Fixed Sampling Points Research Articles

Neural Radiance Field Image Refinement through End‐to‐End Sampling Point Optimization

Neural radiance field (NeRF), capable of synthesizing high‐quality novel viewpoint images, suffers from issues like artifact occurrence due to its fixed sampling points during rendering. This study proposes a method that optimizes sampling points to reduce artifacts and produce more detailed images. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

The Risk of Water Quality Deterioration with Urban Flood Control—A Case in Wuxi

There is a demand for flood control in densely populated river network areas. Therefore, small floodgates are used for long-term and rapid water flow regulation in such contexts. However, people often disregard these floodgates’ potential interference with the natural water environment. This study focused on an urban floodgate-controlled reach and monitored the monthly data of four main pollutant indicators (TN, TP, CODMn, and NH3-N) from 2016 to 2018 at six fixed sampling points (S1–S6). The difference analysis and cluster analysis results indicated that floodgate adjustments were the dominant driving factor of water quality changes in the reach, with pollutant concentration differences observed between the floodgate opening and closing periods. The results of the Canadian Council of Ministers of the Environment Water Quality Index evaluation showed that the water quality of the floodgate-controlled reach was categorized as “marginal” or “poor”. It is particularly important to note that the concentration of nitrogen compounds exceeded the allowable limits. The results of the Mann–Kendall trend and time series analyses revealed an overall upward trend in NH3-N concentration and a localized upward trend in TP concentration and presented periodic concentration fluctuations of four pollutants (TN, TP, CODMn, and NH3-N). This study highlights that flood control management using small floodgates can pose a risk of deteriorating water quality. Therefore, it is necessary to develop scientific water quality management methods.

Hydrodynamic response of a loam soil after wetting with different methods

Different points in a field can be wetted with different natural or artificial rainfall amounts and intensities. Therefore, a non-uniform soil wetting could occur during short-term monitoring of soil hydrodynamic parameters and its global effect on soil characterization is not easily predictable. For an initially dry loam soil, a sequence of three beerkan infiltration runs was performed at fixed sampling points in a period of two weeks. Immediately after the first beerkan run, the soil was perturbed by adding an additional water volume differing with the sampling point by the amount of water (0, 100 or 227 mm) and the application methodology (rainfall simulation, another beerkan run). As compared with the initial conditions, the soil infiltration parameters (mean, initial and final infiltration rates; constant of the Horton’s infiltration model) decreased by 2.9–4.3 times after wetting. A decrease of these parameters by 1.6–2.3 times remained detectable at the end of the sampling period, even if the soil likely returned to wetness conditions close to the initial ones. Relative variability decreased too (coefficients of variation = 52–135%, depending on the parameter, for the initial runs and 36–95% for the final ones). Nevertheless, a link was recognized between the infiltration parameters determined in the subsequent runs of the sequence. The correlation was stronger when the initial runs were compared with the last runs (coefficients of determination, R2 = 0.104–0.749, depending on the parameter; R >0 at P = 0.05 in all cases) than with those performed immediately after the intermediate wetting (R2 varying from a non-significant value of 0.059 to significant 0.196–0.211 values). The hydrodynamic response of a disturbed soil appears overall related to that of the initially undisturbed soil even if the disturbing agent is not exactly the same over the sampled area. Performing other investigations by characterizing the antecedent soil conditions in more detail could help to improve interpretation and modelling of hydrological processes by a more realistic description of temporal variability of soil infiltration parameters.

Local Climate Zones, Sky View Factor and Magnitude of Daytime/Nighttime Urban Heat Islands in Balneário Camboriú, SC, Brazil

For this study on urban climatology, the study area is the city of Balneário Camboriú, belonging to the Brazilian state of Santa Catarina (SC), located at 26°59′42″ south latitude and 48°37′46″ west longitude. As it is the most vertical city in the entire Southern Hemisphere, Balneário Camboriú was selected as the study area for the development of this climate analysis. Then, this study was concerned with analyzing the formation of urban heat islands throughout the daytime and nighttime in the city of Balneário Camboriú, Santa Catarina, Brazil, on some days in October 2020, from the perspective of the local climatic zones. Seven fixed sampling points and one official weather station were selected for this research. These points were selected in order to facilitate analysis of the climatic behaviour of the urban area throughout the day, comparing it with the other points, and also to verify possible changes in the local climate in the most diverse types of LCZ. At these same points, the Sky View Factor (SVF) measurements were taken. to elaborate the map of LCZ of Balneário Camboriú, the WUDAPT method was used. There was a great variation of the SVF between the collection points, and different LCZs were mapped, which contributed to the formation of urban heat islands whose maximum magnitude was 10.8 °C and islands with freshnesses of magnitudes of −4.5 °C.

