Inversion Method for Chlorophyll-a Concentration in High-Salinity Water Based on Hyperspectral Remote Sensing Data.

As one of the important lakes in the "One Lake and Two Seas" of the Inner Mongolia Autonomous Region, the monitoring of water quality in Lake Daihai has attracted increasing attention, and the concentration of chlorophyll-a directly affects the water quality, making the monitoring of chlorophyll-a concentration in Lake Daihai particularly crucial. Traditional methods of monitoring chlorophyll-a concentration are not only inefficient but also require significant human and material resources. Remote sensing technology has the advantages of wide coverage and short update cycles. For lakes such as Daihai with a high salinity content, salinity is considered a key factor when inverting the concentration of chlorophyll-a. In this study, machine learning models, including model stacking from ensemble learning, a ridge regression model, and a random forest model, were constructed. After comparing the training accuracy of the three models on Zhuhai-1 satellite data, the random forest model, which had the highest accuracy, was selected as the final training model. By comparing the accuracy changes before and after adding salinity factors to the random forest model, a high-precision model for inverting chlorophyll-a concentration in hypersaline lakes was obtained. The research results show that, without considering the salinity factor, the root mean square error (RMSE) of the model was 0.056, and the coefficient of determination (R2) was 0.64, indicating moderate model performance. After adding the salinity factor, the model accuracy significantly improved: the RMSE decreased to 0.047, and the R2 increased to 0.92. This study provides a solid basis for the application of remote sensing technology in hypersaline aquatic environments, confirming the importance of considering salinity when estimating chlorophyll-a concentration in hypersaline waters. This research helps us gain a deeper understanding of the water quality and ecosystem evolution in Daihai Lake.

Review of Methods for Automatic Plastic Detection in Water Areas Using Satellite Images and Machine Learning.

Ocean plastic pollution is one of the global environmental problems of our time. "Rubbish islands" formed in the ocean are increasing every year, damaging the marine ecosystem. In order to effectively address this type of pollution, it is necessary to accurately and quickly identify the sources of plastic entering the ocean, identify where it is accumulating, and track the dynamics of waste movement. To this end, remote sensing methods using satellite imagery and aerial photographs from unmanned aerial vehicles are a reliable source of data. Modern machine learning technologies make it possible to automate the detection of floating plastics. This review presents the main projects and research aimed at solving the "plastic" problem. The main data acquisition techniques and the most effective deep learning algorithms are described, various limitations of working with space images are analyzed, and ways to eliminate such shortcomings are proposed.

Impact of Covid-19 lockdown regulations on PM2.5 and trace gases (NO2, SO2, CH4, HCHO, C2H2O2 and O3) over Lahore, Pakistan.

The COVID-19 pandemic altered the human mobility and economic activities immensely, as authorities enforced unprecedented lock down regulations. In order to reduce the spread of COVID-19, a complete lockdown was observed between 24 March - 31 May 2020 in Pakistan. This paper aims at investigating the PM2.5, AOD and column amounts of six trace gases (NO2, SO2, CH4, HCHO, C2H2O2, and O3) by comparing periods of reduced emissions during lockdown periods with reference periods without emission reductions over Lahore, Pakistan. HYSPLIT cluster trajectory analyses were performed, which confirmed similar meteorological flow conditions during lockdown and reference periods. This provides confidence that any change in air quality conditions would be due to changes in human activities and associated emissions. The results show about 38% reduction in ambient surface PM2.5 levels during the lockdown period. This change also positively correlated with MODISDB and AERONETAOD data with a decrease of AOD by 42% and 35%, respectively. Reductions for tropospheric columns of NO2 and SO2 were about 20% and 50%, respectively during a semi lockdown period, while no reduction in the CH4, C2H2O2, HCHO and O3 levels occurred. During the lockdown period NO2, O3 and CH4 were about 50%, 45% and 25% lower, respectively, but no reduction in SO2, C2H2O2 and HCHO levels were noticed compared to the reference lockdown period for Lahore. HYSPLIT cluster trajectory analysis revealed the greatest impact on Lahore air quality through local emissions and regional transport from the east (agricultural burning and industry).

