Zoltan Szantoi | European Space Agency (original) (raw)

Land Cover Monitoring by Zoltan Szantoi

The Copernicus High-Resolution Hot Spot Monitoring activity (C-HSM) delivers a global dataset of ... more The Copernicus High-Resolution Hot Spot Monitoring activity (C-HSM) delivers a global dataset of Key Landscapes for Conservation (KLC), which are characterized by pronounced anthropogenic pressures that require high mapping accuracy. Detailed land cover and land cover change map products are freely available through the activity and include extensive map production accuracy assessments. Without a complete understanding of the map products' spatial, temporal and logical consistencies, quality or quantified confidence levels, usability is reduced and can affect stakeholder decision-making and the implementation of sustainable solutions. For the quantitative accuracy assessment, a stratified random sampling approach was implemented where special emphasis was placed on (i) allocation of sampling units for rare land cover change categories; (ii) effective and accurate labelling of large numbers of sampling units; (iii) accuracy and area estimation in one consistent approach; and (iv) derivation of confidence intervals for all accuracy measures and area estimates. To handle correlations, large uncertainties, and complex probability density functions, bootstrapping was applied instead of analytical equations, which are based on normality assumptions. This paper presents the Quality Assurance and Quality Control framework applied to validate all the C-HSM thematic map products. The Kundelungu-Upemba KLC product results are presented as our use case.

Côte d'Ivoire and Ghana are the largest producers of cocoa in the world. In recent decades the cu... more Côte d'Ivoire and Ghana are the largest producers of cocoa in the world. In recent decades the cultivation of this crop has led to the loss of vast tracts of forest areas in both countries. Efficient and accurate methods for remotely identifying cocoa plantations are essential to the implementation of sustainable cocoa practices and for the periodic and effective monitoring of forests. In this study, a method for cocoa plantation identification was developed based on a multi-temporal stack of Sentinel-1 and Sentinel-2 images and a multi-feature Random Forest (RF) algorithm. The Normalized Difference Vegetation Index (NDVI) and second-order texture features were assessed for their importance in an RF classification, and their optimal combination was used as input variables for the RF model to identify cocoa plantations in both countries. The RF model-based cocoa map achieved 82.89% producer's and 62.22% user's accuracy, detecting 3.69 million hectares (Mha) and 2.15 Mha of cocoa plantations for Côte d'Ivoire and Ghana, respectively. The results demonstrate that a combination of an RF model and multi-feature classification can distinguish cocoa plantations from other land cover/use, effectively reducing feature dimensions and improving classification efficiency. The results also highlight that cocoa farms largely encroach into protected areas (PAs), as 20% of the detected cocoa plantation area is located in PAs and almost 70% of the PAs in the study area house cocoa plantations.

Many land cover changes are a direct threat to nature, especially through its negative effects on... more Many land cover changes are a direct threat to nature, especially through its negative effects on ecosystem services and thus, poses a risk to long-term human health and wellbeing. Land cover data is requested by different users, i.e. it is key for many SDG indicators and therefore accurate and continuously updated land cover information is more essential than ever. However, the "let's produce a map and

Natural resources are increasingly threatened in the world. Threats to biodiversity and human wel... more Natural resources are increasingly threatened in the world. Threats to biodiversity and human wellbeing pose enormous challenges in many vulnerable areas. Effective monitoring and protection of sites with strategic conservation importance require timely monitoring, with a particular focus on certain land cover classes that are especially vulnerable. Larger ecological zones and wildlife corridors also warrant monitoring, as these areas are subject to an even higher degree of pressure and habitat loss as they are not "protected" compared to protected areas (national parks, nature reserves, etc.). To address such a need, a satellite-imagery-based monitoring workflow was developed to cover at-risk areas. The first phase of the programme covered a total area of 560 442 km 2 in sub-Saharan Africa. In this update, we remapped some of the areas using the latest satellite images available, and in addition we included some new areas to be mapped. Thus, in this version we have updated and mapped an additional 852 025 km 2 in the Caribbean, African and Pacific regions, involving up to 32 land cover classes. Medium-to high-spatial-resolution satellite imagery was used to generate dense time series data, from which the thematic land cover maps were derived. Each map and change map was fully verified and validated by an independent team to meet our strict data quality requirements. The independent validation datasets for each key landscape for conservation (KLC) are also described and presented here (all datasets presented are