The use of an automated organism tracking microscope in mesocosm experiments

AbstractA method for automatically counting and measuring sizes and motility behavior of zooplankton and phytoplankton in water samples is presented. Two video cameras are focused on separate optical chambers of different sizes. The chambers are filled and emptied repeatedly by synchronized pumps. Real‐time motion analysis is performed by computer on the respective video feeds. Fluorescence from chlorophyll a (Chl a) is imaged at single pixel resolution. Measured parameters for individual organisms include size, swimming velocity, motility patterns, and chlorophyll fluorescence density. The system was tested during a mesocosm experiment where it was mounted on one of several mesocosm columns. The results were validated against Chl a measurements and microscopy counts. A sampling interval of 1 per day revealed detailed dynamics of chlorophyll activity as well as shifts in both the phytoplankton and zooplankton community structure over the course of a month. A helix coefficient, a metric related to organism motility behavior, showed substantial variation over time, consistent with changing plankton communities. Sampling rates as frequent as 1 per hour enables detailed analysis of diurnal vertical migration and similar phenomena at fixed sampling points.

Construction of a Prediction Model for Distance Education Quality Assessment Based on Convolutional Neural Network.

This paper introduces the principles and operation steps of convolution and pooling of convolutional neural networks in detail. In view of the shortcomings of fixed sampling points and single receptive field in traditional convolution and pooling forms, deformable convolution and deformable pooling are introduced to enhance the network's ability to adapt to image details and large displacement problems. The concepts of warp, loop optimization, and network stack are introduced. In order to improve the optimization performance of the algorithm, three subnetwork structures and stack models are designed, and various methods are used to improve the prediction accuracy of distance education quality assessment. In order to improve the accuracy and timeliness of education quality assessment, this paper proposes a distance education quality assessment model based on mining algorithms. The prediction index is selected by the improved BP neural network. It is required to establish the input layer node as the input vector based on the number of data sources since the input layer is used for data input. The neural network is trained with a quarter of the mining data, and the mining algorithm is further trained with network error trials. A fuzzy relationship matrix is created based on the assessment of teaching quality's hierarchical structure. This leads to the conclusion of the fuzzy thorough evaluation of the effectiveness of distant learning. Experiments show that the proposed model has an average accuracy of 96%, the average teaching quality modeling time is 25.44 ms, and the evaluation speed is fast.

ECOLOGICAL IMPACTS OF ECOHYDROLOGICAL INTERVENTION ON THE WETLANDS ALONG THE SOUTHEASTERN SHORE OF LAKE TANA, ETHIOPIA

Lake Tana watershed wetlands have critical functions of nutrient and sediment retention, biodiversity and hydrological functions. Although ecohydrological solutions have been proposed, a comprehensive study on the impacts of ecohydrology intervention in relation to biological aspects is limited. Information regarding the ecological impacts of ecohydrology based intervention approach in the wetland ecosystem plays an indispensable role in increasing biodiversity and improves water quality. Thus, the aim of this study was to investigate the impacts of ecohydrological intervention in relation to macrophytes, phytoplankton and zooplankton of the wetlands in Lake Tana shore. Quantitative characteristics of macrophytes and plankton sampling were conducted using transect and quadrat method along vegetation and water depth gradients. The biological data were measured on each fixed sampling point. Macrophytes, phytoplankton and zooplankton were significantly different (p<0.05) across the sampling sites. Low disturbed wetland supports higher abundance and diversity of macrophytes, phytoplankton and zooplankton than moderately and highly disturbed wetlands. Overall results of the study confirmed that biological abundance and diversity were associated with the level of disturbance of wetlands. Therefore, emphasis should be given to the factors which may lead to the degradation of wetland resources within and its catchment. Immediate ecohydrology based intervention approaches should be implemented to conserve and minimize further degradation of wetlands along the shore of Lake Tana.

