Spatial Autocorrelation Statistics Research Articles

Spatial Patterns and Spatial Autocorrelations of Wetland Changes in China During 2003-2013

Wetlands play an important role in maintaining ecosystem functions. Wetlands in China have suffered intensive human disturbance, especially before 2000, resulting in great losses and degradation. Therefore, two national wetland resource surveys were carried out by the Chinese government during 1998-2003 and 2008-2013 to determine the status of wetlands, understand their dynamics, and provide substantial data that can aid scientific wetland conservation. Based on the survey data, the spatial pattern and spatial autocorrelation were explored using the standard deviation ellipse as well as global and local spatial autocorrelation statistics. GIS mapping was employed to display the results via the visualization of the spatial patterns and relationships. Results indicate that:① Changes in the wetlands of China are significant and exhibit obvious regional differences. The center of the ellipse of the total wetlands has moved to the west, and the degree of spatial differentiation between natural wetlands and artificial wetlands is increasing. There is no significant spatial autocorrelation for changes in artificial wetlands. ② The changes in natural wetlands are significantly spatially autocorrelated and clustered, which are identified by global Moran's I and local Moran's I. The hotspots of natural wetland change are concentrated primarily in Qinghai, Tibet, and Sichuan. The individual hotspot is in Inner Mongolia and the cold spot is in Henan. The difference in spatial autocorrelation between natural and artificial wetland changes indicates that natural wetland changes have shown spatial continuity, while artificial wetland changes have shown strong spatial randomness. ③ Some useful spatial associations are used to delineate wetland conservation effects. Then, three major or five minor effective protection management regions are identified. Wetland conservation efforts should be continuously strengthened and improved, especially in the middle-lower Yangtze River region of central China. The use of Moran statistics helps to reveal spatial autocorrelation and identify the conservation effects in wetland changes, which can provide a basis for decision-making in regional wetland conservation and management systems.

Spatio-Temporal Variations of Satellite-Based PM2.5 Concentrations and Its Determinants in Xinjiang, Northwest of China.

With the aggravation of air pollution in recent years, a great deal of research on haze episodes is mainly concentrated on the east-central China. However, fine particulate matter (PM2.5) pollution in northwest China has rarely been discussed. To fill this gap, based on the standard deviational ellipse analysis and spatial autocorrelation statistics method, we explored the spatio-temporal variation and aggregation characteristics of PM2.5 concentrations in Xinjiang from 2001 to 2016. The result showed that annual average PM2.5 concentration was high both in the north slope of Tianshan Mountain and the western Tarim Basin. Furthermore, PM2.5 concentrations on the northern slope of the Tianshan Mountain increased significantly, while showing an obviously decrease in the western Tarim Basin during the period of 2001–2016. Based on the result of the geographical detector method (GDM), population density was the most dominant factor of the spatial distribution of PM2.5 concentrations (q = 0.550), followed by road network density (q = 0.423) and GDP density (q = 0.413). During the study period (2001–2016), the driving force of population density on the distribution of PM2.5 concentrations showed a gradual downward trend. However, other determinants, like DEM (Digital elevation model), NSL (Nighttime stable light), LCT (Land cover type), and NDVI (Normalized Difference Vegetation Index), show significant increased trends. Therefore, further effort is required to reveal the role of landform and vegetation in the spatio-temporal variations of PM2.5 concentrations. Moreover, the local government should take effective measures to control urban sprawl while accelerating economic development.