In recent years, the popularity of tree-based ensemble methods for land cover classification has... more In recent years, the popularity of tree-based ensemble methods for land cover classification has increased significantly. Using WorldView-2 image data, we evaluate the potential of the oblique random forest algorithm (oRF) to classify a highly heterogeneous protected area. In contrast to the random forest (RF) algorithm, the oRF algorithm builds multivariate trees by learning the optimal split using a supervised model. The oRF binary algorithm is adapted to a multiclass land cover and land use application using both the“one-against-one”and “one-against-all” combination approaches. Results show that the oRF algorithms are capable of achieving high classification accuracies (>80%). However, there was no statistical difference in classification accuracies obtained by the oRF algorithms and the more popular RF algorithm. For all the algorithms, user accuracies (UAs) and producer accuracies (PAs)>80% were recorded for most of the classes. Both the RF and oRF algorithms poorly classified the indigenous forest class as indicated by the low UAs and PAs. Finally, the results from this study advocate and support the utility of the oRF algorithm for land cover and land use mapping of protected areas using WorldView-2 image data.

Monitoring of crop phenology significantly assists agricultural managing practices and plays an i... more Monitoring of crop phenology significantly assists agricultural managing practices and plays an important role in crop yield predictions. Multi-temporal satellite-based observations allow analyzing vegetation seasonal dynamics over large areas by using vegetation indices or deriving
biophysical variables. The Northern Nile Delta represents about half of all agricultural lands of Egypt. In this region, intensifying farming systems are predominant, which translates into a
pressure on water supply demand. Moreover, double cropping rotations schemes are increasing,
requiring a high temporal and spatial resolution monitoring for capturing successive crop growth
cycles. This study presents a framework for crop phenological characterization based on high
spatial and temporal resolution time series of green Leaf Area Index (LAI). Particularly, NASA's
Harmonized Landsat 8 and Sentinel-2 (HLS) surface reflectance dataset was used. The HLS dataset
provides seamless products from both satellites, enabling global land observations every 2-3 days
at 30m. A green LAI retrieval model was originally trained using ground-based LAI measurements
with Gaussian processes technique and validated for Sentinel-2 (R2: 0.7, RMSE= 0.67m2/m2) (Amin
et al., 2020). Given the compatible spectral bands configuration of both sensors, a new model for
Landsat 8 was adapted from the original one. Both models were implemented in an HLS image
based automated retrieval chain obtaining therefore two different LAI time series, which were
spatially averaged per crop parcel according to the ground data at disposal. The subsequent
analysis was performed based on the time series phenological pre-processing and modelling
implemented in the in-house developed scientific time series toolbox DATimeS (Belda et al., 2020).
The proposed framework permitted to determine the crop patterns for four consecutive years
(2016-2019), identifying one or two seasons per year, for single (e.g. grape, citrus) or doublecropping
(e.g. maize-onion, maize-wheat, rice-clover), respectively. Alongside, each detected crop
was characterized by retrieving a selected set of phenological parameters, which were contrasted
with respect to the established crop type calendar (planting and harvesting dates) and for each
crop type, the annual mean value was computed and the intra annual variability within the four
years was assessed.

— We propose an automated processing workflow to detect and classify changes in bitemporal very h... more — We propose an automated processing workflow to detect and classify changes in bitemporal very high-resolution (VHR) SmallSat imagery. The workflow consists of two preprocessing steps: an image registration method with a cross correlation approach and, second, a radiometric normalization based on regression of automatically detected invariant pixels using the iteratively reweighted multivariate alteration detection (IR-MAD) method. The IR-MAD method also transforms pairs of registered images and detects the pixels of change. Finally, we distinguish different types of change using a clustering algorithm on the original spectral values as well as on the values resulting from different image transformations. We applied this workflow to SkySat images of Rennell Island (Solomon Islands) to detect selective logging and subsequent regrowth within a conservation area. We show how the use of VHR SmallSat images enables the detection of selective logging in the tropics, where significant cloud cover and the small extent of disturbances can prevent detection using data from traditional earth-orbiting platforms. The fine temporal resolution data from SmallSat constellations can generate enough valid observations to rapidly detect disturbances (even when cloud cover is frequent) and partial recovery. In addition, SkySat imagery provides sufficiently high spatial resolution to detect crown-level changes such as those from selective logging. Our change detection workflow is fast and highly accurate, and will be increasingly useful as the data flow from earth-observation sources increases.