Identification of factors controlling heavy metals/metalloid distribution in agricultural soils using multi-source data.

Understanding the factors that controlling the agricultural soil heavy metals/metalloids distribution is vital for cropland soil remediation and management. For this objective, 227 agricultural soils were sampled in the Guanzhong Plain, China, to measure the concentration of five heavy metals (Pb, Cd, Ni, Zn, and Cu) and one metalloid (As) by X-ray fluorescence spectrometer, meanwhile, 24 possible influencing factors to agricultural soil heavy metals/metalloid distribution were collected and grouped into three categories. A sequential multivariate statistical analysis was carried out to provide insight into the controlling factors of soil heavy metals/metalloid distribution, then stepwise multiple linear regression (SMLR) and partial least squares regression (PLS) were used to predict heavy metals/metalloid concentrations in agricultural soil based on the result of soil heavy metals/metalloid controlling factors identification. The results demonstrated the types of soil and land use did not have a substantial effect on soil heavy metals/metalloid distribution, except Zn and Cu. The soil properties category played a major role in influencing the soil heavy metals/metalloid concentration. The concentrations of Mn and Fe, which are the main constitute elements of soil inorganic colloid, were the most significant factors, followed by the concentrations of P, K and Ca. Soil pH and soil organic matter (SOM) content, which are often considered as important factors for soil heavy metals/metalloid distribution, were not important in the present study. The SMLR was more effective than the PLS for predicting soil heavy metals/metalloid content. The results of this study enlighten that future soil heavy metals/metalloid contamination treatment in regions with high pH and low SOM content should concentrate on inorganic colloid particles, which have strong adsorption capacity for soil heavy metals/metalloid and are environmentally friendly. Moreover, the combination of successive multivariate statistical analysis and SMLR provide an effective tool to predict and monitor agricultural soil heavy metals/metalloid distribution, and facilitate the improvement of environmental and territorial management.

Root-zone soil moisture estimation based on remote sensing data and deep learning.

Soil moisture in the root zone is the most important factor in eco-hydrological processes. Even though soil moisture can be obtained by remote sensing, limited to the top few centimeters (<5 cm). Researchers have attempted to estimate root-zone soil moisture using multiple regression, data assimilation, and data-driven methods. However, correlations between root-zone soil moisture and its related variables, including surface soil moisture, always appear nonlinear, which is difficult to extract and express using typical statistical methods. The artificial intelligence (AI) method, which is advantageous for nonlinear relationship analysis and extraction is applied for root-zone soil moisture estimation, but by only considering its separate temporal or spatial correlations. The convolutional long short-term memory (ConvLSTM) model, known to capture spatiotemporal patterns of large-scale sequential datasets with the advantage of dealing with spatiotemporal sequence-forecasting problem, was used in this study to estimate root-zone soil moisture based on remote sensing-based variables. Owing to limitation of regional soil moisture observation data, the physical model Hydrus-1D was used to generate large and spatiotemporal vertical soil moisture dataset for the ConvLSTM model training and verification. Then, normalized difference vegetation index (NDVI) etc. remote sensing-based factors were selected as predictive variables. Results of the ConvLSTM model showed that the fitting coefficients (R2) of the root-zone soil moisture estimation significantly increased compared to those achieved by Global Land Data Assimilation System products, especially for deep layers. For example, R2 increased from 0.02 to 0.60 at depth of 40 cm. This study suggests that a combination of the physical model and AI is a flexible tool capable of predicting spatiotemporally continuous root-zone soil moisture with good accuracy on a large scale.

Query Optimization for Distributed Spatio-Temporal Sensing Data Processing.

