Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Spatial Variation of Soil Respiration in a Cropland under Winter Wheat and Summer Maize Rotation in the North China Plain

  • Ni Huang,

    Affiliation The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China

    ⨯
  • Li Wang ,

    wangli@radi.ac.cn

    Affiliation The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China

    ⨯
  • Yongsen Hu,

    Affiliation The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China

    ⨯
  • Haifeng Tian,

    Affiliations The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China, University of Chinese Academy of Sciences, Beijing, China

    ⨯
  • Zheng Niu

    Affiliation The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, China

    ⨯

Abstract

Spatial variation of soil respiration (Rs) in cropland ecosystems must be assessed to evaluate the global terrestrial carbon budget. This study aims to explore the spatial characteristics and controlling factors of Rs in a cropland under winter wheat and summer maize rotation in the North China Plain. We collected Rs data from 23 sample plots in the cropland. At the late jointing stage, the daily mean Rs of summer maize (4.74 μmol CO2 m-2 s-1) was significantly higher than that of winter wheat (3.77μmol CO2 m-2 s-1). However, the spatial variation of Rs in summer maize (coefficient of variation, CV = 12.2%) was lower than that in winter wheat (CV = 18.5%). A similar trend in CV was also observed for environmental factors but not for biotic factors, such as leaf area index, aboveground biomass, and canopy chlorophyll content. Pearson’s correlation analyses based on the sampling data revealed that the spatial variation of Rs was poorly explained by the spatial variations of biotic factors, environmental factors, or soil properties alone for winter wheat and summer maize. The similarly non-significant relationship was observed between Rs and the enhanced vegetation index (EVI), which was used as surrogate for plant photosynthesis. EVI was better correlated with field-measured leaf area index than the normalized difference vegetation index and red edge chlorophyll index. All the data from the 23 sample plots were categorized into three clusters based on the cluster analysis of soil carbon/nitrogen and soil organic carbon content. An apparent improvement was observed in the relationship between Rs and EVI in each cluster for both winter wheat and summer maize. The spatial variation of Rs in the cropland under winter wheat and summer maize rotation could be attributed to the differences in spatial variations of soil properties and biotic factors. The results indicate that applying cluster analysis to minimize differences in soil properties among different clusters can improve the role of remote sensing data as a proxy of plant photosynthesis in semi-empirical Rs models and benefit the acquisition of Rs in cropland ecosystems at large scales.

Citation: Huang N, Wang L, Hu Y, Tian H, Niu Z (2016) Spatial Variation of Soil Respiration in a Cropland under Winter Wheat and Summer Maize Rotation in the North China Plain. PLoS ONE 11(12): e0168249. https://doi.org/10.1371/journal.pone.0168249

Editor: Ben Bond-Lamberty, Pacific Northwest National Laboratory, UNITED STATES

Received: August 4, 2016; Accepted: November 28, 2016; Published: December 15, 2016

Copyright: © 2016 Huang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data and related metadata underlying the findings reported in the manuscript have been deposited in the Dryad public repository with doi:10.5061/dryad.rk568.

Funding: This work was supported by the National Natural Science Foundation of China (41301498), the Youth Innovation Promotion Association CAS (2014052), the Special Foundation for Young Scientists from the director of the Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, and the National High Technology Research and Development Program of China (863 Program) (2014AA06A511).

Competing interests: The authors have declared that no competing interests exist.

Introduction

Soil respiration (Rs) is an important process in the carbon flux between the terrestrial ecosystem and the atmosphere, and plays a critical role in global carbon cycling [1], [2]. Rs can be divided into autotrophic and heterotrophic respiration based on different biological sources [3]. Autotrophic respiration, also known as root respiration, mainly depends on the supply of photosynthetic substrates [4], [5], [6]. Heterotrophic respiration is the sum of microbial decomposition of soil organic matter [7]. Generally, autotrophic and heterotrophic respirations fluxes are regulated by different mechanisms and interact over different temporal and spatial scales, resulting in the accurate prediction of Rs difficult [8], [9]. Factors affecting Rs must be elucidated to improve the current carbon cycle models and estimate carbon efflux from ecosystems to the atmosphere [10], [11].

Environmental factors, such as soil temperature and soil moisture, are important abiotic regulators of Rs [12], [13], [14], [15]. Plant productivity or photosynthesis capacity proxies, such as leaf area index (LAI), canopy chlorophyll content (Chlcanopy), and biomass, can become the dominant biotic regulators of Rs [16], [17], [18]. This phenomenon occurs especially when Rs is obtained during the time of high root respiration [19], [20] and high rhizodeposition [21], [22], such as the peak growing season of vegetation [23].

Compared with studies on temporal variation in Rs, relatively few studies have explored the spatial variation of Rs. A few reports have contended that spatial patterns of Rs may be controlled more by photosynthesis and productivity than by soil temperature [4], [24], [25], [26]. Temporal patterns of Rs have been simulated using continuous records of temperature, moisture, and other variables [12], [27], [28], [29]. Compared with methods for estimating the temporal variation of Rs, methods for quantifying spatial variation of Rs are limited and difficult [30]. The spatial difference in Rs within a site and between sites is often not explained by climatic variables; instead, the difference is modulated by gradients in the biological activity and differences in the soil properties [26], [27], [31], [32]. These features may provide a basis to design field experiments and conduct data analysis to improve the estimation of soil CO2 emission from an ecosystem.

Researchers have used various statistical methods to disentangle the cross-correlated controlling factors of Rs from one another. For example, a structural equation modeling approach was used to identify direct and indirect affecting factors of Rs in alpine meadow [17] and maize fields [33]. Cluster analysis was performed to identify possible groups of sites where soil CO2 concentration could be affected by different factors [34]. Compared with the structural equation modeling approach, cluster analysis is simple because it does not require significant correlation among the analyzed variables and does not depend on the subjective experience and prior knowledge of the analyst [35], [36], [37].