A fully automatic phenology-based synthesis (PBS) classification algorithm was developed to map l... more A fully automatic phenology-based synthesis (PBS) classification algorithm was developed to map land cover based on medium spatial resolution satellite data using the Google Earth Engine cloud computing platform. Vegetation seasonality, particularly in the tropical dry regions, can lead conventional algorithms based on a single date image classification to "misclassify" land cover types, as the selected date might reflect only a particular stage of the natural phenological cycle. The PBS classifier operates with occurrence rules applied to a selection of single date image classifications of the study area to assign the most appropriate land cover class. Since the launch of Landsat 8 in 2013, it has been possible to acquire imagery at any point on the Earth every 16 days with exceptional radiometric quality. The relatively high global acquisition frequency and the open data policy allow near-realtime land cover mapping and monitoring with automated tools such as the PBS classifier. We mapped four protected areas and their 20-km buffer zones from different ecoregions in Sub-Saharan Africa using the PBS classifier to present its first results. Accuracy assessment was carried out through a visual interpretation of very high resolution images using a Web geographic information system interface. The combined overall accuracy was over 90%, which demonstrates the potential of the classifier and the power of cloud computing in geospatial sciences.

Land cover change is occuring globally and is monitored by governmental and non-governmental agen... more Land cover change is occuring globally and is monitored by governmental and non-governmental agencies with remote sensing expertise. Deriving actual land cover statistics and land cover change information is highly relevant for decision makers and conservation practitioners. Analyses based on remote sensing are often driven by data availability, technical advances and capabilities and perspectives. However ecological requirements are frequently bypassed, making it difficult to evaluate the ecological effects of land cover change.on habitats. In this interdisciplinary study we combine remote sensing technologies with ecological analysis of habitat fragmentation to evaluate the ecological effects of land cover change in African protected areas. Using information on habitat requirements for mammal species and land cover classification based on medium spatial resolution satellite imagery, we derive ecological relevant information on land cover change over time with respect to its effect...

The western "panhandle" region of Florida experienced greater development in the years 2000-2010 ... more The western "panhandle" region of Florida experienced greater development in the years 2000-2010 than the previous 40 years and greater than nearly all other parts of the United States. One reason is the fact that much of the coastal land in the peninsula of Florida is already developed whereas much of the Panhandle is empty. Another reason is the fact that increasing temperatures and less severe winters in the Panhandle have made the region more attractive too people for habitation. Climate change is changing land use and land cover in the Panhandle in a number of ways with the most common shift being from managed forest and agricultural land to urban use. The consequences of this change have been increased pressure on the resources and exacerbated land use conflict. To address the challenges of future land use, a GIS optimization model was developed to determine the spatial and temporal status quo relationships between the drivers and resulting patterns of land uses due to climate change. Resulting models of land use preference and projected population allocation points to the direction of the limiting factor of climate change: saltwater intrusion and increasing water demand.

Remote Sensing and Biodiversity by Zoltan Szantoi

In order to minimize the environmental impacts of the Secretariat's processes, and to contrib... more In order to minimize the environmental impacts of the Secretariat's processes, and to contribute to the Secretary-General's initiative for a C-Neutral UN, this document is printed in limited numbers. Delegates are kindly requested to bring their copies to meetings and not to request additional copies. Note by the Executive Secretary 1. The Executive Secretary is circulating herewith, for the information of participants in the seventeenth meeting of the Subsidiary Body on Scientific, Technical and Technological Advice, the "Review of the use of remotely-sensed data for monitoring biodiversity change and tracking progress towards the Aichi Biodiversity Targets". 2. The report has been prepared by the United Nations Environment Programme – World Conservation Monitoring Centre in association with Biodiversity Indicators Partnership and the Group on Earth Observations Biodiversity Observations Network (GEO BON) with the financial support of the European Commission and t...

