Parcel Level Research Articles

Predicting Detached Housing Vacancy: A Multilevel Analysis

Urban shrinkage is a critical issue in local small- and medium-sized cities in Korea. While there have been several studies to analyze the causes and consequences of vacancy increases, most have only focused on socioeconomic associations at larger scale and failed to consider individual housing level characteristics, primarily due to a lack of appropriate data. Based on data including 52,400 individual parcels, this study analyzes the primary contributors to vacant properties and their spatial distribution through a multilevel model design based on data for each parcel. Then, we identify areas at high risk of vacancy in the future to provide evidence to establish policies for improving the local environment. Results indicate that construction year, building structure, and road access conditions have a significant effect on vacant properties at the individual parcel level, and the presence of schools and hypermarket within 500 m are found to decrease vacant properties. Further, prediction outcomes show that the aged city center and areas with strict regulations on land use are expected to have a higher vacancy rate. These findings are used to provide a set of data-based revitalization strategies through the development of a vacancy prediction model.

Informal Settlement Upgrading and the Rise of Rental Housing in São Paulo, Brazil

Rental housing was historically a minimal feature of urban informality. Now it is surging amid municipal attempts to “upgrade” informal settlements in São Paulo, Brazil. Drawing upon a mixed-methodological study of two favelas on São Paulo’s east side, we analyze how cycles of upgrading shape informal rental housing at the urban, community, block, and parcel levels, providing detailed comparative data for 2010–2020. Our findings suggest that rental housing redevelopment can increase precarity in urban living, but is an important source of low-income housing in already built-up and “consolidated” settlements where access is declining. Our study emphasizes the need for scholars, policy makers, and planners to further explore the praxis of informal renting and rental housing, which can be effective conduits for channeling public investments across consolidated informal settlements and into individual dwellings.

Anomaly Detection and Classification in Multispectral Time Series Based on Hidden Markov Models

Monitoring agriculture from satellite remote sensing data, such as multispectral images, has become a powerful tool since it has demonstrated a great potential for providing timely and accurate knowledge of crops. Detecting anomalies in time series of multispectral remote sensing images for crop monitoring is generally performed using a large sample of historical data at a pixel level. Conversely, this article presents a framework for anomaly detection (AD), localization, and classification that exploits the temporal information contained in a given season at a parcel level to detect and localize outliers using hidden Markov models (HMMs). Specifically, the AD part is based on the learning of HMM parameters associated with unlabeled normal data that are used in a second step to detect abnormal crop parcels referred to as anomalies. The learned HMM can also be used in time segments to temporally localize the anomalies affecting the crop parcels. The detected and localized anomalies are finally classified using a supervised classifier, e.g., based on support vector machines. The proposed framework is applicable to images partially covered by clouds and can handle a set of crop parcels acquired in the same season bypassing problems due to crop rotations. Numerical experiments are conducted on synthetic and real data, where the real data correspond to vegetation indices extracted from several multitemporal Sentinel-2 images of rapeseed crops. The proposed approach is compared to standard AD methods yielding better detection rates with the advantage of allowing anomalies to be localized and characterized.

Parcel Level Staple Crop Type Identification from Time Series Remote Sensing Images Based on Newly Defined Red-Edge Vegetation Indices And Ornn

Parcel Level Staple Crop Type Identification from Time Series Remote Sensing Images Based on Newly Defined Red-Edge Vegetation Indices And Ornn

Monitoring Plant Diversity to Support Agri-Environmental Schemes: Evaluating Statistical Models Informed by Satellite and Local Factors in Southern European Mountain Pastoral Systems