Remote sensing technology is gradually gaining importance in research on the global carbon cycle because of its spatially extensive coverage and low cost [38], [39], [40]. However, the application of remote sensing data in studies of Rs is not always practical and presents several uncertainties. Our previous studies examined the possibility of using remotely sensed data to estimate Rs in croplands [18], grasslands [23], and forests [33]. These studies established the feasibility of remotely-sensed spectral vegetation indices (VIs) in Rs analysis. VIs representing vegetation greenness were correlated with proxies of plant productivity such as gross primary production and leaf area index. However, covariation of plant productivity and other factors (i.e., soil temperature, moisture, and soil properties) [41], [42] complicates the explanation of the relationships between VIs and Rs. When the spatial variations of environmental factors (i.e., temperature and soil moisture) at a county scale were negligible, a model incorporating VIs and soil organic carbon (SOC) content produced satisfactory accuracy for predicting Rs during the peak growing season of maize [43]. Hence, the relationships between VIs and Rs were affected by soil properties. To further improve the role of remotely sensed VIs in Rs estimation at the spatial scale, scholars must determine the mechanism through which soil properties regulate the relationships between VIs and Rs. The present study employs cluster analysis to analyze the spatial variation of Rs in a cropland under winter wheat and summer maize rotation, examines the relationship between Rs and spectral vegetation index in each cluster, and investigates how the soil properties regulate this relationship.

Materials and methods

Ethics statement

No specific permissions were required for the 23 sample plots in this study. We confirmed that the field studies did not involve endangered or protected species. The specific location of the sample plots is provided in the manuscript (Fig 1).

thumbnail
Download:
Fig 1. Location of study site and spatial distributions of sample plots.

The figure in the left bottom corner of Fig 1 is similar to Figure 1 in the reference [43] but not identical to the Figure 1 in the reference [43]. The box in the left bottom corner refers to the South China Sea islands.

https://doi.org/10.1371/journal.pone.0168249.g001

Site description

The field experiment was conducted at Xiaotangshan Precision Agriculture Experimental Base, Changping District, Beijing, North China (40° 10.6′ N, 116° 26.3′ E). This experimental base has been operational since 2001 and is used for precision agriculture research. This site is located in a warm temperate zone with a mean annual rainfall of 507.7 mm and a mean annual temperature of 13°C [44]. The soil at this site is a silt-clay loam [44]. The double cropping system of winter wheat (Triticumaestivum L.) and summer maize (Zeamays L.) is the farming practice at this site and is the dominant farming style in the North China Plain. Winter wheat is usually sowed in October and harvested in June of the following year. Approximately 10 cm-high wheat residuals are left on the field surface after harvest. Summer maize is sowed in June without tillage and then harvested in October. When precipitation is scarce, crops are irrigated depending on soil water status.

Experimental design

A very flat sample area was established at this site with slopes less than 1°and a size of 140 m × 100 m. Land leveling was conducted in this sample area about 2 years ago (November 2013). A single crop cultivar was then planted in this area, which is under uniform water and fertilizer management. Before land leveling, the sample area was subjected to different treatments, such as using different fertilizers and crop cultivars to meet different research needs [44], [45], [46]. Therefore, soil properties showed spatial variations in the sample area because of long-term differences in farm managements (i.e., fertilizer, irrigation and cultivar), which might have led to variations in the vegetation growth and Rs. To determine the spatial variation of Rs, we employed a grid sampling method, where the distance between each sample plot was approximately 20 m (Fig 1). This design was in accordance with the research results on spatial autocorrelation of soil properties at this site [45].

The field experiments were conducted at two continuous sunny days at the late jointing stages of winter wheat (April 20 to 22, 2015) and summer maize (August 3 to 5, 2015), which mostly corresponded to the period of the highest biological activity because of the maximum crop growth rate [18]. On April 13, 2015, the sample plots were fully irrigated to meet the water requirements of winter wheat growth. A heavy rain event occurred on July 27, 2015, approximately 1 week before the summer maize experiment. Therefore, the soil water content was considered to be suitable for crop growth at the time when we conducted the two filed experiments.

Field experiments were conducted at 23 sample plots (Fig 1). We conducted the summer maize experiment at the same sample plots where the winter wheat experiment was conducted using high-precision GPS positioning. Each plot size was 1.5 m × 1.5 m. In each plot, we measured the variables that might explain the spatial variation of Rs: these variables include (1) Rs; (2) biotic factors measured by aboveground biomass (AGB), leaf area index (LAI), and canopy chlorophyll content (Chlcanopy); (3) environmental factors encompassing soil water content at 0–20 cm depth (SWC20) and soil temperature at 10 cm depth (Ts10); (4) soil property factors, including soil total nitrogen (STN) content, soil total carbon (STC) content, soil carbon/nitrogen (C/N), and soil organic carbon (SOC) content; and (5) canopy spectral reflectance of winter wheat and summer maize. It is noteworthy that we only measured soil property factors during the winter wheat experiment because of the short interval between the winter wheat and summer maize experiment (3 months). We assumed that the soil properties in such a short time could be considered constant. Therefore, the measurement data for soil properties were used to analyze the spatial variation of Rs in the winter wheat and summer maize.

Measurements of soil respiration and environmental factors

In each sample plot, Rs was measured using a Rs chamber (6400–09; LiCor, Lincoln, Nebraska, USA) connected to a portable photosynthesis system (LI-6400; LiCor, Lincoln, Nebraska, USA). The Rs chamber was mounted on a PVC soil collar that was sharpened at the bottom. Four soil collars were randomly distributed in each plot for the winter wheat experiment. Six soil collars were installed in each plot for the summer maize experiment. Each Rs measurement was performed between 09:00 h and 12:00 h (local time) because fluxes measured during this time interval usually represent the daily mean flux [18]. Rs measurement procedures, soil collar placement, and Rs data processing were described in previous studies [18], [43]. After the Rs measurement on the PVC soil collar in each plot (S1 Table), the soil temperature at 10 cm depth (Ts10) and soil moisture at 0–20 cm (SM20) were measured in the collar to minimize sample difference. Detailed procedures for soil temperature and soil moisture measurements were previously described by Huang et al. [43].