Medium resolution imagery (typically 10 to 30 m) has a growing role in global vegetation mapping.... more Medium resolution imagery (typically 10 to 30 m) has a growing role in global vegetation mapping. Recent efforts on global deforestation estimation based on Landsat (LS) 7 ETM+ imagery or global land cover monitoring based on LS5 TM and LS7 ETM+ imagery generated interest not only from the scientific community but from the general public as well. However, in some areas, especially where the available scenes are limited, researchers rely on few scenes, even a single image, to generate map products, even though it could have quality issues or seasonality problems. The increased availability of free medium spatial resolution imagery including the upcoming Sentinel-2 A and B and CBERS-4 satellites and the already operating OLI sensor on board LS8, will ease the collection and selection process for good quality, cloud free imagery for a certain date for any geographical region of interest, such as national parks and their buffer zones. However, this calls for dedicated approaches to the ...

Delivering the Sustainable Development Goals (SDGs) requires balancing demands on land between ag... more Delivering the Sustainable Development Goals (SDGs) requires balancing demands on land between agriculture (SDG 2) and
biodiversity (SDG 15). The production of vegetable oils and, in particular, palm oil, illustrates these competing demands and
trade-offs. Palm oil accounts for ~40% of the current global annual demand for vegetable oil as food, animal feed and fuel (210
Mt), but planted oil palm covers less than 5–5.5% of the total global oil crop area (approximately 425 Mha) due to oil palm’s
relatively high yields. Recent oil palm expansion in forested regions of Borneo, Sumatra and the Malay Peninsula, where >90%
of global palm oil is produced, has led to substantial concern around oil palm’s role in deforestation. Oil palm expansion’s direct
contribution to regional tropical deforestation varies widely, ranging from an estimated 3% in West Africa to 50% in Malaysian
Borneo. Oil palm is also implicated in peatland draining and burning in Southeast Asia. Documented negative environmental
impacts from such expansion include biodiversity declines, greenhouse gas emissions and air pollution. However, oil palm
generally produces more oil per area than other oil crops, is often economically viable in sites unsuitable for most other crops
and generates considerable wealth for at least some actors. Global demand for vegetable oils is projected to increase by 46%
by 2050. Meeting this demand through additional expansion of oil palm versus other vegetable oil crops will lead to substantial
differential effects on biodiversity, food security, climate change, land degradation and livelihoods. Our Review highlights
that although substantial gaps remain in our understanding of the relationship between the environmental, socio-cultural and
economic impacts of oil palm, and the scope, stringency and effectiveness of initiatives to address these, there has been little
research into the impacts and trade-offs of other vegetable oil crops. Greater research attention needs to be given to investigating
the impacts of palm oil production compared to alternatives for the trade-offs to be assessed at a global scale.

Oil seed crops, especially oil palm, are among the most rapidly expanding agricultural land uses,... more Oil seed crops, especially oil palm, are among the most rapidly expanding agricultural land uses, and their expansion is known to cause significant environmental damage. Accordingly, these crops often feature in public and policy debates which are hampered or biased by a lack of accurate information on environmental impacts. In particular, the lack of accurate global crop maps remains a concern. Recent advances in deep-learning and remotely sensed data access make it possible to address this gap. We present a map of closed-canopy oil palm (Elaeis guineensis) plantations by typology (industrial versus smallholder plantations) at the global scale and with unprecedented detail (10 m resolution) for the year 2019. The DeepLabv3+ model, a convolutional neural network (CNN) for semantic segmentation, was trained to classify Sentinel-1 and Sentinel-2 images onto an oil palm land cover map. The characteristic backscatter response of closed-canopy oil palm stands in Sentinel-1 and the ability of CNN to learn spatial patterns, such as the harvest road networks, allowed the distinction between industrial and smallholder plantations globally (overall accuracy = 98.52 ± 0.20 %), outperforming the accuracy of existing regional oil palm datasets that used conventional machine-learning algorithms. The user's accuracy, reflecting commission error, in industrial and smallholders was 88.22 ± 2.73 % and 76.56 ± 4.53 %, and the producer's accuracy, reflecting omission error, was 75.78 ± 3.55 % and 86.92 ± 5.12 %, respectively. The global oil palm layer reveals that closed-canopy oil palm plantations are found in 49 countries, covering a mapped area of 19.60 Mha; the area estimate was 21.00 ± 0.42 Mha (72.7 % industrial and 27.3 % smallholder plantations). Southeast Asia ranks as the main producing region with an oil palm area estimate of 18.69 ± 0.33 Mha or 89 % of global closed-canopy plantations. Our analysis confirms significant regional variation in the ratio of industrial versus smallholder growers, but it also confirms that, from a typical land development perspective, large areas of legally defined smallholder oil palm resemble industrial-scale plantings. Since our study identified only closed-canopy oil palm stands, our area estimate was lower than the harvested area reported by the Food and Agriculture Organization (FAO), particularly in West Africa, due to the omission of young and sparse oil palm