Compositional and contextual factors associated with drug overdose deaths in the United States

Background: In 2017, the US Department of Health and Human Services declared the Opioid epidemic a public health emergency. In the US, emergency rooms treat more than 1,000 people each day for drug overdose, and 115 of them die. This study examines compositional and contextual factors associated with drug overdose deaths rates in the US. Methods: Local spatial autocorrelation statistics were used to estimate hot spot areas to identify census tracts with high risk of drug overdose death. Logistic regressions investigated the relationship between drug overdose death rates and various compositional and contextual variables across census tracks. Results: The adjusted logistic model shows that compositional variables: depression (OR = 2.47 [2.37–2.58]), poor mental health (OR = 1.71 [1.63–1.79]), median age 1.41 (1.36–1.47) and the percentage of people with a high school diploma (OR = 1.30 [1.24–1.35]) were positively associated with the rate of drug overdose deaths. On the other hand, contextual variables: the percentage having health insurance (OR = 0.66 [0.64–0.69]), the Theil’s H index (OR = 0.69 [0.66–0.71]), population density (OR = 0.80 [0.77–0.84]), poverty (OR = 0.90 [0.86–0.95]), and median household income (OR = 0.91[0.86–0.96]) were negatively associated with drug overdose deaths. Discussion: The analysis reveals a consistently strong association between compositional mental health factors and census tract-level death rates from drug overdose.

Formulating an Innovative Spatial-Autocorrelation-based Method for Identifying Road Accident Hot Zones

The traditional method of identifying road accident hot zones is through the examination of accident frequency and nature, which sometimes, can be subjective and inaccurate. To overcome the limitation of the traditional method, researchers have applied Geographic Information System (GIS) approaches to identify and visualise road traffic accidences in real-time. However, these approaches still treat accidences as occasional and discrete events and can not support accurate analysis and prediction of accidences at some point. This paper takes the spatial autocorrelation nature of accidents (i.e. the interdependence of accident data and the relationship between the accident and space) into account and proposes an innovative spatial-autocorrelation-based method to identify freeway accident hot zones. Based on the spatial autocorrelation and mathematical statistics, this method constructs a point-line connectivity network to realise the space localisation and validation of accidents. Combined with GIS approaches, our approach can also automatically identify and visualise accident-prone areas. At the moment, the approach has been tentatively applied in a highway in China. The result demonstrates an algorithm behind the approach, which can effectively convert accident data into spatial data, cluster accident hot zones of any length and predict the whereabouts of likely accidents in the future. In conclusion, the robustness and accuracy of the approach innovates this study.

Effective range of non-cell autonomous activator and inhibitor peptides specifying plant stomatal patterning.

ABSTRACTStomata are epidermal valves that facilitate gas exchange between plants and their environment. Stomatal patterning is regulated by the EPIDERMAL PATTERING FACTOR (EPF) family of secreted peptides: EPF1 enforces stomatal spacing, whereas EPIDERMAL PATTERNING FACTOR-LIKE9 (EPFL9), also known as Stomagen, promotes stomatal development. It remains unknown, however, how far these signaling peptides act. Utilizing Cre-lox recombination-based mosaic sectors that overexpress either EPF1 or Stomagen in Arabidopsis cotyledons, we reveal a range within the epidermis and across the cell layers in which these peptides influence patterns. To determine their effective ranges quantitatively, we developed a computational pipeline, SPACE (stomata patterning autocorrelation on epidermis), that describes probabilistic two-dimensional stomatal distributions based upon spatial autocorrelation statistics used in astrophysics. The SPACE analysis shows that, whereas both peptides act locally, the inhibitor EPF1 exerts longer range effects than the activator Stomagen. Furthermore, local perturbation of stomatal development has little influence on global two-dimensional stomatal patterning. Our findings conclusively demonstrate the nature and extent of EPF peptides as non-cell autonomous local signals and provide a means for quantitative characterization of complex spatial patterns in development.This article has an associated ‘The people behind the papers’ interview.

Spatiotemporal Distribution of Tuberculosis during Urbanization in the New Urban Area of Nanchang City, China, 2010-2018.