The spatial monitoring of plant diversity in the endangered species-rich grasslands of European mountain pastoral systems is an important step for fairer and more efficient Agri-Environmental policy schemes supporting conservation. This study assessed the underlying support for a spatially explicit monitoring of plant species richness at parcel level (policy making scale) in Southern European mountain grasslands, with statistical models informed by Sentinel-2 satellite and environmental factors. Twenty-four grassland parcels were surveyed for species richness in the Peneda-Gerês National Park, northern Portugal. Using a multi-model inference approach, three competing hypotheses guided by the species-scaling theoretical framework were established: species–area (P1), species–energy (P2) and species–spectral heterogeneity (P3), each representing a candidate spatial pathway to predict species richness. To evaluate the statistical support of each spatial pathway, generalized linear models were fitted and model selection based on Akaike information criterion (AIC) was conducted. Later, the performance of the most supported spatial pathway(s) was assessed using a leave-one-out cross validation. A model guided by the species–energy hypothesis (P2) was the most parsimonious spatial pathway to monitor plant species richness in mountain grassland parcels (P2, AICc = 137.6, ∆AIC = 0.0, wi = 0.97). Species–area and species–spectral heterogeneity pathways (P1 and P3) were less statistically supported (ΔAICc values in the range 5.7–10.0). The underlying support of the species–energy spatial pathway was based on Sentinel-2 satellite data, namely on the near-infrared (NIR) green ratio in the spring season (NIR/Greenspring) and on its ratio of change between spring and summer (NIR/Greenchange). Both predictor variables related negatively to species richness. Grassland parcels with lower values of near-infrared (NIR) green ratio and lower seasonal amplitude presented higher species richness records. The leave-one-out cross validation indicated a moderate performance of the species–energy spatial pathway in predicting species richness in the grassland parcels covered by the dataset (R2 = 0.44, RMSE = 4.3 species, MAE = 3.5 species). Overall, a species–energy framework based on Sentinel 2 data resulted in a promising spatial pathway for the monitoring of species richness in mountain grassland parcels and for informing decision making on Agri-Environmental policy schemes. The near-infrared (NIR) green ratio and its change in time seems a relevant variable to deliver predictions for plant species richness and further research should be conducted on that.

Monitoring Rainfed Alfalfa Growth in Semiarid Agrosystems Using Sentinel-2 Imagery

The aim of this study was to assess the utility of Sentinel-2 images in the monitoring of the fractional vegetation cover (FVC) of rainfed alfalfa in semiarid areas such as that of Bardenas Reales in Spain. FVC was sampled in situ using 1 m2 surfaces at 172 points inside 18 alfalfa fields from late spring to early summer in 2017 and 2018. Different vegetation indices derived from a series of Sentinel-2 images were calculated and were then correlated with the FVC measurements at the pixel and parcel levels using different types of equations. The results indicate that the normalized difference vegetation index (NDVI) and FVC were highly correlated at the parcel level (R2 = 0.712), whereas the correlation at the pixel level remained moderate across each of the years studied. Based on the findings, another 29 alfalfa plots (28 rainfed; 1 irrigated) were remotely monitored operationally for 3 years (2017–2019), revealing that location and weather conditions were strong determinants of alfalfa growth in Bardenas Reales. The results of this study indicate that Sentinel-2 imagery is a suitable tool for monitoring rainfed alfalfa pastures in semiarid areas, thus increasing the potential success of pasture management.

The heterogeneous effects of agricultural conservation easements on the loss of farmland to development in New England

Farmland near cities can provide diverse non-market benefits, such as recreational open space, landscape appeal, local food production, ecological habitat, and water regulation, which risk being underprovided by markets in the absence of intervention. The rapid loss of farmland to development has long been a concern of conservation non-profits and policymakers in the U.S. and globally, which has recently gained traction due to President Biden's decision to include farmlands in the federal goal to protect 30% of U.S. land by 2030. Agricultural conservation easements are a widespread policy tool to protect farmland from conversion to development in the long term. However, the extent to which these easements causally reduce farmland loss to development within their boundaries is rarely studied at the spatial scale at which easement adoption decisions are made: the individual parcel. Here we estimate the impacts of agricultural conservation easements on farmland loss at the parcel level at a large spatial scale and over a long time horizon. Our case study from six New England states uses a rich dataset of 1.97 million parcels, novel estimates of annual parcel-level land cover change from 1988 to 2016, and quasi-experimental counterfactual estimation strategies to estimate the extent to which 3,959 farmland easements causally avoided conversion of cropland to development. Our results suggest that agricultural conservation easements have significantly reduced farmland loss to development. However, the overall magnitude of avoided rates of farmland loss on easements is very small (0.0067% ± 0.0019% of parcel area per annum), largely because of a low background rate of farmland loss across New England (0.0027% of total area per annum). A spatial allocation of agricultural easements towards more rapidly urbanizing counties (with farmland loss rates of up to 0.0281% of county area per annum) would increase the causal impacts of this instrument. Overall, our findings suggest that the spatial allocation of farmland easements has historically not prioritized the highest-threat locations where impacts would be most noticeable, indicating that allocation is driven by a variety of goals in addition to avoiding farmland loss.