Despite growing awareness about its detrimental effects on tropical biodiversity, land conversion... more Despite growing awareness about its detrimental effects on tropical
biodiversity, land conversion to oil palm continues to increase
rapidly as a consequence of global demand, profitability, and the
income opportunity it offers to producing countries. Although
most industrial oil palm plantations are located in Southeast Asia,
it is argued that much of their future expansion will occur in
Africa.We assessed how this could affect the continent’s primates
by combining information on oil palm suitability and current
land use with primate distribution, diversity, and vulnerability.
We also quantified the potential impact of large-scale oil palm
cultivation on primates in terms of range loss under different
expansion scenarios taking into account future demand, oil palm
suitability, human accessibility, carbon stock, and primate vulnerability.
We found a high overlap between areas of high oil palm
suitability and areas of high conservation priority for primates.
Overall, we found only a few small areas where oil palm could
be cultivated in Africa with a low impact on primates (3.3 Mha,
including all areas suitable for oil palm). These results warn that,
consistent with the dramatic effects of palm oil cultivation on biodiversity
in Southeast Asia, reconciling a large-scale development
of oil palm in Africa with primate conservation will be a great
challenge.

Monitoring is essential for conservation of sites, but capacity to undertake it in the field is o... more Monitoring is essential for conservation of sites, but capacity to undertake it in the field is often limited. Data collected by remote sensing has been identified as a partial solution to this problem, and is becoming a feasible option, since increasing quantities of satellite data in particular are becoming available to conservationists. When suitably classified, satellite imagery can be used to delin-eate land cover types such as forest, and to identify any changes over time. However, the conservation community lacks (a) a simple tool appropriate to the needs for monitoring change in all types of land cover (e.g. not just forest), and (b) an easily accessible information system which allows for simple land cover change analysis and data sharing to reduce duplication of effort. To meet these needs, we developed a web-based information system which allows users to assess land cover dynamics in and around protected areas (or other sites of conservation importance) from multi-temporal medium resolution satellite imagery. The system is based around an open access toolbox that pre-processes and classifies Landsat-type imagery, and then allows users to interactively verify the classification. These data are then open for others to utilize through the online information system. We first explain imagery processing and data accessibility features, and then demonstrate the toolbox and the value of user verification using a case study on Nakuru National Park, Kenya. Monitoring and detection of disturbances can support implementation of effective protection, assist the work of park managers and conservation scientists, and thus contribute to conservation planning, priority assessment and potentially to meeting monitoring needs for Aichi target 11.

Satellite remote sensing is an important tool for monitoring the status of biodiversity and assoc... more Satellite remote sensing is an important tool for monitoring the status of biodiversity and associated environmental parameters, including certain elements of habitats. However, satellite data are currently underused within the biodiversity research and conservation communities. Three factors have significant impact on the utility of remote sensing data for tracking and understanding biodiversity change. They are its continuity, affordability, and access. Data continuity relates to the maintenance of long-term satellite data products. Such products promote knowledge of how biodiversity has changed over time and why. Data affordability arises from the cost of the imagery. New data policies promoting free and open access to government satellite imagery are expanding the use of certain imagery but the number of free and open data sets remains too limited. Data access addresses the ability of conservation biologists and biodiversity researchers to discover, retrieve, manipulate, and extract value from satellite imagery as well as link it with other types of information. Tools are rapidly improving access. Still, more cross-community interactions are necessary to strengthen ties between the biodiversity and remote sensing communities.

High spatial resolution data is increasingly available, however its cost deferring its general us... more High spatial resolution data is increasingly available, however its cost deferring its general use, particularly among conservation and biodiversity scientists. Thus, most researchers working on local, regional or global scale studies rely on lower resolution data, in most cases using the freely available Landsat database. Sumatra and Borneo are the last remaining islands where the orang-utan (Pongo spp.) species are distributed. To monitor the orang-utan's habitats, rapid and accurate remotely sensed information is needed. However, the available forest or land cover maps do not necessary satisfy such requirements, or when imagery was digitized, severely degraded areas were sporadically categorized as primary forests. Unmanned aerial vehicles (UAVs) are increasingly employed in biodiversity monitoring, assessing wildlife (i.e. counting eggs) or vegetation surveys. Their common benefits, compared to manned airplane operations, are: low risk and cost, multispectral imagery at vari...