The unprecedented development of Internet of Things (IoT) technology produces humongous amounts of spatio-temporal sensing data with various geometry types. However, processing such datasets is often challenging due to high-dimensional sensor data geometry characteristics, complex anomalistic spatial regions, unique query patterns, and so on. Timely and efficient spatio-temporal querying significantly improves the accuracy and intelligence of processing sensing data. Most existing query algorithms show their lack of supporting spatio-temporal queries and irregular spatial areas. In this paper, we propose two spatio-temporal query optimization algorithms based on SpatialHadoop to improve the efficiency of query spatio-temporal sensing data: (1) spatio-temporal polygon range query (STPRQ), which aims to find all records from a polygonal location in a time interval; (2) spatio-temporal k nearest neighbors query (STkNNQ), which directly searches the query point's k closest neighbors. To optimize the STkNNQ algorithm, we further propose an adaptive iterative range optimization algorithm (AIRO), which can optimize the iterative range of the algorithm according to the query time range and avoid querying irrelevant data partitions. Finally, extensive experiments based on trajectory datasets demonstrate that our proposed query algorithms can significantly improve query performance over baseline algorithms and shorten response time by 81% and 35.6%, respectively.

Monitoring drought dynamics in China using Optimized Meteorological Drought Index (OMDI) based on remote sensing data sets.

Timely and accurate monitoring of the spatiotemporal changes in drought is very important for the reduction in the social losses caused by drought. The Optimized Meteorological Drought Index (OMDI), originally established in southwestern China, showed great potential for drought monitoring over large regions on a large scale. However, the applicability of the index requires further evaluation, especially when used throughout China, which has a different agricultural divisions, variable climatic conditions, complex terrain and diverse land cover. In addition, the OMDI model relies on training data to construct local parameters for the model. On a large scale, it is of great significance to use multisource remote sensing data sets to construct OMDI model parameters. In this paper, the constrained optimization method was used to establish weights for the MODIS-derived Vegetation Conditional Index (VCI), TRMM-derived Precipitation Condition Index (PCI), and GLDAS-derived Soil Moisture Condition Index (SMCI) and calculate the OMDI based on the Standard Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI) and weather stations. The accuracy of the OMDI model was evaluated by using the correlation coefficient. Moreover, the spatiotemporal changes in drought were also analyzed through trend analysis, Mann-Kendall (MK) statistics and the Hurst index on the monthly and annual scales. The results showed that (1) the highest positive correlation between the OMDI and the SPI was SPI-1, which was higher than that for any other month interval, such as 3 months, 6 months, 9 months and 12 months of the SPI. The results indicated that the OMDI was suitable to monitor meteorological drought. (2) In the nine agricultural subareas in China, the degree of drought in the Yangtze River (DYR) area had the most severe evolution and change frequency. This region was very sensitive to drought in the past two decades. (3) The area with OMDI variation coefficient less than 0.1 accounted for 94%, indicating that the degree of drought fluctuates little; The linear tendency rate is 0.0004, and the area greater than 0 reaches 66.44%, indicating that the drought is developing in a lightning trend. (4) The Hurst index value is mostly higher than 0.5 (the area ratio is 56.31%), and the area of "Positive-Consistent" and "Negative- Opposite" accounted for 54.02%, indicating that more than half of China's area drought changes will show a trend of mitigation in the future.

Hybrid Deep Learning Predictor for Smart Agriculture Sensing Based on Empirical Mode Decomposition and Gated Recurrent Unit Group Model.