Canopy reflectance measurements and vegetation index calculation

Canopy reflectance was measured after the installation of soil collars. A portable spectroradiometer (FS-FR2500, ASD, USA) was used to measure winter wheat and summer maize canopy radiance between 350 and 2500 nm with a 1 nm waveband width. The procedures for canopy reflectance measurements were described in detail by Huang et al. [18]. Based on the measured canopy reflectance data, three VIs, namely, normalized difference vegetation index (NDVI), red edge chlorophyll index (CIred edge), and enhanced vegetation index (EVI), were calculated to analyze their relationships to the biotic factors of winter wheat and summer maize. Three formulas used for the calculation of these VIs were described by Huang et al. [18].

Biotic factor measurements

LAI was measured with a LAI-2000 plant canopy analyzer (LI-COR Inc., Lincoln, Nebraska, USA). In each plot, five representative positions were selected for LAI measurement, and two repeated measurements were performed at each position. Chlleaf was obtained with a portable chlorophyll meter (SPAD-502, New Jersey, USA). The Chlleaf measurement procedures and Chlcanopy calculation were described in detail by Huang et al. [18]. AGB was measured by randomly harvesting the aboveground fresh winter wheat plants in three subplots (0.2 m×0.2 m) and three maize plants in each plot. The fresh samples were oven dried at 65°C until the mass of the sample became constant. AGB measurement damaged the samples. Thus, we conducted this measurement when all the other measurements were finished. To reduce spatial sampling and measurement errors, we averaged the LAI, Chlleaf, and AGB derived from each plot for both winter wheat and summer maize for further analysis.

Soil property measurements

Soil inside the four PVC collars in each plot was destructively sampled after measuring Rs, soil temperature, and soil moisture in the winter wheat experiment. The collected soil samples were stored at room temperature and rapidly transported to the nearby laboratory (approximately 200 meters from the sampling site) for analysis. Soil sampling procedures and soil sample processing were described elsewhere [43]. SOC content was estimated by the standard Mebius method [47]. STN and STC content were measured by an elemental analyzer (Isoprime-EuroEA3000, Milan Italy). Soil C/N was calculated from the ratio of the STC and STN content.

Data analysis

Correlation analysis was employed to examine the relationships among Rs, biotic factors, environmental factors, and soil properties. The coefficient of variation (CV) was used to represent the spatial variation of Rs and its various affecting factors. The relationships between biotic factors (i.e. LAI, AGB, and Chlcanopy) and VIs (i.e., NDVI, CIred edge, and EVI) were examined using regression analysis. The optimum VI was selected based on the determination coefficient (R2).

Previous studies [45], [46] revealed that soil properties in our experimental area exhibited spatial variance. Spatial clustering of sample plots based on soil property factors is advisable to detect the possible confounding effects of soil properties on the relationship between Rs and other biotic or abiotic factors, and elucidate the relationship between Rs and VIs. In the present study, cluster analysis was performed based on the soil property factors to quantify the similarity in the 23 sample plots. Hanesch et al. [48] demonstrated that using all the variables causes over-information in cluster analysis and leads to insufficiently distinguishable samples from one another. The high correlation among input variables will over-represent one variable and bias the cluster results [49], [50]. Correlation analysis of the soil properties (Table 1) demonstrated that the SOC content highly correlated with the STN content (Pearson’s correlation coefficient r = 0.83, p<0.001) and STC content (r = 0.86, p < 0.001). Soil C/N displayed no significant correlation (p > 0.05) with the SOC, STN, and STC content (Table 1). Therefore, the SOC content and soil C/N were considered in the cluster analysis. Moitinho et al. [51] also demonstrated that SOC and soil C/N ratio are the two most important soil property variables that affect spatial variation of Rs in a sugarcane field. Before cluster analysis, the variables were standardized using the methods of Jiang et al. [37]. Based on the results of the cluster analysis, linear regression between Rs and optimal VI was used to detect the possible relationship between Rs and the photosynthesis proxy factor derived from remote sensing data in each cluster. One-way ANOVA with the least significant difference (LSD) multiple comparison test was used to analyze differences in Rs, biotic factors, environmental factors, and soil properties among different clusters. All the statistical analyses were performed with the Statistical Package for the Social Sciences (SPSS, Chicago, Illinois, USA).

thumbnail
Download:
Table 1. Correlation coefficients among soil respiration (Rs), leaf area index (LAI), aboveground biomass (AGB), canopy chlorophyll content (Chlcanopy), soil water content at 0–20 cm depth (SWC20), soil temperature at 10 cm depth (Ts10, °C), soil total nitrogen (STN) content, soil total carbon (STC) content, soil carbon/nitrogen (C/N) ratio, and soil organic carbon (SOC) content at the late jointing stage of winter wheat and summer maize in North China plain.

https://doi.org/10.1371/journal.pone.0168249.t001

Results

Spatial variations of soil respiration, biotic, and abiotic factors

Based on the spatially measured data from 23 sample plots in the winter wheat experiment, the daily mean Rs at the late jointing stage of winter wheat was 3.77 μmol CO2 m-2 s-1 with a range of 2.40 μmol CO2 m-2 s-1 to 4.88 μmol CO2 m-2 s-1. The CV of Rs for winter wheat was 18.5% (Table 2). Biotic factors, such as LAI, AGB, and Chlcanopy, displayed high spatial variability with CV ranging from 18.2% to 25.1%. Compared with the soil temperature (Ts10, CV = 4.8%), the soil water content (SWC20, CV = 15.7%) showed larger spatial variation. Among soil property factors, SOC content demonstrated higher spatial variation (15.1%) than the other three soil properties (7.2%–10.8%; Table 2).

thumbnail
Download:
Table 2. Descriptive statistics for soil respiration, biotic and abiotic factors.

https://doi.org/10.1371/journal.pone.0168249.t002

At the late jointing stage of summer maize, the daily mean Rs was 4.74 μmol CO2 m-2 s-1 with a minimum of 3.32 μmol CO2 m-2 s-1and a maximum of 5.70 μmol CO2 m-2 s-1. The daily mean Rs was significantly higher than the corresponding values for winter wheat (p < 0.05). The spatial variation of Rs showed difference in the same field when winter wheat and summer maize were planted. The CV of Rs for summer maize (12.2%) was lower than that for winter wheat (18.5%). The similar trend in CV was observed for environmental factors (SWC20 and Ts10). The CVs of LAI and Chlcanopy for summer maize were consistently lower than their corresponding values for winter wheat.