Conservation of the Sumatran orangutans' (Pongo abelii) habitat is threatened by change in land u... more Conservation of the Sumatran orangutans' (Pongo abelii) habitat is threatened by change in land use/land cover (LULCC), due to the logging of its native primary forest habitat, and the primary forest conversion to oil palm, rubber tree, and coffee plantations. Frequent LULCC monitoring is vital to rapid conservation interventions. Due to the costs of high-resolution satellite imagery, researchers are forced to rely on cost-free sources (e.g. Landsat), those, however, provide images at a moderate-to-low resolution (e.g. 15-250 m), permitting identification only general LULC classes, and limit the detection of small-scale deforestation or degradation. Here, we combine Landsat imagery with very high-resolution imagery obtained from an unmanned aircraft system (UAS). The UAS imagery was used as 'drone truthing' data to train image classification algorithms. Our results show that UAS data can successfully be used to help discriminate similar land-cover/use classes (oil palm plantation vs. reforestation vs. logged forest) with consistently high identification of over 75% on the generated thematic map, where the oil palm detection rate was as high as 89%. Because UAS is employed increasingly in conservation projects, this approach can be used in a large variety of them to improve land-cover classification or aid-specific mapping needs.

This l)al)wr describes the latest version of the University of Florida (UF)'s uninanned autoioino... more This l)al)wr describes the latest version of the University of Florida (UF)'s uninanned autoioinoits vehicle (Rf,W). named the MAKO MAKO. The NIAKO MAKO can operate in fully autonomous w%avipoint navigation in1ode. including autoniomus takeoff and landing, allowing for repeatable, predictable grotind coverage. Other features essential for itht mission profile include hand-launch ability for takeoff, as wIell as a waterproof Fuselage for aquatic landing. Low-altitude, high-r,esolution imaging is tacilitated by a cruise speed of I5 ni/s. -Fle sensor payload on the MAKO is a digital single lens reflex (D)SLR) camera olpcrawed at the shortesi exposure to minimize the etftect ofi motion bluiiing. The resulting images are capable of resolving objects on the ground as small as six cm. The MAKO has been used for a number of applications, including mapping ivading bird nests in the Florida Everglades and elsewhere, nionitloring the etlicacy of oetuliant spray progranis in Lake Okeechobee. identification of invasive exotic vegetation, and m1app)ing of bisoln.

Unmanned aircraft systems (UASs) are proposed as a useful alternative to manned aircraft for some... more Unmanned aircraft systems (UASs) are proposed as a useful alternative to manned aircraft for some aerial wildlife surveys.

In order to monitor natural and anthropogenic disturbance effects to wetland ecosystems, it is ne... more In order to monitor natural and anthropogenic disturbance effects to wetland ecosystems, it is necessary to employ both accurate and rapid mapping of wet graminoid/sedge communities. Thus, it is desirable to utilize automated classification algorithms so that the monitoring can be done regularly and in an efficient manner. This study developed a classification and accuracy assessment method for wetland mapping of at-risk plant communities in marl prairie and marsh areas of the Everglades National Park. Maximum likelihood (ML) and Support Vector Machine (SVM) classifiers were tested using 30.5 cm aerial imagery, the normalized difference vegetation index (NDVI), first and second order texture features and ancillary data. Additionally, appropriate window sizes for different texture features were estimated using semivariogram analysis. Findings show that the addition of NDVI and texture features increased classification accuracy from 66.2% using the ML classifier (spectral bands only) to 83.71% using the SVM classifier (spectral bands, NDVI and first order texture features).