Advanced Chirp Transform Spectrometer with Novel Digital Pulse Compression Method for Spectrum Detection

Based on chirp transform and pulse compression technology, chirp transform spectrometers (CTSs) can be used to perform high-resolution and real-time spectrum measurements. Nowadays, they are widely applied for weather and astronomical observations. The surface acoustic wave (SAW) filter is a key device for pulse compression. The system performance is significantly affected by the dispersion characteristics match and the large insertion loss of the SAW filters. In this paper, a linear phase sampling and accumulating (LPSA) algorithm was developed to replace the matched filter for fast pulse compression. By selecting and accumulating the sampling points satisfying a specific periodic phase distribution, the intermediate frequency (IF) chirp signal carrying the information of the input signal could be detected and compressed. Spectrum measurements across the entire operational bandwidth could be performed by shifting the fixed sampling points in the time domain. A two-stage frequency resolution subdivision method was also developed for the fast pulse compression of the sparse spectrum, which was shown to significantly improve the calculation speed. The simulation and experiment results demonstrate that the LPSA method can realize fast pulse compression with adequate high amplitude accuracy and frequency resolution. Compared to existing digital pulse compression technology, this method can significantly reduce the number of required calculations, especially for measurements of sparse signals.

A Blind Demodulation Algorithm for Underwater Acoustic MPSK Signal

Multiple phase-shift keying (MPSK) is a common carrier modulation in underwater acoustic (UWA) communication. This paper proposes a robust blind demodulation algorithm to improve the performance and practicability of blind demodulation of MPSK signals under impulsive noise and UWA sparse multipath channels. The proposed algorithm adopts a $T$ /2-spaced blind equalizer based on quasi-affine projection, so it has strong adaptability to sparse UWA multipath channels. First, the received signal is preprocessed to suppress the impulsive interference, and the parameters are estimated. Then, timing synchronization is performed point by point based on sliding discrete Fourier transform (SDFT) to avoid equalizer performance degradation or even lock-loss caused by large clock frequency deviation and long-time error accumulation. Based on this, the over-sampling signal with two fixed sampling points per symbol is obtained. Meanwhile, following the idea of memory improved proportionate affine projection algorithm and normalized factor, the cost function of the blind equalizer is constructed by using the $l_{0}$ norm penalty on the tap coefficients. In this way, the updating formula of the tap coefficients is derived, and the blind equalization is then conducted on the signal after timing synchronization to eliminate or weaken the channel influence. Finally, the residual frequency offset and phase offset of the carrier are removed through the M-power transform and the second-order decision feedback digital phase-locked loop. Simulation experiments and practical signal demodulation results indicate that the proposed algorithm achieves higher BER performance and modulation parameter robustness for impulsive noise and UWA multipath channel.

Simultaneous time-course measurements of metoprolol and α-hydroxyl metoprolol in fingermarks after oral administration by liquid chromatography tandem mass spectrometry

The aim of this study was to evaluate whether fingerprints are suitable to be applied as the biometric identification samples by testing the orally administered drugs needs to be taken daily. The dosage of BETALOC® was administered to subjects following single and multiple doses and its active ingredient metoprolol and its main metabolite α-hydroxyl metoprolol were selected as the analytes. The subjects washed their hands and pressed fingertips onto glass slides at fixed sampling points (from 1 h to 7 days), and the analytes were extracted using cotton swabs 30 times followed by ultrasonic assistance in 30℃ methanol solution for 5 min with working power of 2000 W after optimization. The drugs in blood were taken from their elbow vein and deproteinized before analysis. Analysis were performed using liquid chromatography–tandem mass spectrometry (LC/MS/MS), and their concentration time course in fingerprints and blood were evaluated and compared. Results showed that metoprolol was detected 1 h after ingestion both in fingerprints and blood, while α-hydroxyl metoprolol was detected from sampling points of 2 h in fingerprint and 3 h in blood, respectively. Drugs could be detected for longer periods in blood than in fingerprint in single dose administration. However, in multiple doses, they could reach a steady detectable state in fingerprints from the fifth day after oral administration, and could serve as a more rapid and simpler alternative for drug analysis. We demonstrate that fingerprints could be applied as the biometric identification samples for orally administered drugs in multiple-doses, and could be applied to drug testing in criminal investigations.