Within-Field Rice Yield Estimation Based on Sentinel-2 Satellite Data

Rice is considered one of the most important crops in the world. According to the Food and Agriculture Organization of the United Nations (FAO), rice production has increased significantly (156%) during the last 50 years, with a limited increase in cultivated area (24%). With the recent advances in remote sensing technologies, it is now possible to monitor rice crop production for a better understanding of its management at field scale to ultimately improve rice yields. In this work, we monitor within-field rice production of the two main rice varieties grown in Valencia (Spain) JSendra and Bomba during the 2020 season. The sowing date of both varieties was May 22–25, while the harvesting date was September 15–17 for Bomba and October 5–8 for JSendra. Rice yield data was collected over 66.03 ha (52 fields) by harvesting machines equipped with onboard sensors that determine the dry grain yield within irregular polygons of 3–7 m width. This dataset was split in two, selecting 70% of fields for training and 30% for validation purposes. Sentinel-2 surface reflectance spectral data acquired from May until September 2020 was considered over the test area at the two different spatial resolutions of 10 and 20 m. These two datasets were combined assessing the best combination of spectral reflectance bands (SR) or vegetation indices (VIs) as well as the best timing to infer final within-field yields. The results show that SR improves the performance of models with VIs. Furthermore, the correlation of each spectral band and VIs with the final yield changes with the dates and varieties. Considering the training data, the best correlation with the yields is obtained on July 4, with R2 for JSendra of 0.72 at 10 m and 0.76 at 20 m resolution, while the R2 for Bomba is 0.87 at 10 m and 0.92 at 20 m resolution. Based on the validation dataset, the proposed models provide within-field yield modelling Mean Absolute Error (MAE) of 0.254 t×ha−1 (Mean Absolute Percentage Error, MAPE, of 3.73%) for JSendra at 10 m (0.240 t×ha−1; 3.48% at 20 m) and 0.218 t×ha−1 (MAPE 5.82%) for Bomba (0.223 t×ha−1; 5.78% at 20 m) on July 4, that is three months before harvest. At parcel level the model’s MAE is 0.176 t×ha−1 (MAPE 2.61%) for JSendra and 0.142 t×ha−1 (MAPE 4.51%) for Bomba. These results confirm the close correlation between the rice yield and the spectral information from satellite imagery. Additionally, these models provide a timeliness overview of underperforming areas within the field three months before the harvest where farmers can improve their management practices. Furthermore, it highlights the importance of optimum agronomic management of rice plants during the first weeks of rice cultivation (40–50 days after sowing) to achieve high yields.

Understanding E-Scooter Incidents Patterns in Street Network Perspective: A Case Study of Travis County, Texas

Dockless electric scooter (E-scooters) services have emerged in the United States as an alternative form of micro transit in the past few years. With the increasing popularity of E-scooters, it is important for cities to manage their usage to create and maintain safe urban environments. However, E-scooter safety in U.S. urban environments remains unexplored due to the lack of traffic and crash data related to E-scooters. Our study objective is to better understand E-scooter crashes from a street network perspective. New parcel level street network data are obtained from Zillow and curated in Geographic Information System (GIS). We conducted local Moran’s I and independent Z-test to compare where and how the street network that involves E-scooter crash differs spatially with traffic incidents. The analysis results show that there is a spatial correlation between E-scooter crashes and traffic incidents. Nevertheless, E-scooter crashes do not fully replicate characteristics of traffic incidents. Compared to traffic incidents, E-scooter incidents tend to occur adjacent to traffic signals and on primary roads.

The Role of Demographic Data Bias in the Under-Provision of Retail: A Quasi-Experimental Case Study of Low Food Access Neighborhoods in Twenty-Five Southern Metros

Municipal policy attempting to remediate low food access neighborhoods tends to focus on improving demand conditions in these neighborhoods. We investigate the role of two fundamental measures of food demand (population and income) and the biases inherent in these data in creating low food access neighborhoods. Population and income data were collected for a 1-mile radius surrounding seventy-one grocery desert sites in Southern U.S. metro areas, those results were compared to thirty-eight low-income, non-desert sites in the same metros. No significant difference between the demand characteristics of desert and non-desert sites was found in our sample—suggesting that policy may need to be refocused on issues other than demand metrics. In addition, we detected significant demand underestimation bias from one source commonly used by grocery stores. Given these findings we believe that parcel level characteristics such as visibility, accessibility, and buildability may play a larger role in remediating low food access than addressing demand conditions.