Mapping of wetlands (marsh vs. swamp vs. upland) is a common remote sensing application.Yet, disc... more Mapping of wetlands (marsh vs. swamp vs. upland) is a common remote sensing application.Yet, discriminating between similar freshwater communities such as graminoid/sedge from remotely sensed imagery is more difficult. Most of this activity has been performed using medium to low resolution imagery. There are only a few studies using high spatial resolution imagery and machine learning image classification algorithms for mapping heterogeneous wetland plant communities. This study addresses this void by analyzing whether machine learning classifiers such as decision trees (DT) and artificial neural networks (ANN) can accurately classify graminoid/sedge communities using high resolution aerial imagery and image texture data in the Everglades National Park, Florida. In addition to spectral bands, the normalized difference vegetation index, and first-and second-order texture features derived from the near-infrared band were analyzed. Classifier accuracies were assessed using confusion tables and the calculated kappa coefficients of the resulting maps. The results indicated that an ANN (multilayer perceptron based on back propagation) algorithm produced a statistically significantly higher accuracy (82.04 %) than the DT (QUEST) algorithm (80.48 %) or the maximum likelihood (80.56 %) classifier (α<0.05). Findings show that using multiple window sizes provided the best results. First-order texture features also provided computational advantages and results that were not significantly different from those using second-order texture features. Environ Monit Assess ( 2 0 1 5 ) 1 8 7 : 2 6 2

Food security has become a global concern for humanity with rapid population growth, requiring a ... more Food security has become a global concern for humanity with rapid population growth,
requiring a sustainable assessment of natural resources. Soil is one of the most important sources
that can help to bridge the food demand gap to achieve food security if well assessed and managed.
The aim of this study was to determine the soil quality index (SQI) for El Fayoum depression in the
Western Egyptian Desert using spatial modeling for soil physical, chemical, and biological properties
based on the MEDALUS methodology. For this purpose, a spatial model was developed to evaluate
the soil quality of the El Fayoum depression in theWestern Egyptian Desert. The integration between
Digital Elevation Model (DEM) and Sentinel-2 satellite image was used to produce landforms and
digital soil mapping for the study area. Results showed that the study area located under six classes
of soil quality, e.g., very high-quality class represents an area of 387.12 km2 (22.7%), high-quality class
occupies 441.72 km2 (25.87%), the moderate-quality class represents 208.57 km2 (12.21%), slightly
moderate-quality class represents 231.10 km2 (13.5%), as well as, a low-quality class covering an area
of 233 km2 (13.60%), and very low-quality class occupies about 206 km2 (12%). The Agricultural Land
Evaluation System for arid and semi-arid regions (ALESarid) was used to estimate land capability.
Land capability classes were non-agriculture class (C6), poor (C4), fair (C3), and good (C2) with an
area 231.87 km2 (13.50%), 291.94 km2 (17%), 767.39 km2 (44.94%), and 416.07 km2 (24.4%), respectively.
Land capability along with the normalized difference vegetation index (NDVI) used for validation
of the proposed model of soil quality. The spatially-explicit soil quality index (SQI) shows a strong
significant positive correlation with the land capability and a positive correlation with NDVI at R2
0.86 (p < 0.001) and 0.18 (p < 0.05), respectively. In arid regions, the strategy outlined here can easily
be re-applied in similar environments, allowing decision-makers and regional governments to use
the quantitative results achieved to ensure sustainable development.

Several preprocessing steps have to be performed to reliably study mountainous terrains with sate... more Several preprocessing steps have to be performed to reliably study mountainous terrains with satellite imagery, and one of the most important is topographic correction. The illumination conditions of these images often vary due to unequal physical properties, such as sun elevation angles and different illumination levels, while the temporal resolution of the imagery has to be accounted for as well. Two digital elevation models, a pre-classification/stratification approach and several correction methods were tested on selected medium resolution sensors. The processed images were selected to encompass different land cover types and temporal variations in solar illumination and a range of topography. It has been demonstrated over several study sites that the empirical-statistical method in combination with a pre-classification/stratification approach provided exceptional results in correcting topographic effects of the satellite imagery using the ASTER Global Digital Elevation Model. The pre-classification/stratification approach was used to split the different land cover types into "strata" which were corrected individually with the selected topographic correction method to achieve better reduction of the terrain effects.

Cities are increasingly promoting policies that increase and conserve urban forests based largely... more Cities are increasingly promoting policies that increase and conserve
urban forests based largely on biophysical and land use-cover metrics. This study
demonstrates how socioeconomic factors need to be considered in geospatial analyses
when formulating urban greening policies. Using remote sensing, geographical
information systems, spatial field and census data, and policy analyses, we analyzed
the effectiveness of urban forest cover policies that included socioeconomic factors
when quantifying urban forest cover. We found that urban forest cover was heterogeneous
across the study area and non-white residents younger than 19 and greater
than 45 years old living in rentals were more likely to reside in areas with less urban
forest cover than any other age cohort. Our analyses also indicated that urban forest
cover was temporally variable and demographic factors unique to Miami-Dade
County bring to light the complexity of establishing homogenous, county-wide “tree
canopy” and urban greening policy goals. We present a localized socioeconomic and
ecologically based geospatial approach for formulating urban forest cover goals.

