification. Thus, while strong correlations between the remotely sensed and ground data estimates of gap area (r ≥ 0.96) and perimeter (r ≥ 0.87), based on a sample of 36 gaps, were derived from all classifications, the accuracy with which gap properties were estimated was generally highest when a soft classification was used. For example, the use of a soft rather than hard classification resulted in the root mean square error in estimating gap area declining from 144.90 to 132.87 m2. Furthermore, the soft classification allowed the sharpness of the gap boundary to be estimated, enabling further gap properties to be inferred. In particular, the soft classification output enabled the direction of the wind event causing the initial damage to be estimated, and it may aid the definition of sites with a future risk of windthrow. FOR. SCI. 49(3):444–454.
Crowdsourcing is a popular means of acquiring data, but the use of such data is limited by concerns with its quality. This is evident within cartography and geographical sciences more generally, with the quality of volunteered geographic information (VGI) recognized as a major challenge to address if the full potential of citizen sensing in mapping applications is to be realized. Here, a means to characterize the quality of volunteers, based only on the data they contribute, was used to explore issues connected with the quantity and quality of volunteers for attribute mapping. The focus was on data in the form of annotations or class labels provided by volunteers who visually interpreted an attribute, land cover, from a series of satellite sensor images. A latent class model was found to be able to provide accurate characterisations of the quality of volunteers in terms of the accuracy of their labelling, irrespective of the number of cases that they labelled. The accuracy with which a volunteer could be characterized tended to increase with the number of volunteers contributing but was typically good at all but small numbers of volunteers. Moreover, the ability to characterize volunteers in terms of the quality of their labelling could be used constructively. For example, volunteers could be ranked in terms of quality which could then be used to select a sub-set as input to a subsequent mapping task. This was particularly important as an identified subset of volunteers could undertake a task more accurately than when part of a larger group of volunteers. The results highlight that both the quantity and quality of volunteers need consideration and that the use of VGI may be enhanced through information on the quality of the volunteers derived entirely from the data provided without any additional information.
The kappa coefficient is not an index of accuracy, indeed it is not an index of overall agreement but one of agreement beyond chance. Chance agreement is, however, irrelevant in an accuracy assessment and is anyway inappropriately modelled in the calculation of a kappa coefficient for typical remote sensing applications. The magnitude of a kappa coefficient is also difficult to interpret. Values that span the full range of widely used interpretation scales, indicating a level of agreement that equates to that estimated to arise from chance alone all the way through to almost perfect agreement, can be obtained from classifications that satisfy demanding accuracy targets (e.g. for a classification with overall accuracy of 95% the range of possible values of the kappa coefficient is −0.026 to 0.900). Comparisons of kappa coefficients are particularly challenging if the classes vary in their abundance (i.e. prevalence) as the magnitude of a kappa coefficient reflects not only agreement in labelling but also properties of the populations under study. It is shown that all of the arguments put forward for the use of the kappa coefficient in accuracy assessment are flawed and/or irrelevant as they apply equally to other, sometimes easier to calculate, measures of accuracy. Calls for the kappa coefficient to be abandoned from accuracy assessments should finally be heeded and researchers are encouraged to provide a set of simple measures and associated outputs such as estimates of per-class accuracy and the confusion matrix when assessing and comparing classification accuracy.
The potential of citizens as a source of geographical information has been recognized for many years. Such activity has grown recently due to the proliferation of inexpensive location aware devices and an ability to share data over the internet. Recently, a series of major projects, often cast as citizen observatories, have helped explore and develop this potential for a wide range of applications. Here, some of the experiences and learnings gained from part of one such project, which aimed to further the role of citizen science within Earth observation and help address environmental challenges, LandSense, are shared. The key focus is on quality assurance of citizen generated data on land use and land cover especially to support analyses of remotely sensed data and products. Particular focus is directed to quality assurance checks on photographic image quality, privacy, polygon overlap, positional accuracy and offset, contributor agreement, and categorical accuracy. The discussion aims to provide good practice advice to aid future studies and help fulfil the full potential of citizens as a source of volunteered geographical information (VGI).