Relationships between soil respiration and biotic or abiotic factors

During the late jointing stage of winter wheat and summer maize, none of the factors were statistically significant important in explaining the spatial variation of Rs based on the Pearson’s correlations between Rs and various biotic or abiotic factors in the 23 sample plots (Table 1). Therefore, the spatial variation of Rs was poorly explained by the spatial variations of biotic factors, environmental factors, or soil properties alone for winter wheat and summer maize. The STN, STC, and SOC content had relatively high correlation coefficients, which ranged from 0.78 to 0.86. The correlations between soil C/N and other soil properties were not statistically significant (p > 0.05). Significant correlations were found among LAI, AGB, and Chlcanopy for winter wheat and summer maize.

Cluster analysis of soil properties

During cluster analyses, the relative variance of SOC content and soil C/N significantly decreased with an increasing cluster number (Fig 2A). When the cluster number was 3, the relative variance of SOC content and soil C/N decreased to around 25%. The further increase of the cluster number did not cause a significant decline in the relative variance of SOC content and soil C/N. Thus, all the samples were classified into three clusters, according to changes in relative variance and number of available data. Based on the standardized soil C/N and standardized SOC content, Clusters I, II, and III were clearly separated from one another (Fig 2B). Fig 1C shows the spatial distribution of the 23 sample plots in the three clusters. Sample plots belonging to the same cluster indicated a high degree of similarity in their soil properties. Except for the sample plots in Cluster III, spatial continuity was obvious in Clusters I and II (Fig 1C).

thumbnail
Download:
Fig 2. Results of cluster analysis based on soil carbon/nitrogen ratio (soil C/N) and soil organic carbon (SOC) content.

(a) Changed pattern of relative variance with the increase in cluster number; (b) Points of the clusters in the standardized soil C/N-standardized SOC space. Here, cluster I, cluster II, and cluster III separated quite clearly from each other.

https://doi.org/10.1371/journal.pone.0168249.g002

Relationships between biotic factors and vegetation indices

Pearson’s correlation analysis revealed that three biotic factors (i.e., LAI, AGB, and Chlcanopy) exhibited high correlations for winter wheat and summer maize (Table 1). Therefore, LAI was selected to analyze the relationships between the biotic factors and VIs. Based on regression analysis, EVI consistently demonstrated the optimal linear relationship to LAI, with R2 = 0.75 for winter wheat and R2 = 0.71 for summer maize (S1 and S2 Figs). However, with increasing LAI, the NDVI of the winter wheat canopy showed obvious saturation, especially when LAI was larger than 3. The CIred edge and EVI greatly improved this problem, and EVI appeared to be the optimal factor among the three VIs. The logarithmic fit quantitatively illustrated this point (S1 Fig). Thus, EVI was selected for the following analysis.

Relationships between soil respiration and enhanced vegetation index

After combining all data from the 23 sample plots, EVI did not display a statistically significant relationship with Rs in the sample field under a winter wheat and summer maize rotation (Fig 3). After grouping all the data from the 23 sample plots into three clusters based on the cluster analysis of soil C/N and SOC content, the relationship between Rs and EVI was apparently improved (Fig 4). For the three clusters, the relationship between Rs and EVI could be empirically fitted as a linear function; EVI explained 72%–87% spatial variation of Rs for winter wheat and 67%–77% of that for summer maize (Fig 4). However, the linear fitting functions differed among the three clusters for both crop types; that is, the same increase in EVI corresponded to a significantly different magnitude of variation in Rs in the three clusters.

thumbnail
Download:
Fig 3. Relationship between soil respiration (Rs) and enhanced vegetation index (EVI) for (a) winter wheat and (b) summer maize.

https://doi.org/10.1371/journal.pone.0168249.g003

thumbnail
Download:
Fig 4.

Relationships between soil respiration (Rs) and enhanced vegetation index (EVI) based on the data from the three clusters for (a) winter wheat and (b) summer maize. Cluster analysis was conducted based on soil carbon/nitrogen (soil C/N) and soil organic carbon (SOC) content.

https://doi.org/10.1371/journal.pone.0168249.g004

Discussion

Response of soil respiration to soil temperature

The measured Rs (2.40–4.88 μmol CO2 m-2 s-1) was consistent with the range reported from a winter wheat plot (2.58–5.04 μmol CO2 m-2 s-1), close to the present study site [18]. The mean Rs at the jointing period of winter wheat (3.77 μmol CO2 m-2 s-1) in the present study was higher than that in the semi-arid Loess Plateau (approximately 3 μmol CO2 m-2 s-1) [52] and the Tibetan Plateau (1.42 μmol CO2 m-2 s-1) [53] but lower than that in the temperate region of the North China Plain (approximately 5 μmol CO2 m-2 s-1) [54]. At the late jointing stage of summer maize, the measured mean Rs (4.74 μmol CO2 m-2 s-1) was lower than that in the North China Plain (5.47 μmol CO2 m-2 s-1) [18] and higher than that in the Northeast China plain (approximately 3.7 μmol CO2 m-2 s-1) [55]. The present results supported the previous studies, which suggested that Rs of the same crop in different regions might vary with climate and cropland management practices [56], [57].

For the field sampling of winter wheat and summer maize, each Rs measurement was conducted between 09:00 h and 12:00 h (local time) on two continuous sunny days. Within the Ts10 range of 12.7°C– 14.9°C for winter wheat and 22.1°C– 25.2°C for summer maize, the spatial variation of Rs did not display the predicted increase with the increase in soil temperature (Table 1). This result did not contradict the general positive response of Rs to temperature but indicated that temperature is not necessarily the most important factor in explaining the spatial variation of Rs. Previous studies revealed that the spatial variation of Rs within an ecosystem was poorly explained by the spatial variation of soil temperature [25], [27], [43]. In addition, the spatial variation of Ts10 was low, with CVs of 4.8% and 3.5% for winter wheat and summer maize, respectively. The inadequate temperature range would limit the significant regression of Rs versus temperature [58], [59].