Background: Urbanization will play a key role in ending the tuberculosis (TB) epidemic by 2030, but understanding the relationship between urbanization and the health threats posed by TB is incomplete. Therefore, this study aimed to explore the spatiotemporal distribution of TB at the township level during urbanization in the new urban area of Nanchang. Methods: Seasonal-trend decomposition of time series analysis was used to explore the seasonal distribution and trend analysis. Global and local spatial autocorrelation statistics, and space–time scan statistics were performed to detect the spatiotemporal clusters of TB cases in the new urban area of Nanchang from 2010 to 2018. Results: A total of 3245 TB cases were reported in the study area from 2010 to 2018. Of all the TB cases, 68% occurred in individuals older than 40 years old, 73.2% were male cases, and 56.6% were farmers. The primary seasonal peak was in late spring (April), and a smaller peak was in early autumn (September). The results of local indicators of spatial association showed that Jiaoqiao town and Changleng town might be “High–High” clusters. The most likely spatiotemporal cluster was located in the southwest of the study area in 2010, which included five towns, and then shifted to the northeast gradually. Across 2010 to 2018, nine spatiotemporal clusters were identified. The most likely cluster was located at the northeast of the study area. The center of this area was in Nanji town with a circle radius of 43.74 kilometers. Conclusions: The spatial clusters of TB incidence shifted to the rural region and the fringe of the new urban area of Nanchang. Targeted management strategies for urban migrants in the process of urbanization should be strengthened.

Mesothelioma mortality in Great Britain: how much longer will dockyards dominate?

ObjectivesWe aimed to investigate whether there has been a geographic shift in the distribution of mesothelioma deaths in Great Britain given the decline of shipbuilding and progressive exposure regulation.MethodsWe calculated...

Epidemiological Characteristics and Spatiotemporal Analysis of Mumps from 2004 to 2018 in Chongqing, China.

Mumps vaccines have been widely used in recent years, but frequent mumps outbreaks and re-emergence around the world have not stopped. Mumps still remains a serious public health problem with a high incidence in China. The status of mumps epidemics in Chongqing, the largest city in China, is still unclear. This study aimed to investigate the epidemiological and spatiotemporal characteristics of mumps and to provide a scientific basis for formulating effective strategies for its prevention and control. Surveillance data of mumps in Chongqing from January 2004 to December 2018 were collected from the National Notifiable Diseases Reporting Information System. A descriptive analysis was conducted to understand the epidemiological characteristics. Hot spots and spatiotemporal patterns were identified by performing a spatial autocorrelation analysis, a purely spatial scan, and a spatiotemporal scan at the county level based on geographic information systems. A total of 895,429 mumps cases were reported in Chongqing, with an annual average incidence of 36.34 per 100,000. The yearly incidence of mumps decreased markedly from 2004 to 2007, increased sharply from 2007 to 2011, and then tapered with a two-year cyclical peak after 2011. The onset of mumps showed an obvious bimodal seasonal distribution, with a higher peak of mumps observed from April to July of each year. Children aged 5–9 years old, males, and students were the prime high-risk groups. The spatial distribution of mumps did not exhibit significant global autocorrelation in most years, but local indicators of spatial autocorrelation and scan statistics detected high-incidence clusters which were mainly located in the midwestern, western, northeastern, and southwestern parts of Chongqing. The aggregation time frame detected by the purely temporal scan was between March 2009 and July 2013. The incidence of mumps in Chongqing from 2004 to 2018 featured significant spatial heterogeneity and spatiotemporal clustering. The findings of this study might assist public health agencies to develop real-time space monitoring, especially in the clustering regions and at peak periods; to improve immunization strategies for long-term prevention; and to deploy health resources reasonably.

An epidemiological and spatiotemporal analysis to identify high risk areas of malaria in Visakhapatnam district of Andhra Pradesh, India, 1999–2015

Malaria is a major public health problem in Vishakhapatnam district of Andhra Pradesh, India. To understand malaria prevalence a retrospective surveillance study was conducted in the district from 1995 to 2015. A total of 204,229 malaria cases were reported from 1999 to 2015. Plasmodium falciparum and Plasmodium vivax are the major parasites that accounted for 66.8% and 33.2% of the total cases. Tribal population (67%) affected more than the coastal population (33%). Similarly, males were affected (56%) more than female (44%) populace and the highest prevalence was observed in > 15 years age group (83.74%). The spatial analysis reveals that the distribution of malaria is having high spatial autocorrelation (0.231 to 0.493) and scan statistics declare that the malaria cases were significantly clustered in spatial, temporal and spatiotemporal distribution. The most likely spatiotemporal cluster of malaria (LLR = 26,562.24, RR = 6.62, P < 0.001) occurred in the Northern part of the district covering 11 mandals with the time frame from April 2010 to September 2015. The results confirm that the presence of spatial and space–time clusters concentrated in the North and North-eastern region of the district, which contribute for better understanding of disease spreading dynamics in high-risk areas for future malaria prevention and control.