Coastal communities in the southeast United States have regularly experienced severe hurricane im... more Coastal communities in the southeast United States have regularly experienced severe hurricane impacts. To better facilitate recovery efforts in these communities following natural disasters, state and federal agencies must respond quickly with information regarding the extent and severity of hurricane damage and the amount of tree debris volume. A tool was developed to detect downed trees and debris volume to better aid disaster response efforts and tree debris removal. The tool estimates downed tree debris volume in hurricane affected urban areas using a Leica Airborne Digital Sensor (ADS40) and very high resolution digital images. The tool employs a Sobel edge detection algorithm combined with spectral information based on color filtering using 15 different statistical combinations of spectral bands. The algorithm identified downed tree edges based on contrasts between tree stems, grass, and asphalt and color filtering was then used to establish threshold values. Colors outside these threshold values were replaced and excluded from the detection processes. Results were overlaid and an "edge line" was placed where lines or edges from longer consecutive segments and color values within the threshold were met. Where two lines were paired within a very short distance in the scene a polygon was drawn automatically and, in doing so, downed tree stems were detected. Tree stem diameter-volume bulking factors were used to estimate post-hurricane tree debris volumes. Images following Hurricane Ivan in 2005 and Hurricane Ike in 2008 were used to assess the error of the tool by comparing downed tree counts and subsequent debris volume estimates with post-hurricane photo-interpreted downed tree counts and actual field measured estimates of downed tree debris volume. The errors associated with the use of the tool and potential applications are also presented.

Cities are increasingly promoting policies that increase and conserve urban forests based largely... more Cities are increasingly promoting policies that increase and conserve urban forests based largely on biophysical and land use-cover metrics. This study demonstrates how socioeconomic factors need to be considered in geospatial analyses when formulating urban greening policies. Using remote sensing, geographical information systems, spatial field and census data, and policy analyses, we analyzed the effectiveness of urban forest cover policies that included socioeconomic factors when quantifying urban forest cover. We found that urban forest cover was heterogeneous across the study area and non-white residents younger than 19 and greater than 45 years old living in rentals were more likely to reside in areas with less urban forest cover than any other age cohort. Our analyses also indicated that urban forest cover was temporally variable and demographic factors unique to Miami-Dade County bring to light the complexity of establishing homogenous, county-wide "tree canopy" and urban greening policy goals. We present a localized socioeconomic and ecologically based geospatial approach for formulating urban forest cover goals.

EDU can potentially reduce O 3 visible injury in cutleaf coneflower, but may have phytotoxic effe... more EDU can potentially reduce O 3 visible injury in cutleaf coneflower, but may have phytotoxic effects regarding plant productivity.

EDU effectively protect plants against ambient ozone.

EDU can potentially ameliorate negative effects of O 3 on nutritive quality in purple coneflower.... more EDU can potentially ameliorate negative effects of O 3 on nutritive quality in purple coneflower. Abstract Purple coneflower plants (Echinacea purpurea) were placed into open-top chambers (OTCs) for 6 and 12 weeks in 2003 and 2004, respectively, and exposed to charcoal-filtered air (CF) or twice-ambient (2Â) ozone (O 3 ) in 2003, and to CF, 2Â or non-filtered (NF), ambient air in 2004. Plants were treated with ethylenediurea (EDU) weekly as a foliar spray. Foliar symptoms were observed in >95% of the plants in 2Â-treated OTCs in both years. Above-ground biomass was not affected by 2Â treatments in 2003, but root and total-plant biomass decreased in 2004. As a result of higher concentrations of select cell wall constituents (% ADF, NDF and lignin) nutritive quality was lower for plants exposed to 2Â-O 3 in 2003 and 2004 (26% and 17%, respectively). Significant EDU Â O 3 interactions for concentrations of cell wall constituents in 2003 indicated that EDU ameliorated O 3 effects on nutritive quality. Interactions observed in 2004 were inconsistent.