The regression analyses undertaken commonly in remote sensing are aspatial, ignoring the locational information associated with each sample site at which the variables under study were measured. Typically, basic ordinary least squares regression analysis is used to derive a relationship that is believed to be uniformly applicable across the study area. Although such global analyses may appear satisfactory, often with large coefficients of determination derived, they may provide an inappropriate description of the relationship between the variables under study. In particular, a global regression analysis may miss local detail that can be significant if the relationship is spatially non-stationary. Local statistical approaches, such as geographically weighted regression, include the spatial coordinates of the sample sites in the analysis and may provide a more appropriate basis for the investigation of the relationship between variables. The potential value of geographically weighted regression to the remote sensing community is illustrated with reference to the relationship between the normalised difference vegetation index (NDVI) and rainfall over north Africa and the Middle East over an 8-year period. For each year, spatial non-stationarity was evident, particularly with regard to the slope parameter of the regression model. Moreover, the conventional ordinary least squares regression models, while superficially strong (minimum R 2=0.67), were relatively poor local descriptors of the relationship. Relative to this, the geographically weighted approach to regression provided considerably stronger relationships from the same data sets (minimum R 2=0.96) as well as highlighting areas of local variation. The implications of the difference in the outputs from the two types of regression analysis are illustrated with reference to the use of the derived NDVI–rainfall relationships in mapping desert extent. For example, with the data relating to 1987 the southern limit of the Sahara was generally estimated to lie at a more southerly position when the relationship derived from OLS rather than geographically weighted regression was used.
A classification of training data can be a useful source of information to increase thematic classification accuracy. Not only does it indicate the distribution of misclassification but it can provide data which can enable the modification of the classifiers decision rules. This can lead to significant increases in classification accuracy. Rule modification also permits the definition of uncertainty classes which may be used to delimit transitional zones between cover-types. Both rule modification and uncertainty class definition are discussed in relation to the classification of semi-natural vegetation in Surrey from Landsat-R1 data.
In order to inform the development of a remote sensing drought monitoring system over Sabah, Borneo, this paper explored how a relationship between remotely sensed data and rainfall developed at one site in Sabah transferred to four other sites in Sabah. By way of developing relationships between rainfall statistics collected at five Sabah rainforest sites and remotely sensed data (acquired by NOAA AVHRR) processed to MIR reflectance, the VI3 index, the Ts/NDVI index and Ts/VI3 index, two points were concluded. The first was that the relationship between remotely sensed data and rainfall is non-stationary across Sabah during the 1997/1998 ENSO and the second was that the index Ts/VI3 is an effective means of using the remotely sensed data available.
Accuracy assessment should be a fundamental component of an image classification analysis and is typically undertaken following either a non-site- or a site-specific methodology. The assessment of classification accuracy is, however, often difficult, with many challenges associated with the ground data typically required. Using a series of classifications of two test sites, this paper shows that accuracy assessment from both perspectives is possible through the use of a latent class modeling approach in the absence of ground data. This is possible because the parameters of a latent class model that explains the observed associations in class labeling made by a series of classifications provide estimates of class cover and conditional probabilities of class membership that equate to popular non-site- and site-specific (producer's accuracy) measures of accuracy, respectively. Additionally, the latent class model provides a new classification that could be evaluated by traditional means if ground data are available. The classification of each test site derived from the latent class model was accurate, being of equivalent accuracy to a conventional ensemble classification that was based on the same series of classifications for a site. The ability to derive a highly accurate classification and yield estimates of classification accuracy without ground data to form a testing set indicates the considerable promise of the method and a means to reduce demands for costly ground data that may also be a source of error due to imperfections.
The age of tropical secondary forests was estimated using a post-classification comparison of a multi-temporal sequence of Landsat sensor imagery for an area north of Manaus, Brazil. Five stages of regenerating secondary forest together with primary forest and agricultural land were identified. Most secondary forests at this site were younger than 14 years and therefore represented the regeneration stages at which carbon accumulation was at a maximum. This study has implications for the regional or global mapping of tropical secondary forests for global carbon studies.
Post-disaster building damage maps are an important component of the disaster response chain and may be derived from remotely sensed data. The usefulness of the maps is a function of their accuracy and hence information on map accuracy is desirable. The assessment of building damage map accuracy is, however, a challenging task as high quality ground reference data are usually scarce or absent. Here, binary and ordinal level latent class analyses were used to evaluate the accuracy of five maps produced after the 2010 earthquake in Haiti. The quality of the estimates derived from the analyses could be evaluated in this case as a ground data set was available. The results showed that the latent class analyses were able to yield an accurate assessment of the relative accuracy of the maps, allowing maps to be ranked in order of quality. This feature may help disaster relief activities by ensuring the highest quality data are used.