Spatiotemporal patterns and ecological factors of tuberculosis notification: A spatial panel data analysis in Guangxi, China.

BackgroundGuangxi is one of the provinces having the highest notification rate of tuberculosis in China. However, spatial and temporal patterns and the association between environmental diversity and tuberculosis notification are still unclear.ObjectiveTo detect the spatiotemporal pattern of tuberculosis notification rates from 2010 to 2016 and its potential association with ecological environmental factors in Guangxi Zhuang autonomous region, China.MethodsWe performed a spatiotemporal analysis with prediction using time series analysis, Moran’s I global and local spatial autocorrelation statistics, and space-time scan statistics to detect temporal and spatial clusters of tuberculosis notifications in Guangxi between 2010 and 2016. Spatial panel models were employed to identify potential associating factors.ResultsThe number of reported cases peaked in spring and summer and decreased in autumn and winter. The predicted number of reported cases was 49,946 in 2017. Moran's I global statistics were greater than 0 (0.363–0.536) during the study period. The most significant hot spots were mainly located in the central area. The eastern area exhibited a low-low relation. By the space-time scanning, the clusters identified were similar to those of the local autocorrelation statistics, and were clustered toward the early part of 2016. Duration of sunshine, per capita gross domestic product, the treatment success rate of tuberculosis and participation rate of the new cooperative medical care insurance scheme in rural areas had a significant negative association with tuberculosis notification rates.ConclusionThe notification rate of tuberculosis in Guangxi remains high, with the highest notification cluster located in the central region. The notification rate is associated with economic level, treatment success rate and participation in the new cooperative medical care insurance scheme.

Households or Hotspots? Defining Intervention Targets for Malaria Elimination in Ratanakiri Province, Eastern Cambodia.

BackgroundMalaria “hotspots” have been proposed as potential intervention units for targeted malaria elimination. Little is known about hotspot formation and stability in settings outside sub-Saharan Africa.MethodsClustering of Plasmodium infections at the household and hotspot level was assessed over 2 years in 3 villages in eastern Cambodia. Social and spatial autocorrelation statistics were calculated to assess clustering of malaria risk, and logistic regression was used to assess the effect of living in a malaria hotspot compared to living in a malaria-positive household in the first year of the study on risk of malaria infection in the second year.ResultsThe crude prevalence of Plasmodium infection was 8.4% in 2016 and 3.6% in 2017. Living in a hotspot in 2016 did not predict Plasmodium risk at the individual or household level in 2017 overall, but living in a Plasmodium-positive household in 2016 strongly predicted living in a Plasmodium-positive household in 2017 (Risk Ratio, 5.00 [95% confidence interval, 2.09–11.96], P < .0001). There was no consistent evidence that malaria risk clustered in groups of socially connected individuals from different households.ConclusionsMalaria risk clustered more clearly in households than in hotspots over 2 years. Household-based strategies should be prioritized in malaria elimination programs in this region.

Spatial autocorrelation for massive spatial data: verification of efficiency and statistical power asymptotics