Health and equity opportunities of commercial corridors revitalization in Colorado

BACKGROUND AND AIM: In the U.S. transport corridors have received large attention as an opportunity area for urban regeneration. AIM. This study aims to quantify the health and equity impacts of land-use changes from low-density commercial corridors to high-density mixed-use commercial corridors with special attention on affordable housing and green spaces in Fort Collins and Denver, Colorado. METHODS: This study quantified opportunities on housing, employment, and green space of commercial corridors regeneration, based on land use scenarios at the parcel level. For the health assessment, we quantified health impacts on mortality and disease incidence related to increases of urban vegetation at parcel level in both cities through a quantitative health impact assessment approach. We developed two main policy scenarios: a) transforming low-dense commercial parcels into mixed-use developments; and b) transform vacant and parking surface lots into green spaces. RESULTS:In Denver, commercial corridors revitalization based on shifting low-density commercial parcels to mixed-use developments could absorb a 22% increment in city inhabitants, 29% new dwelling units, 59% new affordable housing, and 16% new jobs. In the same commercial corridors, the revitalization of vacant and parking surface lots to green spaces will increase 715 acres of green spaces across the city, resulting in a reduction of 35 premature deaths. In Fort Collins, commercial corridors revitalization could absorb a 77% increment in city inhabitants, 115% new dwelling units, 600% new affordable housing, and 63% new jobs. In the same commercial corridors, the revitalization of vacant and parking surface lots to green spaces will increase 54 acres of green spaces across the city, resulting in a reduction of 5 premature deaths. CONCLUSIONS:The regeneration of commercial corridors in Colorado could help to reduce urban sprawl, increase affordable housing, employment, green spaces, and health equity-related benefits. KEYWORDS: Commercial corridors, affordable housing, sprawl, green spaces, equity, health impact assessment

The missing exposure-outcome link: “equivalent” outdoor ambient temperatures for assigning heat-health effects to individual-level temperature exposures in three U.S. cities

BACKGROUND AND AIM: Heat-health epidemiologic studies often lack information on key determinants of exposure, including occupation, housing, and air conditioning, at the individual or parcel (home) level. Therefore, effect estimates specific to people in subgroups defined by various combinations of these determinants are often not available. We derived subgroup estimates from modeled individually-experienced temperatures (IETs) during 10 days around a past city-specific extreme heat event for three U.S. cities of varying climates: Atlanta, Georgia (hot-humid), Detroit, Michigan (temperate), and Phoenix, Arizona (hot-dry). METHODS: IETs were previously estimated for 2.5 million people in the three cities using modeled data: parcel-linked population microdata, housing-specific indoor temperatures, temporospatially-varying outdoor ambient temperatures, and time activity patterns. To each of the IETs we linked the daily outdoor ambient temperatures as measured at the regional airport—the temperature exposure often used in heat-health studies. We then fit a mixed-effects regression model to predict this often-used temperature exposure, ‘equivalent’ outdoor airport temperature, based on IET and housing type, demographics, air conditioning, and person-specific random effects. Using the equivalent temperatures, we assigned risk ratios (RRs, for increases in outdoor airport temperature) from existing literature on mortality, emergency department visits, and preterm births to each person-day and estimated attributable burdens (using age/race/sex) health event counts. RESULTS:The equivalent temperatures, RRs, and attributable burdens, by design, differed between population subgroups due to variability in indoor and outdoor temperature exposure. For example, heat-associated deaths were 150% higher among Blacks vs. Whites in Atlanta and 40% higher in Detroit. Emergency department visits were 19%, 13%, and 25% higher in Atlanta, Detroit, and Phoenix, respectively, among those without vs. with air conditioning. CONCLUSIONS:When IETs are known, these person-specific equivalent temperatures can be used to estimate subgroup and spatially-varying heat-health burdens for resource allocation for future heat waves and environmental and energy justice policies. KEYWORDS: climate change, indoor temperature exposure, urban heat island, climate vulnerability, extreme weather preparedness