Synchronization for hybrid coupled reaction-diffusion neural networks with stochastic disturbances via spatial sampled-data control strategy

The synchronization of reaction-diffusion neural networks with state and spatial couplings is investigated in this article, and the time-varying delay and stochastic disturbances are considered in the proposed systems. Due to the development and merits of digital controllers, sampled-data control is a natural choice to establish synchronization in continuous-time systems. Here, we suggest a spatial sampled-data controller design, where the sampled-data (in space) measurements of the state are taken in a finite number of fixed sampling points in the spatial domain. It is assumed that the sampling intervals in space are bounded. Based on the Lyapunov stability theory, Young’s and Wirtinger’s inequalities techniques, some sufficient conditions are presented to synchronize the hybrid coupling reaction-diffusion neural networks with stochastic disturbances. Finally, the efficiency of the derived criteria will be demonstrated by resorting to two numerical examples.

Optimal sampling design to survey riparian bird populations with low detection probability

ABSTRACT Capsule: Linear censusing and occupancy models based on fixed sampling points are alternative widely used techniques to determine bird densities in riparian ecosystems, although it cannot be always properly executed. Aims: The aim is to assess the survey efficiency for river birds using occupancy models in contexts of impaired visibility owing to dense vegetation along the banks. Methods: We tested whether increasing sampling periods within each survey unit (point) at occupancy models would result in increasing detection probability values. We used two approaches in order to identify the ‘best’ design for White-throated Dippers Cinclus cinclus along forested river stretches: minimizing survey effort of standard single-season site occupancy modelling and exploratory power analysis. Results: With a detection probability of 0.26 (i.e. much lower than in previous studies), a design with 60 sites surveyed 10 min 6 times a year would be the option to survey White-throated Dippers in forested habitats if an acceptable power is required. Simulations further revealed the consistency of the results. Conclusion: We provide guidelines to establish a cost-effective survey design for any long-term monitoring citizen-based programme of a White-throated Dipper population when detection probabilities are low. A strength and novelty of the method is to take advantage of advanced probabilistic approaches (e.g. GRTS) to select the survey sites providing, among other major interests, a spatially balanced geographic coverage.

Spacial sampled-data control for [formula omitted] output synchronization of directed coupled reaction–diffusion neural networks with mixed delays

This work investigates the H∞ output synchronization (HOS) of the directed coupled reaction–diffusion (R–D) neural networks (NNs) with mixed delays. Firstly, a model of the directed state coupled R–D NNs is introduced, which not only contains some discrete and distributed time delays, but also obeys a mixed Dirichlet–Neumann boundary condition. Secondly, a spacial sampled-data controller is proposed to achieve the HOS of the considered networks. This type of controller can reduce the update rate in the process of control by measuring the state of networks at some fixed sampling points in the space region. Moreover, some criteria for the HOS are established by designing an appropriate Lyapunov functional, and some quantitative relations between diffusion coefficients, mixed delays, coupling strength and control parameters are given accurately by these criteria. Thirdly, the case of directed spatial diffusion coupled networks is also studied and, the following finding is obtained: the spatial diffusion coupling can suppress the HOS while the state coupling can promote it. Finally, one example is simulated as the verification of the theoretical results.

A temporally stratified extension of space-for-time Cormack-Jolly-Seber for migratory animals.

Understanding drivers of temporal variation in demographic parameters is a central goal of mark-recapture analysis. To estimate the survival of migrating animal populations in migration corridors, space-for-time mark-recapture models employ discrete sampling locations in space to monitor marked populations as they move past monitoring sites, rather than the standard practice of using fixed sampling points in time. Because these models focus on estimating survival over discrete spatial segments, model parameters are implicitly integrated over the temporal dimension. Furthermore, modeling the effect of time-varying covariates on model parameters is complicated by unknown passage times for individuals that are not detected at monitoring sites. To overcome these limitations, we extended the Cormack-Jolly-Seber (CJS) framework to estimate temporally stratified survival and capture probabilities by including a discretized arrival time process in a Bayesian framework. We allow for flexibility in the model form by including temporally stratified covariates and hierarchical structures. In addition, we provide tools for assessing model fit and comparing among alternative structural models for the parameters. We demonstrate our framework by fitting three competing models to estimate daily survival, capture, and arrival probabilities at four hydroelectric dams for over 200 000 individually tagged migratory juvenile salmon released into the Snake River,USA.