Being a hot topic in recent years, many studies have been conducted with spatial data containing massive numbers of observations. Because initial developments for classical spatial autocorrelation statistics are based on rather small sample sizes, in the context of massive spatial datasets, this paper presents extensions to efficiency and statistical power comparisons between the Moran coefficient and the Geary ratio for different variable distribution assumptions and selected geographic neighborhood definitions. The question addressed asks whether or not earlier results for small n extend to large and massively large n, especially for non-normal variables; implications established are relevant to big spatial data. To achieve these comparisons, this paper summarizes proofs of limiting variances, also called asymptotic variances, to do the efficiency analysis, and derives the relationship function between the two statistics to compare their statistical power at the same scale. Visualization of this statistical power analysis employs an alternative technique that already appears in the literature, furnishing additional understanding and clarity about these spatial autocorrelation statistics. Results include: the Moran coefficient is more efficient than the Geary ratio for most surface partitionings, because this index has a relatively smaller asymptotic as well as exact variance, and the superior power of the Moran coefficient vis-a-vis the Geary ratio for positive spatial autocorrelation depends upon the type of geographic configuration, with this power approaching one as sample sizes become increasingly large. Because spatial analysts usually calculate these two statistics for interval/ration data, this paper also includes comments about the join count statistics used for nominal data.

Analysis on spatial-temporal distribution characteristics of smear positive pulmonary tuberculosis in China, 2004–2015

BackgroundIn China, tuberculosis (TB) is still a major infectious disease threatening people’s health. Smear positive pulmonary TB is one of the most common infectious forms of TB and it might easily cause the outbreak in some areas. With a better understanding of the spatial-temporal variations of smear positive PTB, we would reach the targets for TB prevention and controlling, identify high-risk areas and periods. Thus, the aim of this study was to investigate the spatial-temporal variations of smear positive PTB. MethodsProvincial level data of reported smear positive PTB monthly cases and incidence from January 2004 to December 2015 were obtained from the National Scientific Data Sharing Platform for Population and Health of China. Purely spatial-temporal descriptive analysis was used to characterize the distribution patterns of smear positive PTB. The global spatial auto-correlation statistics (Moran’s I) and the local indicators of spatial association (LISA) were conducted to identify the spatial auto-correlation and high risk areas of smear positive PTB cases. Furthermore, the space-time scan statistic was adopted to detect the spatial-temporal clusters in different periods. ResultsA total of 4,711,571 smear positive PTB cases were notified in China with an average annual incidence of 29.59/100,000. The proportion of male in different age groups were obviously higher than that of women. The largest number of cases was reported in the 20–24 years age group. Time-series analysis indicated that monthly incidence appeared a clearly seasonality and periodicity, which the seasonal peaks occurred in January and March. Smear positive PTB cases had a positive global spatial auto-correlation in 2013–2015 (Moran‘s I=0.186, P=0.046). Spatial clusters were identified in four periods, located in the different regions. The time period of 2004–2006, the most likely spatial-temporal cluster (RR=1.69, P<0.001) was mainly located in Hubei, Hunan, Jiangxi and Anhui of central China, clustering in the time frame from January 2005 to June 2006. During 2007–2009, the most likely spatial-temporal cluster (RR=5.65, P<0.001) was located in Guizhou, clustering in the time frame from January to December 2009. The spatial-temporal clustering in the years 2010–2012 showed the most likely cluster (RR=1.44, P<0.001) was distributed in Anhui, Hunan, Hubei, Jiangxi and Guangdong with the time frame from January 2010 to June 2011. During 2013–2015, the most likely cluster (RR=1.86, P<0.001) was detected in Hunan, Hubei, Jiangxi and Guangdong from February 2013 to June 2014. ConclusionsThis study identified the spatial-temporal patterns of smear positive PTB in China and demonstrated the capability and utility of the spatial-temporal approach in epidemiology. The results of this study would contribute to estimating the high risk periods and areas, and to providing more useful information for policy-making.