Smart agricultural sensing has enabled great advantages in practical applications recently, making it one of the most important and valuable systems. For outdoor plantation farms, the prediction of climate data, such as temperature, wind speed, and humidity, enables the planning and control of agricultural production to improve the yield and quality of crops. However, it is not easy to accurately predict climate trends because the sensing data are complex, nonlinear, and contain multiple components. This study proposes a hybrid deep learning predictor, in which an empirical mode decomposition (EMD) method is used to decompose the climate data into fixed component groups with different frequency characteristics, then a gated recurrent unit (GRU) network is trained for each group as the sub-predictor, and finally the results from the GRU are added to obtain the prediction result. Experiments based on climate data from an agricultural Internet of Things (IoT) system verify the development of the proposed model. The prediction results show that the proposed predictor can obtain more accurate predictions of temperature, wind speed, and humidity data to meet the needs of precision agricultural production.

Surface Temperature Simulation of Lunar Dayside and Its Geological Applications: A Case in Sinus Iridum.

Lunar surface temperature is one of the fundamental thermophysical parameters of the lunar regolith, which is of great significance to the interpretation of remote-sensing thermal data. In this study, a daytime surface temperature model is established focusing on the lunar superficial layer with high spatial-temporal resolution. The physical parameters at the time of interest are adopted, including effective solar irradiance, lunar libration, large-scale topographic shading, and surrounding diffuse reflection. Thereafter, the 1/64° temperature distributions at five local times are quantitatively generated and analyzed in Sinus Iridum. Also, combined with Chang'E-2 microwave radiometer (CELMS) data and Diviner thermal infrared (TIR) data, the spectral emissivity distributions are estimated as a potential geological application of the simulated surface temperature. The results are as follows: (1) daytime surface temperature in Sinus Iridum is significantly affected by the local topography and observation time, and the influence of diffuse reflection energy is obvious; (2) the emissivity distributions provide a new way to understand the thermophysical properties difference of lunar regolith at different depths; (3) the influence of lunar orbiting revolution and precession on surface temperature should be analyzed carefully, which shows the importance of using the parameters at the time of interest.

An Approach to Dynamic Sensing Data Fusion.

For the research and development of sensor systems, the collection and fusion of sensing data is the core. In order to make sensor data acquisition change with the change in environment, a dynamic data acquisition and fusion method based on feedback control is proposed in this paper. According to the sensing data acquisition and fusion model, the optimal acquisition of sensor data is achieved through real-time dynamic judgment of the collected data, decision-making of the next acquisition time interval, and adjustment. This model enables the sensor system to adapt to different environments. An experimental study of the proposed model was carried out on an experimental platform, and the results show that the proposed model can not only reflect the change in sensing data but also improve the transmission efficiency.

Design and Implementation of an Integrated IoT Blockchain Platform for Sensing Data Integrity.

With the rapid development of communication technologies, the Internet of Things (IoT) is getting out of its infancy, into full maturity, and tends to be developed in an explosively rapid way, with more and more data transmitted and processed. As a result, the ability to manage devices deployed worldwide has been given more and advanced requirements in practical application performances. Most existing IoT platforms are highly centralized architectures, which suffer from various technical limitations, such as a cyber-attack and single point of failure. A new solution direction is essential to enhance data accessing, while regulating it with government mandates in privacy and security. In this paper, we propose an integrated IoT platform using blockchain technology to guarantee sensing data integrity. The aim of this platform is to afford the device owner a practical application that provides a comprehensive, immutable log and allows easy access to their devices deployed in different domains. It also provides characteristics of general IoT systems, allows for real-time monitoring, and control between the end user and device. The business logic of the application is defined by the smart contract, which contains rules and conditions. The proposed approach is backed by a proof of concept implementation in realistic IoT scenarios, utilizing Raspberry Pi devices and a permissioned network called Hyperledger Fabric. Lastly, a benchmark study using various performance metrics is made to highlight the significance of the proposed work. The analysis results indicate that the designed platform is suitable for the resource-constrained IoT architecture and is scalable to be extended in various IoT scenarios.

Research on Mechanical Fault Prediction Method Based on Multifeature Fusion of Vibration Sensing Data.