Improving leaf area index retrieval using spectral characteristic parameters and data splitting

ABSTRACT Spectral variables such as spectral characteristic parameters (SCPs) commonly change with intraday phenology. Empirical retrieval methods, which are generally used in leaf area index (LAI) retrieval due to their simplicity and computational efficiency, typically relate the biophysical parameter of interest to the spectral variable during the whole observation period. Whilst information regarding diurnal changes in spectral variables is necessary and useful in applied contexts. We analysed the diurnal change characteristics of canopy spectral reflectance and SCPs of winter wheat in the jointing stage based on field data collected at fixed sampling points with different vegetation canopies, and validated the effectiveness of data splitting strategy with field data collected in random sample pattern. The key results are as follows: (i) Canopy spectral reflectance of winter wheat in the jointing stage exhibited clear intraday variability, typically presenting a double-peak characteristic occurring from 11:35 to 12:34, where the reflectance changed substantively during this period. (ii) The SCPs of winter wheat in the jointing stage exhibited different diurnal patterns. Specifically, the blue edge position presented ‘blue shifts’, the yellow edge position generally exhibited steady fluctuations, and the red edge position followed divergent trends between the two sampling points due to differences in the vegetation canopy. Amplitude and area parameters exhibited a double-peak characteristic but there were slight differences between them. (iii) By dividing the whole observation period into sub-periods, the coefficient of variation (CV) of each spectral characteristic parameter can be greatly reduced, whilst the coefficient of determination (R 2) of LAI retrieval can be greatly increased. Optimal spectral parameters and sub-periods for LAI retrieval were confirmed based on the diurnal variation of SCPs. To optimize LAI retrieval the suggested spectral parameters are blue edge amplitude, red edge amplitude, and red edge area, and the sub-periods are 09:50–11:35, 11:35–12:34, 12:34–13:50, and 13:50–15:00, respectively. The 11:35–12:34 sub-period should be carefully considered due to possible midday depression of photosynthesis.

Disentangling physical and biological drivers of phytoplankton dynamics in a coastal system

This proof-of-concept study integrates the surface currents measured by high-frequency coastal radars with plankton time-series data collected at a fixed sampling point from the Mediterranean Sea (MareChiara Long Term Ecological Research site in the Gulf of Naples) to characterize the spatial origin of phytoplankton assemblages and to scrutinize the processes ruling their dynamics. The phytoplankton community generally originated from the coastal waters whereby species succession was mainly regulated by biological factors (life-cycle processes, species-specific physiological performances and inter-specific interactions). Physical factors, e.g. the alternation between coastal and offshore waters and the horizontal mixing, were also important drivers of phytoplankton dynamics promoting diversity maintenance by i) advecting species from offshore and ii) diluting the resident coastal community so as to dampen resource stripping by dominant species and thereby increase the numerical importance of rarer species. Our observations highlight the resilience of coastal communities, which may favour their persistence over time and the prevalence of successional events over small time and space scales. Although coastal systems may act differently from one another, our findings provide a conceptual framework to address physical–biological interactions occurring in coastal basins, which can be generalised to other areas.

Reliability of cardiac output measurements using LiDCOrapid\u2122 and FloTrac/Vigileo\u2122 across broad ranges of cardiac output values

Knowing a patient’s cardiac output (CO) could contribute to a safe, optimized hemodynamic control during surgery. Precise CO measurements can serve as a guide for resuscitation therapy, catecholamine use, differential diagnosis, and intervention during a hemodynamic crisis. Despite its invasiveness and intermittent nature, the thermodilution technique via a pulmonary artery catheter (PAC) remains the clinical gold standard for CO measurements. LiDCOrapid™ (LiDCO, London, UK) and FloTrac/Vigileo™ (Edwards Lifesciences, Irvine, CA) are less invasive continuous CO monitors that use arterial waveform analysis. Their calculations are based on arterial waveform characteristics and do not require calibration. Here, we evaluated LiDCOrapid™ and FloTrac/Vigileo™ during off-pump coronary artery bypass graft (OPCAB) and living-donor liver transplantation (LDLT) surgery. This observational, single-center study included 21 patients (11 OPCAB and 10 LDLT). We performed simultaneous measurements of CO at fixed sampling points during surgery using both devices (LiDCOrapid™ version 1.04-b222 and FloTrac/Vigileo™ version 3.02). The thermodilution technique via a PAC was used to obtain the benchmark data. LiDCOrapid™ and FloTrac/Vigileo™ were used in an uncalibrated fashion. We analyzed the measured cardiac index using a Bland–Altman analysis (the method of variance estimates recovery), a polar plot method (half-moon method), a 4-quadrant plot and compared the widths of the limits of agreement (LOA) using an F test. One OPCAB patient was excluded because of the use of an intra-aortic balloon pumping during surgery, and 20 patients (10 OPCAB and 10 LDLT) were ultimately analyzed. We obtained 149 triplet measurements with a wide range of cardiac index. For the FloTrac/Vigileo™, the bias and percentage error were −0.44 L/min/m2 and 74.4 %. For the LiDCOrapid™, the bias and percentage error were −0.38 L/min/m2 and 53.5 %. The polar plot method showed an angular bias (FloTrac/Vigileo™ vs. LiDCOrapid™: 6.6° vs. 5.8°, respectively) and radial limits of agreement (−63.9 to 77.1 vs. −41.6 to 53.1). A 4-quadrant plot was used to obtain concordance rates (FloTrac/Vigileo™ vs. PAC and LiDCOrapid™ vs. PAC: 84.0 and 92.4 %, respectively). We could compare CO measurement devices across broad ranges of CO and SVR using LDLT and OPCAB surgical patients. An F test revealed no significant difference in the widths of the LoA for both devices when sample sizes capable of detecting a more than two-fold difference were used. We found that both devices tended to underestimate the calculated CIs when the CIs were relatively high. These proportional bias produced large percentage errors in the present study.