Upscaling ecosystem service maps to administrative levels: beyond scale mismatches

As Ecosystem Services (ES) are the products of complex socio–ecological systems, their mapping requires a deep understanding of the spatial relationships and pattern that underpin ES provision. Upscaling ES maps is often carried out to avoid mismatches between the scale of ES assessment and that of their level of management. However, so far only a few efforts have been made to quantify how information loss occurs as data are aggregated to coarser scales. In the present study this was analyzed for three distinct case studies in the eastern Alps by comparing ES maps of outdoor recreation at the municipality level and at finer scales, i.e. high-resolution grids. Specifically, we adopt an innovative and flexible methodology based on Exploratory Spatial Data Analysis (ESDA), to disentangle the problem of the scale from the perspective of different levels of jurisdiction, by assessing in an iterative process how ES patterns change when upscaling high-resolution maps. Furthermore, we assess the sensitivity to the modifiable areal unit problem (MAUP) by calculating global statistics over three grid displacements. Our results demonstrate that spatial clusters tend to disappear when their extent becomes smaller than the features to which values are upscaled, leading to substantial information loss. Moreover, cross-comparison among grids and the municipality level highlights local anomalies that global spatial autocorrelation indicators fail to detect, revealing hidden clusters and inconsistencies among multiple scales. We conclude that, whenever ES maps are aggregated to a coarser scale, our methodology represents a suitable and flexible approach to explore clustering trends, shape and position of upscaling units, through graphs and maps showing spatial autocorrelation statistics. This can be crucial to finding the best compromise among scale mismatches, information loss and statistical bias that can directly affect the targeted ES mapping.

Spatial autocorrelation and data uncertainty in the American Community Survey: a critique

ABSTRACTWe argue that the use of American Community Survey (ACS) data in spatial autocorrelation statistics without considering error margins is critically problematic. Public health and geographical research has been slow to recognize high data uncertainty of ACS estimates, even though ACS data are widely accepted data sources in neighborhood health studies and health policies. Detecting spatial autocorrelation patterns of health indicators on ACS data can be distorted to the point that scholars may have difficulty in perceiving the true pattern. We examine the statistical properties of spatial autocorrelation statistics of areal incidence rates based on ACS data. In a case study of teen birth rates in Mecklenburg County, North Carolina, in 2010, Global and Local Moran’s I statistics estimated on 5-year ACS estimates (2006–2010) are compared to ground truth rate estimates on actual counts of births certificate records and decennial-census data (2010). Detected spatial autocorrelation patterns are found to be significantly different between the two data sources so that actual spatial structures are misrepresented. We warn of the possibility of misjudgment of the reality and of policy failure and argue for new spatially explicit methods that mitigate the biasedness of statistical estimations imposed by the uncertainty of ACS data.

Analysis of Lipid Order States and Domains in Lipid Bilayer Simulations.

We propose a general procedure to analyze lipid order states and domains in lipid bilayer simulations using surface areas and hydrophobic thicknesses of lipids. In our approach, the observable order states of individual lipids are inferred by a hidden Markov model analysis of their time series and by considering the deformation of a lipid in different packing environments. The assigned lipid order states are mapped onto the Voronoi tessellation of lipids, from which the ordered and disordered lipids are robustly clustered by the Getis-Ord local spatial autocorrelation statistics. The usefulness of this method is illustrated by its application to the quinary mixed bilayers consisting of cholesterol (Chol), 1,2-dimyristoyl- sn-glycero-3-phosphocholine (DMPC), 1,2-dimyristoyl- sn-glycero-3-phosphoethanolamine (DMPE), 1-palmitoyl-2-oleoyl- sn-glycero-3-phosphocholine (POPC), and 1-palmitoyl-2-oleoyl- sn-glycero-3-phosphoethanolamine (POPE), where any phospholipid type does not show strong preference over the other types to be enriched in lipid domains. The independent order state analysis for each lipid type allows straightforward applications of our method to arbitrarily complex bilayer simulations.

TERRITORIAL COHESION OF COMMUNES IN POLAND IN 2005-2015 WITH REFERENCE TO TECHNICAL INFRASTRUCTURE DEVELOPMENT IN TERMS OF SPATIAL AUTOCORRELATION MEASURES

Territorial cohesion refers to integrated space management, that mitigates and prevents its polarization. The aim of the analysis was to delimit the phenomenon of territorial cohesion in terms of the development of technical infrastructure using spatial autocorrelation statistics to identify homogeneous groups – clusters of municipalities with similar characteristics and hot spots – regions with outliers in relation to the environment. The survey covered the entire population of communes in Poland (2,478), in 2005, 2010 and 2015. The results indicate the existence of positive and increasing spatial autocorrelation for the aggregate measure of technical infrastructure development in subsequent years. For the size of clusters of local Moran statistics, the differentiating features were the type and population type of communes and the spatial location in voivodship. Territorial cohesion, ie compliance in relation to the type of homogenous group has been identified in a small part of the surveyed local government units.

