Tonglin Zhang

Agriculture intensifies soil moisture decline in Northern China

Northern China is one of the most densely populated regions in the world. Agricultural activities have intensified since the 1980s to provide food security to the country. However, this intensification has likely contributed to an increasing scarcity in water resources, which may in turn be endangering food security. Based on in-situ measurements of soil moisture collected in agricultural plots during 1983–2012, we find that topsoil (0–50 cm) volumetric water content during the growing season has declined significantly (p < 0.01), with a trend of −0.011 to −0.015 m3 m−3 per decade. Observed discharge declines for the three large river basins are consistent with the effects of agricultural intensification, although other factors (e.g. dam constructions) likely have contributed to these trends. Practices like fertilizer application have favoured biomass growth and increased transpiration rates, thus reducing available soil water. In addition, the rapid proliferation of water-expensive crops (e.g., maize) and the expansion of the area dedicated to food production have also contributed to soil drying. Adoption of alternative agricultural practices that can meet the immediate food demand without compromising future water resources seem critical for the sustainability of the food production system.

Spatial scan statistics in loglinear models

The likelihood ratio spatial scan statistic has been widely used in spatial disease surveillance and spatial cluster detection applications. In order to better understand cluster mechanisms, an equivalent model-based approach is proposed to the spatial scan statistic that unifies currently loosely coupled methods for including ecological covariates in the spatial scan test. In addition, the utility of the model-based approach with a Wald-based scan statistic is demonstrated to account for overdispersion and heterogeneity in background rates. Simulation and case studies show that both the likelihood ratio-based and Wald-based scan statistics are comparable with the original spatial scan statistic.

Spatial scan statistics with overdispersion

The spatial scan statistic has been widely used in spatial disease surveillance and spatial cluster detection for more than a decade. However, overdispersion often presents in real-world data, causing not only violation of the Poisson assumption but also excessive type I errors or false alarms. In order to account for overdispersion, we extend the Poisson-based spatial scan test to a quasi-Poisson-based test. The simulation shows that the proposed method can substantially reduce type I error probabilities in the presence of overdispersion. In a case study of infant mortality in Jiangxi, China, both tests detect a cluster; however, a secondary cluster is identified by only the Poisson-based test. It is recommended that a cluster detected by the Poisson-based scan test should be interpreted with caution when it is not confirmed by the quasi-Poisson-based test.

Cluster Detection Based on Spatial Associations and Iterated Residuals in Generalized Linear Mixed Models

SUMMARY Spatial clustering is commonly modeled by a Bayesian method under the framework of generalized linear mixed effect models (GLMMs). Spatial clusters are commonly detected by a frequentist method through hypothesis testing. In this article, we provide a frequentist method for assessing spatial properties of GLMMs. We propose a strategy that detects spatial clusters through parameter estimates of spatial associations, and assesses spatial aspects of model improvement through iterated residuals. Simulations and a case study show that the proposed method is able to consistently and efficiently detect the locations and magnitudes of spatial clusters.

Credible and confidence sets for restricted parameter spaces

Abstract Recent experiments in high-energy physics have sparked interest in problems where a parameter is restricted to a portion of its natural range—for example, a positive normal mean or a Poisson mean that is known to exceed some background level. A recent article has shown that the Bayesian credible intervals that arise when the parameter is given a uniform distribution over the restricted range have good frequentist coverage probabilities. Here, the latter result is examined for robustness when nuisance parameters are included in the model. It is shown that the frequentist coverage probabilities are still high, but there is a mild surprise in the nature of the intervals.

Big Data Dimension Reduction Using PCA

Principal component analysis (PCA) is a powerful tool in dimensional reduction for highly correlated data. Classical PCA approaches cannot be applied to big data because ofmemory and storage barriers. To solve the problem, the article proposes a new approach. The basic idea is to derive an array of sufficient statistics by scanning data by rows. It shows that the proposed approach can provide exact solutions if the linear regression approach is used in the follow up analysis.

Identification of local clusters for count data: a model-based Moran's I test

We set out I DR as a loglinear-model-based Morans I test for Poisson count data that resembles the Morans I residual test for Gaussian data. We evaluate its type I and type II error probabilities via simulations, and demonstrate its utility via a case study. When population sizes are heterogeneous, I DR is effective in detecting local clusters by local association terms with an acceptable type I error probability. When used in conjunction with local spatial association terms in loglinear models, I DR can also indicate the existence of first-order global cluster that can hardly be removed by local spatial association terms. In this situation, I DR should not be directly applied for local cluster detection. In the case study of St. Louis homicides, we bridge loglinear model methods for parameter estimation to exploratory data analysis, so that a uniform association term can be defined with spatially varied contributions among spatial neighbors. The method makes use of exploratory tools such as Morans I scatter plots and residual plots to evaluate the magnitude of deviance residuals, and it is effective to model the shape, the elevation and the magnitude of a local cluster in the model-based test.

On Moran's I coefficient under heterogeneity

Morans I is the most popular spatial test statistic, but its inability to incorporate heterogeneous populations has been long recognized. This article provides a limiting distribution of the Morans I coefficient which can be applied to heterogeneous populations. The method provides a unified framework of testing for spatial autocorrelation for both homogeneous and heterogeneous populations, thereby resolving a long standing issue for Morans I . For Poisson count data, a variance adjustment method is provided that solely depends on populations at risk. Simulation results are shown to be consistent with theoretical results. The application of Nebraska breast cancer data shows that the variance adjustment method is simple and effective in reducing type I error rates, which in turn will likely reduce potential misallocation of limited resources.

Cyberinfrastructure for isotope analysis and modeling

As the quantity and complexity of scientific data expand, accessible interfaces for data manipulation and analysis are needed to support broad and efficient data use. The Isoscapes Modeling, Analysis, and Prediction (IsoMAP; http://isomap.org) Web-based geographical information system (GIS) gateway is an example of such a resource. Recently launched with support from the U.S. National Science Foundation (NSF) Division of Biological Infrastructure, IsoMAP enables analysis and integration of diverse light stable isotope and environmental data by a broad-based user community. It provides an intuitive, spatial interface that streamlines data analysis, modeling, and exploration in research ranging from greenhouse gas biogeochemistry to food science.

Detection and localization of hidden radioactive sources with spatial statistical method

The detection of radioactive materials has become a critical issue for environmental services, public health, and national security. This paper proposes a spatial statistical method to detect and localize a hidden radioactive source. Based on a detection system of multiple radiation detectors, the statistical model assumes that the counts of radiation particles received by those detectors are spatially distributed of Poisson distribution, and each comprises a signal and a background. By considering the physical law of signal degradation with distance, the paper provides a numerical method to compute the maximum likelihood estimates of the strength and location of the source. Based on these estimates, a likelihood ratio statistic is used to test the existence of the source. Because of the special properties of the model, the test statistic does not converge asymptotically to the standard chi-square distribution. Thus a bootstrap method is proposed to compute the p-value in the test. The simulation results show that the proposed method is efficient for detecting and localizing the hidden radioactive source.