Response of soil respiration to soil water content

The experimental field in the present study was under strict water management (i.e., irrigating based on crop growth and climate condition) to avoid drought. Approximately 1 week prior to the winter wheat field experiment, the sample field was fully irrigated. The soil water content was considered suitable for winter wheat growth after 1-week soil evaporation and plant water use. In addition, precipitation was prevalent in late July at our study site. After 4–5 days of a heavy rain event at the end of July, we conducted the Rs experiment in the summer maize field. Therefore, the soil water content might be optimal for Rs of winter wheat and summer maize when we conducted field experiments. It was expected that the soil water content did not significantly affect the spatial variation of Rs in the field under a winter wheat and summer maize rotation in the North China Plain. Previous studies also demonstrated that Rs shows minimal response to soil water content for a broad range of near-optimum soil water content [60], [61], [62]. However, the soil water content could become the dominant factor controlling Rs under extremely wet or dry conditions [13], [62], [63].

Influence of biotic factors on soil respiration

Several factors, such as soil, climate, and human management, affect crop growth in croplands [64], [65]. The soil property is just one of these factors. In the present study, biotic and abiotic factors were evaluated to explain the spatial variation of Rs (Table 1). Biotic factors (i.e., LAI, AGB, and Chlcanopy) reflected the conditions of crop growth and displayed considerably higher spatial variation (with a mean CV of approximately 23% for winter wheat and 21% for summer maize) than soil properties (with a mean CV of approximately 10%).

Biotic factors affect root respiration and consequently Rs when environmental factors are not limiting for Rs [2], [24], [66]. Typically, the root activity of crops was high at the vegetative growth stage and low at the reproductive growth stage. We conducted field experiments at the late jointing stage of winter wheat and summer maize, which nearly corresponded with the peak of the vegetative growth period. Thus, the proportion of live root respiration accounted for total Rs might be high during our measurement periods. However, we did not obtain the field-measured root respiration to support this assumption. A previous study [67] demonstrated that root respiration accounted for approximately 60% of the total Rs in a winter wheat stand at the late jointing stage; the spatial variability in Rs mainly represented the spatial variability of the autotrophic component. Ding et al. [68] demonstrated that autotrophic respiration of a maize cropland could reach up to 70% of the total Rs at the jointing period. Although we did not measured live root biomass for each plot in the present study, we measured live root biomass of winter wheat and summer maize during the growing season at the same study site (in 2011) and a nearby sample plot (in 2010). LAI showed a good linear relationship to live root biomass at the seasonal time scale for winter wheat and summer maize (S3 Fig). Therefore, live root biomass can be inferred from LAI in this study.

Relationships between spectral vegetation indices and LAI

Among the three VIs, NDVI tends to be saturated at high vegetation densities and showed a poor linear relationship to LAI than EVI and CIred edge for winter wheat and summer maize (S1 and S2 Figs). This trend may be attributed to the fact that EVI and CIred edge improve the canopy background reflectance; both are also more sensitive to variation in dense vegetation than NDVI [69], [70]. As a greenness vegetation index, EVI could serve as a strong proxy for plant productivity [71], [72]. In the present study, the biotic factor LAI was used as surrogate to processes related to total carbon uptake (or plant photosynthesis) by crops. The strong correlation between EVI and LAI supported these assumptions (S1 and S2 Figs).

Influence of soil properties on the relationship between soil respiration and spectral vegetation index

A single linear function could not describe the relationship between Rs and EVI for winter wheat and summer maize in this study (Fig 3). However, after clustering based on soil C/N and SOC content, the linear relationships between Rs and EVI were significant (p < 0.01) in each cluster for both crops (Fig 4). These findings indicated that the spatial variation of Rs in the present study may be attributed to the spatial variations of soil properties and biotic factors. Similarity, Xu and Qi [28] obtained data from two 20 m × 20 m plots and reported that biological factors and soil properties dominated the spatial variation of soil CO2 efflux in a young ponderosa pine plantation. This result also agreed well with Huang et al. [43], who reported that the LAI and SOC content directly affected the spatial variability of Rs during the peak growing season of maize in three counties in North China. However, it should be noted that these results were obtained from a 140 m × 100 m sample area (present study), two 20 m × 20 m plots [28] and three counties [43] because physical and biological controls on soil CO2 efflux might differ for ecosystems at large scales. For example, the soil water content could be an important factor affecting the spatial variation of soil CO2 efflux at large scales, where the soil drainage class varies across landscapes [16], [73], [74].

In each cluster, the strong linear relationship between EVI and Rs at the late jointing stage of winter wheat and summer maize was mainly caused by background correlation of both quantities with biotic factors [e.g. LAI; S1(C), S2(C) and S4 Figs]. EVI and Rs displayed a strong linear relationship in each cluster (Fig 4) because of the close relationship between plant photosynthesis and Rs [5], [75]. Inconsistencies in the relationships between EVI and Rs before and after the cluster analysis were possibly due to the confounding influences of soil properties. Under the field conditions, the effects of biotic and abiotic factors on Rs are often confounded between each other [55]. The difference between the soil properties in each cluster was reduced by cluster analysis (Fig 2), which clarified the relationship between EVI and Rs (Fig 4). In each cluster, a significant positive linear relationship was observed between EVI and Rs (Fig 4). This observation suggested that Rs was higher at sites with higher photosynthetic capacity when the values of soil properties were maintained at a certain range. This phenomenon agreed with several previous studies, where the spatial variability of vegetation productivity affected spatial variation of Rs in the absence of other restricting factors [20], [24], [76].