Using big data for last mile performance evaluation: An accessibility-based approach

ObjectiveThe ‘last mile’ of public transportation describes the final leg of a transit journey. This paper brings an accessibility-based approach to last-mile performance evaluation at the parcel level by measuring desirable destinations reasonably reachable through accessible transit stations. MethodsAn accessibility-based last mile performance measure is developed to include destinations, attractiveness, and transit connectivity. Google Map API data is used to identify potential destinations and further evaluate their popularity. ResultsThe range of last-mile performance scores was 0–91.7954%, with a mean of 49.82% and a standard deviation of 61.61%, indicating high variation of the last mile performance in Chicago. Last mile problem areas in Chicago tend to cluster in more economically challenged areas. Income levels and housing sale price had positive relationships with last mile performance scores. ConclusionAreas with low last-mile accessibility performance are more likely to cluster in communities that have greater economic disadvantages, lower density, and less mixed land use, implying spatial inequality and disparity in overall accessibility. PracticeThe described approach can inform the development of strategic planning interventions to improve transit connectivity and to reduce the disparity of transit connectivity and accessibility across neighborhoods. ImplicationsThe evaluation of last mile connectivity needs to consider both access to transit station and access to potential destinations. The last mile performance score is highly influenced by neighborhood socioeconomic status.

Sentinel-1 Time Series for Crop Identification in the Framework of the Future CAP Monitoring

In this upcoming Common Agricultural Policy (CAP) reform, the use of satellite imagery is taking an increasing role for improving the Integrated Administration and Control System (IACS). Considering the operational aspect of the CAP monitoring process, the use of Sentinel-1 SAR (Synthetic Aperture Radar) images is highly relevant, especially in regions with a frequent cloud cover, such as Belgium. Indeed, SAR imagery does not depend on sunlight and is barely affected by the presence of clouds. Moreover, the SAR signal is particularly sensitive to the geometry and the water content of the target. Crop identification is often a pre-requisite to monitor agriculture at parcel level (ploughing, harvest, grassland mowing, intercropping, etc.) The main goal of this study is to assess the performances and constraints of a SAR-based crop classification in an operational large-scale application. The Random Forest object-oriented classification model is built on Sentinel-1 time series from January to August 2020 only. It can identify crops in the Walloon Region (south part of Belgium) with high performance: 93.4% of well-classified area, representing 88.4% of the parcels. Among the 48 crop groups, the six most represented ones get a F1-score higher or equal to 84%. Additionally, this research documents how the classification performance is affected by different parameters: the SAR orbit, the size of the training dataset, the use of different internal buffers on parcel polygons before signal extraction, the set of explanatory variables, and the period of the time series. In an operational context, this allows to choose the right balance between classification accuracy and model complexity. A key result is that using a training dataset containing only 3.2% of the total number of parcels allows to correctly classify 91.7% of the agricultural area. The impact of rain and snow is also discussed. Finally, this research analyses how the classification accuracy depends on some characteristics of the parcels like their shape or size. This allows to assess the relevance of the classification depending on those characteristics, as well as to identify a subset of parcels for which the global accuracy is higher.

Oilseed Rape (Brassica napus L.) Phenology Estimation by Averaged Stokes-Related Parameters