Synchronization for distributed parameter NNs with mixed delays via sampled-data control

A sampled-data synchronization scheme is developed for a class of distributed parameter neural networks with mixed time delays. The design approach of the desired sampled data controller is also presented. We propose a sampled-data controller design, in which the sampled-data (in time) measurements of the state are taken in a finite number of fixed sampling points in the spatial domain. It is supposed that the sampling intervals in time and in space are bounded. Based on the extended Wirtinger’s inequality, an appropriate Lyapunov–Krasovskii functional and some free-weighting matrices are used to establish simple and less conservative synchronization criteria, which can ensure that the master systems synchronize with the slave systems via direct Lyapunov approach in terms of linear matrix inequalities. Finally, numerical simulations are carried out to demonstrate the effectiveness of the proposed synchronization method.

Application of inverse path distance weighting for high-density spatial mapping of coastal water quality patterns

One of the primary goals of coastal water quality monitoring is to characterize spatial variation. Generally, this monitoring takes place at a limited number of fixed sampling points. The alternative sampling methodology explored in this paper involves high-density sampling from an on-board flow-through water analysis system (Dataflow). Dataflow (West Palm Beach, FL, USA) has the potential to provide better spatial resolution of water quality features because it generates many closely spaced (<10 m) measurements. Regardless of the measurement technique, parameter values at unsampled locations must be interpolated from nearby measurement points in order to generate a comprehensive picture of spatial variations. Standard Euclidean interpolations in coastal settings tend to yield inaccurate results because they extend through barriers in the landscape such as peninsulas, islands, and submerged banks. We recently developed a method for non-Euclidean interpolation by inverse path distance weighting (IPDW) in order to account for these barriers. The algorithms were implemented as part of an R package and made available from R repositories. The combination of IPDW with Dataflow provided more accurate estimates of salinity patterning relative to Euclidean inverse distance weighting (IDW). IPDW was notably more accurate than IDW in the presence of intense spatial gradients.

A site-directed CO2 detection method for monitoring the spoilage of stored grains by insects and fungi in Chinese horizontal warehouses

Detection of CO2 concentration in grain bulk is an efficient way to monitor insect and microbial activities. To develop a practical CO2 detection system suitable for Chinese horizontal warehouses, multiple fixed sampling points in the grain bulk were used to determine the factors associated with CO2 detection in grain warehouses containing over 4000 tonnes of wheat or corn. CO2 detection was significantly (P < 0.01) affected by pipe size, and pipes that were less than 30-m in length (2-mm internal diameter × 4-mm external diameter) were the most suitable for CO2 detection in 6-m-high grain bulks. The gas flow needed be set to the maximum rate of the CO2 sensor in order to minimize detection time. The CO2 produced by grain respiration was relatively low, and the activity of insect and mould significantly affected the CO2 concentration in the bulk. Spatial changes in CO2 concentration around artificial hot spots were sensitively detected within 2 m of the hot spot in both the horizontal (P < 0.05) and downward directions (P < 0.01), as well as within 1 m of the hot spot in the upward direction. Therefore, the multiple fixed-point CO2 detection method has great potential as an indicator of grain spoilage in Chinese horizontal warehouses.