The linear relationship between Rs and EVI in each cluster for both winter wheat and summer maize (Fig 4) could not be described by a single function, which indicated that the photosynthetic dependence of Rs was influenced by processes related to soil properties and crop types. With increasing EVI, a different increasing rate of Rs in each cluster was observed (Fig 4) because of the interaction effects of soil (belowground) and vegetation (aboveground) on Rs (S2 Table). For winter wheat and summer maize, significant differences were noted in biotic factors (i.e., LAI, AGB and Chlcanopy) and soil property factors (i.e., STC content, Soil C/N, and SOC content) between Cluster Ⅰ and Cluster Ⅲ, and between Cluster Ⅱ and Cluster Ⅲ. These results indicated that the application of cluster analysis to minimize differences in soil properties among different clusters may improve the role of remote sensing data as a substitute of plant photosynthesis in semi-empirical Rs models and benefit the acquisition of Rs in cropland ecosystems at large scales.

Supporting Information

S1 Table. Description of 23 sample plots where soil respiration measurements were taken for winter wheat and summer maize.

https://doi.org/10.1371/journal.pone.0168249.s001

(DOCX)

S2 Table. Effect of different clusters on average soil respiration, biotic and abiotic factors for winter wheat and summer maize.

Different letters indicate significant differences among clusters (p < 0.05).

https://doi.org/10.1371/journal.pone.0168249.s002

(DOCX)

S1 Fig. Relationships between leaf area index (LAI) and spectral vegetation indices (VIs) at the jointing stage of winter wheat.

The VIs are normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and red edge chlorophyll index (CIred edge). All relationships were statistically significant at p < 0.0001.

https://doi.org/10.1371/journal.pone.0168249.s003

(TIF)

S2 Fig. Relationships between leaf area index (LAI) and spectral vegetation indices (VIs) at the jointing stage of summer maize.

The VIs are normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), and red edge chlorophyll index (CIred edge). All relationships were statistically significant at p < 0.0001.

https://doi.org/10.1371/journal.pone.0168249.s004

(TIF)

S3 Fig.

Relationships between leaf area index (LAI) and live root biomass (a) during the 2011 growing season of winter wheat at our study site and (b) during the 2010 growing season of summer maize at a nearby sample plot.

https://doi.org/10.1371/journal.pone.0168249.s005

(TIF)

S4 Fig.

Relationships between soil respiration (Rs) and leaf area index (LAI) based on the data from the three clusters for (a) winter wheat and (b) summer maize. Cluster analysis was conducted based on soil carbon/nitrogen (soil C/N) and soil organic carbon (SOC) content.

https://doi.org/10.1371/journal.pone.0168249.s006

(TIF)

Acknowledgments

This work was supported by the National Natural Science Foundation of China (41301498), the Youth Innovation Promotion Association CAS (2014052), the Special Foundation for Young Scientists from the director of the Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, and the National High Technology Research and Development Program of China (863 Program) (2014AA06A511).

Author Contributions

  1. Conceptualization: NH LW ZN.
  2. Data curation: NH.
  3. Formal analysis: NH.
  4. Funding acquisition: LW.
  5. Investigation: NH YSH HFT.
  6. Methodology: NH.
  7. Project administration: NH.
  8. Resources: LW.
  9. Software: NH.
  10. Supervision: NH LW ZN.
  11. Visualization: NH.
  12. Writing – original draft: NH.
  13. Writing – review & editing: NH.