Accurate and timely knowledge of crop phenology assists in planning and/or triggering appropriate farming activities. The multiple Polarimetric Synthetic Aperture Radar (PolSAR) technique shows great potential in crop phenology retrieval for its characterizations, such as short revisit time, all-weather monitoring and sensitivity to vegetation structure. This study aims to explore the potential of averaged Stokes-related parameters derived from multiple PolSAR data in oilseed rape phenology identification. In this study, the averaged Stokes-related parameters were first computed by two different wave polarimetric states. Then, the two groups of averaged Stokes-related parameters were generated and applied for analyzing averaged Stokes-related parameter sensitivity to oilseed rape phenology changes. At last, decision tree (DT) algorithms trained using 60% of the data were used for oilseed rape phenological stage classification. Four Stokes parameters (g0, g1, g2 and g3) and eight sub parameters (degree of polarization m, entropy H, ellipticity angle χ, orientation angle φ, degree of linear polarization Dolp, degree of circular polarization Docp, linear polarization ratio Lpr and circular polarization ratio Cpr) were extracted from a multi-temporal RADARSAT-2 dataset acquired during the whole oilseed rape growth cycle in 2013. Their sensitivities to oilseed rape phenology were analyzed versus five main rape phenology stages. In two groups (two different wave polarimetric states) of this study, g0, g1, g2, g3, m, H, Dolp and Lpr showed high sensitivity to oilseed rape growth stages while χ, φ, Docp and Cpr showed good performance for phenology classification in previous studies, which were quite noisy during the whole oilseed rape growth circle and showed unobvious sensitivity to the crop’s phenology change. The DT algorithms performed well in oilseed rape phenological stage identification. The results were verified at the parcel level with left 40% of the point dataset. Five phenology intervals of oilseed rape were identified with no more than three parameters by simple but robust decision tree algorithm groups. The identified phenology stages agree well with the ground measurements; the overall identification accuracies were 71.18% and 79.71%, respectively. For each growth stage, the best performance occurred at stage S1 with the accuracy of 95.65% for Group 1 and 94.23% for Group 2, and the worst performance occurred at stage S3 and S5 with the values around 60%. Most of the classification errors may resulted from the indistinguishability of S3 and S5 using Stokes-related parameters.

Insights into the Impacts of Mega Transport Infrastructures on the Transformation of Urban Fabric: Case of BRT Lahore

State of the art mass transit systems, such as bus rapid transit (BRT), have appeared to be an innovative solution to meet the mobility needs of many world cities. However, their ability to transform surrounding urban fabric remains less explored in the developing world. This paper examines the impacts of BRT Lahore on land development patterns, considering the phenomena of land use revitalization, densification, and property reconfiguration. We have used quantitative approaches to gauge the changes in the urban fabric, with respect to the local neighborhood conditions and parcel level attributes, using multilevel logit models. The results of the logit models reveal heterogeneous impacts on conversions to commercial plazas, apartment buildings, and mixed-use buildings. The distance to the BRT station and size of the property is significantly associated with these conversions. Overall, one can notice a lack of a sense of place making and an absence of transit-oriented zoning around BRT stations. The desire of sustainable transit-oriented development to intensify land uses according to local conditions requires local urban designers and planners to think ahead of the curve by working on the regulatory and zoning restrictions governing the designs of built environment and also address the issues of gentrification and social inequity.

Development of Supply Curves for Biodiversity Offsets on Agricultural Land – a Case Study from the Stuttgart Region

Impacts on nature and landscape are to be offset in accordance with different nature conservation acts in various European countries. In Germany in particular, biodiversity offsets can also be made in advance, for instance, by booking them into eco-accounts, and then allocating them to an intervention. In Baden-Württemberg, these offset measures are assessed in eco credits in accordance with the Eco Account Regulation (ÖKVO). As a means of income diversification, farmers can voluntarily implement offset measures on their land, and then generate and sell corresponding eco credits. Using a geodata-based model, the potential for implementing biodiversity offsets on arable land – areas with major eco credit potential – is analysed from an economic perspective. The Stuttgart Region is a steadily growing conurbation in south-west Germany. It serves as a study region since the loss of farmland due to large-scale construction measures and the related offsetting are a major issue here. In the analysis, the gross margins of the crops grown, their yield capacity, the associated standard land values and the costs of possible offset measures are used to determine the net present value of the arable land at parcel level. From a theoretical point of view and depending on the market price for eco credits, there is a significant potential for offset measures on arable land. Production-integrated compensation (PIC) – an extensification of arable land use – is less economically viable than the conversion of arable land into grassland or its utilisation for nature conservation. There are major spatial disparities between the city of Stuttgart and the surrounding districts. The implementation of biodiversity offsets is not economically viable at a price of less than € 1.00 per eco credit in the city of Stuttgart. By contrast, in surrounding districts, offset measures may be economically viable and implemented on a large scale for less than € 0.30. This is particularly relevant as the districts concerned are located in the same natural area as the city of Stuttgart and the eco credits can, therefore, be attributed in the event of interventions. Based on derived supply curves, decision-makers can see the scale of additional costs of biodiversity offset measures if they are implemented in a spatially restricted region. The analyses presented here can help decision-makers to more easily weigh up the desired natural characteristics and economic effects in the context of agricultural land.