Vibration sensing data is an important resource for mechanical fault prediction, which is widely used in the industrial sector. Artificial neural networks (ANNs) are important tools for classifying vibration sensing data. However, their basic structures and hyperparameters must be manually adjusted, which results in the prediction accuracy easily falling into the local optimum. For data with high levels of uncertainty, it is difficult for an ANN to obtain correct prediction results. Therefore, we propose a multifeature fusion model based on Dempster-Shafer evidence theory combined with a particle swarm optimization algorithm and artificial neural network (PSO-ANN). The model first used the particle swarm optimization algorithm to optimize the structure and hyperparameters of the ANN, thereby improving its prediction accuracy. Then, the prediction error data of the multifeature fusion using a PSO-ANN is repredicted using multiple PSO-ANNs with different single feature training to obtain new prediction results. Finally, the Dempster-Shafer evidence theory was applied to the decision-level fusion of the new prediction results preprocessed with prediction accuracy and belief entropy, thus improving the model's ability to process uncertain data. The experimental results indicated that compared to the K-nearest neighbor method, support vector machine, and long short-term memory neural networks, the proposed model can effectively improve the accuracy of fault prediction.

Performance Sensing Data Prediction for an Aircraft Auxiliary Power Unit Using the Optimized Extreme Learning Machine.

The aircraft auxiliary power unit (APU) is responsible for environmental control in the cabin and the main engines starting the aircraft. The prediction of its performance sensing data is significant for condition-based maintenance. As a complex system, its performance sensing data have a typically nonlinear feature. In order to monitor this process, a model with strong nonlinear fitting ability needs to be formulated. A neural network has advantages of solving a nonlinear problem. Compared with the traditional back propagation neural network algorithm, an extreme learning machine (ELM) has features of a faster learning speed and better generalization performance. To enhance the training of the neural network with a back propagation algorithm, an ELM is employed to predict the performance sensing data of the APU in this study. However, the randomly generated weights and thresholds of the ELM often may result in unstable prediction results. To address this problem, a restricted Boltzmann machine (RBM) is utilized to optimize the ELM. In this way, a stable performance parameter prediction model of the APU can be obtained and better performance parameter prediction results can be achieved. The proposed method is evaluated by the real APU sensing data of China Southern Airlines Company Limited Shenyang Maintenance Base. Experimental results show that the optimized ELM with an RBM is more stable and can obtain more accurate prediction results.

Visual IoT Security: Data Hiding in AMBTC Images Using Block-Wise Embedding Strategy.

This study investigates combining the property of human vision system and a 2-phase data hiding strategy to improve the visual quality of data-embedded compressed images. The visual Internet of Things (IoT) is indispensable in smart cities, where different sources of visual data are collected for more efficient management. With the transmission through the public network, security issue becomes critical. Moreover, for the sake of increasing transmission efficiency, image compression is widely used. In order to respond to both needs, we present a novel data hiding scheme for image compression with Absolute Moment Block Truncation Coding (AMBTC). Embedding secure data in digital images has broad security uses, e.g., image authentication, prevention of forgery attacks, and intellectual property protection. The proposed method embeds data into an AMBTC block by two phases. In the intra-block embedding phase, a hidden function is proposed, where the five AMBTC parameters are extracted and manipulated to embed the secret data. In the inter-block embedding phase, the relevance of high mean and low mean values between adjacent blocks are exploited to embed additional secret data in a reversible way. Between these two embedding phases, a halftoning scheme called direct binary search is integrated to efficiently improve the image quality without changing the fixed parameters. The modulo operator is used for data extraction. The advantages of this study contain two aspects. First, data hiding is an essential area of research for increasing the IoT security. Second, hiding in compressed images instead of original images can improve the network transmission efficiency. The experimental results demonstrate the effectiveness and superiority of the proposed method.

Improvement of Kafka Streaming Using Partition and Multi-Threading in Big Data Environment.