References

  1. 1. Raich JW, Potter CS, Bhagawati D. Interannual variability in global soil respiration, 1980–94. Global Change Biology. 2002; 8: 800–812.
  2. 2. Ryan MG, Law BE. Interpreting, measuring, and modeling soil respiration. Biogeochemistry. 2005; 73: 3–27.
  3. 3. Hanson PJ, Edwards NT, Garten CT, Andrews JA. Separating root and soil microbial contributions to soil respiration: a review of methods and observations. Biogeochemistry. 2000; 48: 115–146.
  4. 4. Högberg P, Nordgren A, Buchmann N, Taylor AFS, Ekblad A, Högberg MN, et al. Large-scale forest girdling shows that current photosynthesis drives soil respiration. Nature. 2001; 411: 789–792. pmid:11459055
  5. 5. Tang JW, Baldocchi DD, Xu L. Tree photosynthesis modulates soil respiration on a diurnal time scale. Global Change Biology. 2005; 11: 1298–1304.
  6. 6. Wang B, Zha TS, Jia X, Wu B, Zhang YQ, Qin SG. Soil moisture modifies the response of soil respiration to temperature in a desert shrub ecosystem. Biogeosciences. 2014; 11: 259–268.
  7. 7. Kuzyakov Y, Larionova AA. Root and rhizomicrobial respiration: A review of approaches to estimate respiration by autotrophic and heterotrophic organisms in soil. Journal of Soil Science and Plant Nutrition. 2005; 168: 503–520.
  8. 8. Olsson P, Linder S, Giesler R, Högberg P. Fertilization of boreal forest reduces both autotrophic and heterotrophic soil respiration. Global Change Biology. 2005; 11: 1745–1753.
  9. 9. Zhang C, Niu D, Hall SJ, Wen H, Li X, Fu H, et al. Effects of simulated nitrogen deposition on soil respiration components and their temperature sensitivities in a semiarid grassland. Soil Biology and Biochemistry. 2014; 75: 113–123.
  10. 10. Davidson EA, Janssens IV, Luo YQ. On the variability of respiration in terrestrial ecosystems: moving beyond Q10. Global Change Biology. 2006; 12: 154–164.
  11. 11. Vargas R, Carbone MS, Reichstein M, Baldocchi DD. Frontiers and challenges in soil respiration research: from measurements to model-data integration. Biogeochemistry. 2011; 102, 1–13.
  12. 12. Davidson EA, Belk E, Boone RD. Soil water content and temperature as independent or confound factors controlling soil respiration in a temperature mixed hardwood forest. Global Change Biology. 1998; 4: 217–227.
  13. 13. Gaumont-Guay D, Black TA, Griffis TJ, Barr AG, Jassal RS, Nesic Z. Interpreting the dependence of soil respiration on soil temperature and water content in a boreal aspen stand.agric for meteorol. Agricultural and Forest Meteorology. 2006; 140: 220–235.
  14. 14. Phillips SC, Varner RK, Frolking S, Munger JW, Bubier JL, Wofsy SC, Crill PM. Interannual, seasonal, and diel variation in soil respiration relative to ecosystem respiration at a wetland to upland slope at Harvard Forest. Journal of Geophysical Research: Biogeosciences. 2010; 115: G02019.
  15. 15. Chang CT, Sabaté S, Sperlich D, Poblador S, Sabater F, Gracia C. Does soil moisture overrule temperature dependence of soil respiration in mediterranean riparian forests? Biogeosciences. 2014; 11: 6173–6185.
  16. 16. Reichstein M, Rey A, Freibauer A, Tenhunen J, Valentini R, et al. Modeling temporal and large-scale spatial variability of soil respiration from soil water availability, temperature and vegetation productivity indices. Global Biogeochemical Cycles. 2003; 17: 1104.
  17. 17. Geng Y, Wang Y, Yang K, Wang S, Zeng H, et al. Soil respiration in Tibetan alpine grasslands: belowground biomass and soil moisture, but not soil temperature, best explain the large-scale patterns. PloS one. 2012; 7: e34968. pmid:22509373
  18. 18. Huang N, Niu Z, Zhan YL, Tappertc MC, Wu CY, et al. Relationships between soil respiration and photosynthesis-related spectral vegetation indices in two cropland ecosystems. Agricultural and Forest Meteorology. 2012; 160: 80–89.
  19. 19. Aubrey DP, Teskey RO. Root‐derived CO2 efflux via xylem stream rivals soil CO2 efflux. New Phytologist. 2009; 184: 35–40. pmid:19674328
  20. 20. Kuzyakov Y, Richkovaw O. Time lag between photosynthesis and carbon dioxide efflux from soil: a review of mechanisms and controls. Global Change Biol. 2010; 16: 3386–3406.
  21. 21. Nguyen C. Rhizodeposition of organic c by plants: mechanisms and controls. Agronomie. 2003; 23: 375–396.
  22. 22. Amos B, Walters DT. Maize root biomass and net rhizodeposited carbon: An analysis of the literature. Soil Science Society of America Journal. 2006; 70: 1489–1503.
  23. 23. Huang N, He JS, Niu Z. Estimating the spatial pattern of soil respiration in tibetan alpine grasslands using landsat tm images and modis data. Ecological Indicators. 2013; 26: 117–125.
  24. 24. Raich JW, Tufekcioglu A. Vegetation and soil respiration: Correlations and controls. Biogeochemistry. 2000; 48: 71–90.
  25. 25. Janssens IA, Lankreijer H, Matteucci G, Buchmann N. 2001; Productivity overshadows temperature in determining soil and ecosystem respiration across European forests. Global Change Biology. 7: 269–278.
  26. 26. Chen Z, Yu G, Zhu X, Wang Q, Niu S, Hu Z. Covariation between gross primary production and ecosystem respiration across space and the underlying mechanisms: a global synthesis. Agricultural and Forest Meteorology. 2015; 203: 180–190.
  27. 27. Huang N, Zheng N. Estimating soil respiration using spectral vegetation indices and abiotic factors in irrigated and rainfed agroecosystems. Plant and Soil. 2013; 367: 535–550.
  28. 28. Xu M, Qi Y. Soil-surface CO2 efflux and its spatial and temporal variations in a young ponderosa pine plantation in northern California. Global Change Biology. 2001; 7: 667–677.
  29. 29. Zhang Q, Lei HM, Yang DW. Seasonal variations in soil respiration, heterotrophic respiration and autotrophic respiration of a wheat and maize rotation cropland in the north china plain. Agricultural and Forest Meteorology. 2013; 180: 34–43.
  30. 30. Moitinho MR, Padovan MP, Panosso AR, Teixeira DDB, Ferraudo AS, Scala NL. On the spatial and temporal dependence of CO2 emission on soil properties in sugarcane (saccharum spp.) production. Soil & Tillage Research. 2015; 148: 127–132.
  31. 31. Rayment MB, Jarvis PG. Temporal and spatial variation of soil CO2efflux in a canadian boreal forest. Soil Biology and Biochemistry. 2000; 32: 35–45.
  32. 32. Katayama A, Kume T, Komatsu H, Ohashi M, Nakagawa M, Yamashita M, et al. Effect of forest structure on the spatial variation in soil respiration in a bornean tropical rainforest. Agricultural and Forest Meteorology. 2009; 149: 1666–1673.
  33. 33. Huang N, Gu L, Niu Z. Estimating soil respiration using spatial data products: a case study in a deciduous broadleaf forest in the midwest USA. Journal of Geophysical Research: Atmospheres. 2014; 119: 6393–6408.