A Scalable Machine Learning Pipeline for Paddy Rice Classification Using Multi-Temporal Sentinel Data

The demand for rice production in Asia is expected to increase by 70% in the next 30 years, which makes evident the need for a balanced productivity and effective food security management at a national and continental level. Consequently, the timely and accurate mapping of paddy rice extent and its productivity assessment is of utmost significance. In turn, this requires continuous area monitoring and large scale mapping, at the parcel level, through the processing of big satellite data of high spatial resolution. This work designs and implements a paddy rice mapping pipeline in South Korea that is based on a time-series of Sentinel-1 and Sentinel-2 data for the year of 2018. There are two challenges that we address; the first one is the ability of our model to manage big satellite data and scale for a nationwide application. The second one is the algorithm’s capacity to cope with scarce labeled data to train supervised machine learning algorithms. Specifically, we implement an approach that combines unsupervised and supervised learning. First, we generate pseudo-labels for rice classification from a single site (Seosan-Dangjin) by using a dynamic k-means clustering approach. The pseudo-labels are then used to train a Random Forest (RF) classifier that is fine-tuned to generalize in two other sites (Haenam and Cheorwon). The optimized model was then tested against 40 labeled plots, evenly distributed across the country. The paddy rice mapping pipeline is scalable as it has been deployed in a High Performance Data Analytics (HPDA) environment using distributed implementations for both k-means and RF classifiers. When tested across the country, our model provided an overall accuracy of 96.69% and a kappa coefficient 0.87. Even more, the accurate paddy rice area mapping was returned early in the year (late July), which is key for timely decision-making. Finally, the performance of the generalized paddy rice classification model, when applied in the sites of Haenam and Cheorwon, was compared to the performance of two equivalent models that were trained with locally sampled labels. The results were comparable and highlighted the success of the model’s generalization and its applicability to other regions.

How does urban green space feature influence physical activity diversity in high-density built environment? An on-site observational study

Urban green space (UGS) plays an important role in maintaining and promoting public health because it provides opportunities for outdoor physical activities (PA). Many studies have proved that different UGS characteristics, such as area, shape, distance, etc., have a certain impact on the intensity or frequency of PA. It is noteworthy that most studies are completed on a city scale, which explains how ensemble characteristics of a park affect PA. However, few researchers have explored the relations between UGS features and physical activities at the parcel level. To address this gap, this study has investigated the effect of multiple features of UGS on the diversity of PA, with downtown Shanghai as the study site.To identify types of PA and features of UGS based on on-site observation of a total of 101 sites in five urban parks located in a central district of Shanghai, China, we established an ordinal logistic regression with total types of PA (PA diversity hereafter) as the dependent variable and UGS characteristics as independent variables. The results show that green coverage ratio and diversity of shrubs are positively related to PA diversity, while diversity of trees has an inverse impact. Moreover, the paved area shape index and green view ratio are negatively correlated with PA diversity, which suggests that the irregular shape of paved areas with winding edges and excessive green view ratio are factors detrimental to PA diversity. These findings provide pertinent guidance in the design of UGS for the promotion of PA.

Remotely Sensed Phenology Monitoring and Land-cover Classification for the Localization of the Endemic Argan Tree in the Southern-west of Morocco

ABSTRACT Argania spinosa also known as the argan tree is an endemic plant of Morocco. Despite having the ability to subsist in extreme drought conditions, it is threatened by soil land clearing, overexploitation, and absence of natural regeneration, causing a worrying decline in both spatial extent and density. The spatial extent of dryland forests is debated, as estimates of forest areas in drylands are uncertain. The present study aims to map and locate the spatial distribution of the argan trees at Smimou community located in Essaouira province, south-eastern Morocco, using satellite images and a double-classification process to overcome separability problems. The work focuses on the characterization and comparison of the unique phenological patterns of argan with the other present land-cover classes. NDVI products were derived from a Sentinel-2 time-series covering one year (2018 to 2019), then ground samples were used to extract phenological profiles at parcel level then at tree level, to feed representative calibration samples to Support Vector Machine classifier. The outcome was integrated with an elevation model in a Decision Tree to reclassify mixed areas. The results indicated an F1-score and an overall accuracy of 91.27% and 92.60% respectively, a promising technique for updating argan extent at national scale.