The large amount of programmable logic controller (PLC) sensing data has rapidly increased in the manufacturing environment. Therefore, a large data store is necessary for Big Data platforms. In this paper, we propose a Hadoop ecosystem for the support of many features in the manufacturing industry. In this ecosystem, Apache Hadoop and HBase are used as Big Data storage and handle large scale data. In addition, Apache Kafka is used as a data streaming pipeline which contains many configurations and properties that are used to make a better-designed environment and a reliable system, such as Kafka offset and partition, which is used for program scaling purposes. Moreover, Apache Spark closely works with Kafka consumers to create a real-time processing and analysis of the data. Meanwhile, data security is applied in the data transmission phase between the Kafka producers and consumers. Public-key cryptography is performed as a security method which contains public and private keys. Additionally, the public-key is located in the Kafka producer, and the private-key is stored in the Kafka consumer. The integration of these above technologies will enhance the performance and accuracy of data storing, processing, and securing in the manufacturing environment.

Describing temporal variation in reticuloruminal pH using continuous monitoring data.

Reticuloruminal pH has been linked to subclinical disease in dairy cattle, leading to considerable interest in identifying pH observations below a given threshold. The relatively recent availability of continuously monitored data from pH boluses gives new opportunities for characterizing the normal patterns of pH over time and distinguishing these from abnormal patterns using more sensitive and specific methods than simple thresholds. We fitted a series of statistical models to continuously monitored data from 93 animals on 13 farms to characterize normal variation within and between animals. We used a subset of the data to relate deviations from the normal pattern to the productivity of 24 dairy cows from a single herd. Our findings show substantial variation in pH characteristics between animals, although animals within the same farm tended to show more consistent patterns. There was strong evidence for a predictable diurnal variation in all animals, and up to 70% of the observed variation in pH could be explained using a simple statistical model. For the 24 animals with available production information, there was also a strong association between productivity (as measured by both milk yield and dry matter intake) and deviations from the expected diurnal pattern of pH 2 d before the productivity observation. In contrast, there was no association between productivity and the occurrence of observations below a threshold pH. We conclude that statistical models can be used to account for a substantial proportion of the observed variability in pH and that future work with continuously monitored pH data should focus on deviations from a predictable pattern rather than the frequency of observations below an arbitrary pH threshold.

A Study on the Design of Fog Computing Architecture Using Sensor Networks.

It is expected that the number of devices connecting to the Internet-of-Things (IoT) will increase geometrically in the future, with improvement of their functions. Such devices may create a huge amount of data to be processed in a limited time. Under the IoT environment, data management should play the role of an intermediate level between objects and devices that generate data and applications that access to the data for analysis and the provision of services. IoT interactively connects all communication devices and allows global access to the data generated by a device. Fog computing manages data and computation at the edge of the network near an end user and provides new types of applications and services, with low latency, high frequency bandwidth and geographical distribution. In this paper, we propose a fog computing architecture for efficiently and reliably delivering IoT data to the corresponding IoT applications while ensuring time sensitivity. Based on fog computing, the proposed architecture provides efficient power management in IoT device communication between sensors and secure management of data to be decrypted based on user attributes. The functional effectiveness and the safe data management of the method proposed are compared through experiments.

Dynamic Context-Aware Event Recognition Based on Markov Logic Networks.

Event recognition in smart spaces is an important and challenging task. Most existing approaches for event recognition purely employ either logical methods that do not handle uncertainty, or probabilistic methods that can hardly manage the representation of structured information. To overcome these limitations, especially in the situation where the uncertainty of sensing data is dynamically changing over the time, we propose a multi-level information fusion model for sensing data and contextual information, and also present a corresponding method to handle uncertainty for event recognition based on Markov logic networks (MLNs) which combine the expressivity of first order logic (FOL) and the uncertainty disposal of probabilistic graphical models (PGMs). Then we put forward an algorithm for updating formula weights in MLNs to deal with data dynamics. Experiments on two datasets from different scenarios are conducted to evaluate the proposed approach. The results show that our approach (i) provides an effective way to recognize events by using the fusion of uncertain data and contextual information based on MLNs and (ii) outperforms the original MLNs-based method in dealing with dynamic data.