Spatial Autocorrelation Statistics of Areal Prevalence Rates under High Uncertainty in Denominator Data

We propose a new estimator of spatial autocorrelation of areal incidence or prevalence rates in small areas, such as crime and health indicators, for correcting spatially heterogeneous sampling errors in denominator data. The approach is dubbed the heteroscedasticity‐consistent empirical Bayes (HC‐EB) method. As American Community Survey (ACS) data have been released to the public for small census geographies, small‐area estimates now form the demographic landscape of neighborhoods. Meanwhile, there is growing awareness of the diminished statistical validity of global and local Moran’s I when such small‐area estimates are used in denominator data. Using teen birth rates by census tracts in Mecklenburg County, North Carolina, we present comparisons of conventional and new HC‐EB estimates of Global and Local Moran’s I statistics created on ACS data, along with estimates on ground truth values from the 2010 decennial census. Results show that the new adjustment method dramatically enhances the statistical validity of global and local spatial autocorrelation statistics.

BUILT-UP AREA DETECTION FROM HIGH-RESOLUTION SATELLITE IMAGES USING MULTI-SCALE WAVELET TRANSFORM AND LOCAL SPATIAL STATISTICS

Abstract. Recently, built-up area detection from high-resolution satellite images (HRSI) has attracted increasing attention because HRSI can provide more detailed object information. In this paper, multi-resolution wavelet transform and local spatial autocorrelation statistic are introduced to model the spatial patterns of built-up areas. First, the input image is decomposed into high- and low-frequency subbands by wavelet transform at three levels. Then the high-frequency detail information in three directions (horizontal, vertical and diagonal) are extracted followed by a maximization operation to integrate the information in all directions. Afterward, a cross-scale operation is implemented to fuse different levels of information. Finally, local spatial autocorrelation statistic is introduced to enhance the saliency of built-up features and an adaptive threshold algorithm is used to achieve the detection of built-up areas. Experiments are conducted on ZY-3 and Quickbird panchromatic satellite images, and the results show that the proposed method is very effective for built-up area detection.

Effects of landscape fragmentation on land loss

Coastal Louisiana has been facing a serious land loss problem over the past several decades, and extensive research has been undertaken to address the problem. However, the importance of landscape fragmentation on land loss has seldom been examined. This paper evaluates the effects of landscape fragmentation on land loss in the Lower Mississippi River Basin region. The research hypothesis is that the higher the degree of fragmentation in a locality, the greater the amount of land loss in the next time period. We used Landsat-TM data with a pixel size of 30m×30m in 1996 and 2010 and transformed the images into either land or water pixels. We then calculated the fractal dimension and Moran's I spatial autocorrelation statistics and used them to represent the degree of landscape fragmentation. Four sample box sizes, including sizes of 101×101, 71×71, 51×51, and 31×31pixels, were used to detect if there is a relationship between fragmentation and land loss at different neighborhood (context) scales. For each box size, 100 samples were randomly selected. To isolate the fragmentation effect so that it can be better evaluated, we used only sample boxes with a 50% land-water ratio. Regression results between fragmentation and land loss show that the R2 values for box sizes of 71×71, 51×51 and 31×31 were statistically significant (0.20, 0.45, 0.35; p<0.001 for Moran's I) but not for the 101×101 box size. These results imply that land protection may be most effective by prioritizing areas with land patches that have the least fragmentation. Furthermore, the neighborhood scale at which the R2 value is the highest indicates the scale at which the effects are most likely to be observed (51×51 box size, approximately 1.5×1.5km2, R2=0.45), which suggests that future land loss modeling using this neighborhood scale would be most effective.