Design and Implementation of a Smart Home System Using Multisensor Data Fusion Technology.

This paper aims to develop a multisensor data fusion technology-based smart home system by integrating wearable intelligent technology, artificial intelligence, and sensor fusion technology. We have developed the following three systems to create an intelligent smart home environment: (1) a wearable motion sensing device to be placed on residents' wrists and its corresponding 3D gesture recognition algorithm to implement a convenient automated household appliance control system; (2) a wearable motion sensing device mounted on a resident's feet and its indoor positioning algorithm to realize an effective indoor pedestrian navigation system for smart energy management; (3) a multisensor circuit module and an intelligent fire detection and alarm algorithm to realize a home safety and fire detection system. In addition, an intelligent monitoring interface is developed to provide in real-time information about the smart home system, such as environmental temperatures, CO concentrations, communicative environmental alarms, household appliance status, human motion signals, and the results of gesture recognition and indoor positioning. Furthermore, an experimental testbed for validating the effectiveness and feasibility of the smart home system was built and verified experimentally. The results showed that the 3D gesture recognition algorithm could achieve recognition rates for automated household appliance control of 92.0%, 94.8%, 95.3%, and 87.7% by the 2-fold cross-validation, 5-fold cross-validation, 10-fold cross-validation, and leave-one-subject-out cross-validation strategies. For indoor positioning and smart energy management, the distance accuracy and positioning accuracy were around 0.22% and 3.36% of the total traveled distance in the indoor environment. For home safety and fire detection, the classification rate achieved 98.81% accuracy for determining the conditions of the indoor living environment.

Spatial distribution estimation of malaria in northern China and its scenarios in 2020, 2030, 2040 and 2050.

BACKGROUND: Malaria is one of the most severe parasitic diseases in the world. Spatial distribution estimation of malaria and its future scenarios are important issues for malaria control and elimination. Furthermore, sophisticated nonlinear relationships for prediction between malaria incidence and potential variables have not been well constructed in previous research. This study aims to estimate these nonlinear relationships and predict future malaria scenarios in northern China. METHODS: Nonlinear relationships between malaria incidence and predictor variables were constructed using a genetic programming (GP) method, to predict the spatial distributions of malaria under climate change scenarios. For this, the examples of monthly average malaria incidence were used in each county of northern China from 2004 to 2010. Among the five variables at county level, precipitation rate and temperature are used for projections, while elevation, water density index, and gross domestic product are held at their present-day values. RESULTS: Average malaria incidence was 0.107 ‰ per annum in northern China, with incidence characteristics in significant spatial clustering. A GP-based model fit the relationships with average relative error (ARE) = 8.127 % for training data (R(2) = 0.825) and 17.102 % for test data (R(2) = 0.532). The fitness of GP results are significantly improved compared with those by generalized additive models (GAM) and linear regressions. With the future precipitation rate and temperature conditions in Special Report on Emission Scenarios (SRES) family B1, A1B and A2 scenarios, spatial distributions and changes in malaria incidences in 2020, 2030, 2040 and 2050 were predicted and mapped. CONCLUSIONS: The GP method increases the precision of predicting the spatial distribution of malaria incidence. With the assumption of varied precipitation rate and temperature, and other variables controlled, the relationships between incidence and the varied variables appear sophisticated nonlinearity and spatially differentiation. Using the future fluctuated precipitation and the increased temperature, median malaria incidence in 2020, 2030, 2040 and 2050 would significantly increase that it might increase 19 to 29 % in 2020, but currently China is in the malaria elimination phase, indicating that the effective strategies and actions had been taken. While the mean incidences will not increase even reduce due to the incidence reduction in high-risk regions but the simultaneous expansion of the high-risk areas.