  34. 34. Giammanco S, Bonfanti P. Cluster analysis of soil CO2 data from mt.etna (italy) reveals volcanic influences on temporal and spatial patterns of degassing. Bulletin of Volcanology. 2009; 71: 201–218.
  35. 35. Ortega RA, Santibáñez OA. Determination of management zones in corn (zea mays l.) based on soil fertility. Computers and Electronics in Agriculture. 2007; 58: 49–59.
  36. 36. Hanesch M, Scholger R, Dekkers MJ. The application of fuzzy c-means cluster analysis and non-linear mapping to a soil data set for the detection of polluted sites. Physics & Chemistry of the Earth Part A Solid Earth & Geodesy. 2001; 26: 885–891.
  37. 37. Jiang Y, Guo H, Jia Y, Cao Y, Hu C. Principal component analysis and hierarchical cluster analyses of arsenic groundwater geochemistry in the hetao basin, inner mongolia. Chemie der Erde—Geochemistry. 2015; 75.
  38. 38. Asner GP, Townsend AR, Bustamante MMC, Nardoto GB, Oler LP. Pasture degradation in the central amazon: linking changes in carbon and nutrient cycling with remote sensing. Global Change Biology. 2004; 10: 844–862.
  39. 39. Turner DP, Ollinger SV, Kimball JS. Integrating remote sensing and ecosystem process models for landscape- to regional-scale analysis of the carbon cycle. Bioscience. 2009; 54: 573–584.
  40. 40. Irisarri JGN, Oesterheld M, Paruelo JM, Texeira MA. Patterns and controls of above-ground net primary production in meadows of Patagonia. A remote sensing approach. Journal of Vegetation Science. 2012; 23: 114–126.
  41. 41. Chen D, Lan Z, Bai X, Grace JB, Bai Y. Evidence that acidification-induced declines in plant diversity and productivity are mediated by changes in below-ground communities and soil properties in a semi-arid steppe. Journal of Ecology. 101, 2013; 1322–1334.
  42. 42. Alexandrov GA, Yamagata Y. A peaked function for modeling temperature dependence of plant productivity. Ecological Modelling. 2007; 200: 189–192.
  43. 43. Huang N, Wang L, Guo Y, Hao P, Niu Z. Modeling spatial patterns of soil respiration in maize fields from vegetation and soil property factors with the use of remote sensing and geographical information system. Plos One. 2014; 9, e105150–e105150. pmid:25157827
  44. 44. Huang WJ, Lamb DW, Niu Z, Zhang YJ, Liu LY, Wang JD. Identification of yellow rust in wheat using in-situ spectral reflectance measurements and airborne hyperspectral imaging. Precision Agriculture. 2007; 8: 187–197.
  45. 45. Song XY, Wang JH, Huang WJ, Liu LY, Yan GJ, Pu RL. The delineation of agricultural management zones with high resolution remotely sensed data. Precision Agriculture. 2009; 10: 471–487.
  46. 46. Zhao CJ, Jiang AN, Huang WJ, Liu KL, Liu LY, Wang JH. Evaluation of variable‐rate nitrogen recommendation of winter wheat based on spad chlorophyll meter measurement. New Zealand Journal of Agricultural Research. 2007; 50: 735–741.
  47. 47. Nelson DW, Sommers LE. Total carbon, organic carbon, and organic matter. In: Page AL, Miller RH, Keeney DR. (Eds.), Methods of Soil Analysis. American Society of Agronomy and Soil Science Society of American, Madison. 1982; 101–129.
    • 48. Hanesch M, Scholger R, Dekkers MJ. The application of fuzzy c-means cluster analysis and non-linear mapping to a soil data set for the detection of polluted sites. Physics & Chemistry of the Earth Part A Solid Earth & Geodesy. 2001; 26: 885–891.
    • 49. Kaufman L, Rousseeuw PJ. Finding groups in data: an introduction to cluster analysis. Journal of the American Statistical Association. 1990.
    • 50. Everitt B, Rabe-Hesketh S. Cluster analysis. Quality & Quantity. 2009; 14: 75–100.
    • 51. Moitinho MR, Padovan MP, Panosso AR, Teixeira DDB, Ferraudo AS, Scala NL. On the spatial and temporal dependence of CO2 emission on soil properties in sugarcane (saccharum spp.) production. Soil & Tillage Research. 2015; 148: 127–132.
    • 52. Jiang J, Guo S, Zhang Y, Liu Q, Wang R, et al. Changes in temperature sensitivity of soil respiration in the phases of a three-year crop rotation system. Soil & Tillage Research. 2015; 150: 139–146.
    • 53. Shi PL, Zhang XZ, Zhong ZM, Ouyang H. Diurnal and seasonal variability of soil CO2 efflux in a cropland ecosystem on the tibetan plateau. Agricultural and Forest Meteorology. 2006; 137: 220–233.
    • 54. Zhang Q, Lei HM, Yang DW. Seasonal variations in soil respiration, heterotrophic respiration and autotrophic respiration of a wheat and maize rotation cropland in the north china plain. Agricultural and Forest Meteorology. 2013; 180: 34–43.
    • 55. Han G, Zhou G, Xu Z, Yang Y, Liu J, Shi K. Biotic and abiotic factors controlling the spatial and temporal variation of soil respiration in an agricultural ecosystem. Soil Biology and Biochemistry. 2007; 39: 418–425.
    • 56. Parkin TB, Kaspar TC, Senwo Z, Prueger JH, Hatfield JL. Relationship of soil respiration to crop and landscape in the walnut creek watershed. Journal of Hydrometeorology. 2005; 6: 812.
    • 57. Eugster W, Moffat AM, Ceschia E, Aubinet M, Ammann C, et al. Management effects on European cropland respiration. Agriculture Ecosystems & Environment. 2010; 139: 346–362.
    • 58. Reichstein M, Falge E, Baldocchi D, Papale D, Aubinet M, Berbigier P, et al. On the separation of net ecosystem exchange into assimilation and ecosystem respiration: review and improved algorithm. Global Change Biology. 2005; 11: 1424–1439.
    • 59. Janssens IA, Pilegaard K. Large seasonal changes in q10 of soil respiration in a beech forest. Global Change Biology. 2003; 9: 911–918.
    • 60. Balogh J, Pintér K, Fóti S, Cserhalmi D, Papp M, Nagy Z. Dependence of soil respiration on soil moisture, clay content, soil organic matter, and CO2 uptake in dry grasslands. Soil Biology and Biochemistry. 2011; 43: 1006–1013.
    • 61. Wagle P, Kakani VG. Confounding effects of soil moisture on the relationship between ecosystem respiration and soil temperature in switch grass. Bioenergy Research. 2014; 7: 789–798.
    • 62. Cook FJ, Orchard VA. Relationships between soil respiration and soil moisture. Soil Biology and Biochemistry. 2008; 40: 1013–1018.
    • 63. Chang CT, Sabaté S, Sperlich D, Poblador S, Sabater F, Gracia C. Does soil moisture overrule temperature dependence of soil respiration in mediterranean riparian forests? Biogeosciences 2014; 11: 6173–6185.
    • 64. Supit I, Diepen CAV, Wit AJWD, Wolf J, Kabat P, Baruth B, et al. Assessing climate change effects on european crop yields using the crop growth monitoring system and a weather generator. Agricultural and Forest Meteorology. 2012; 164: 96–111.
    • 65. Stadler A, Rudolph S, Kupisch M, Langensiepen M, Kruk JVD, Ewert F. Quantifying the effects of soil variability on crop growth using apparent soil electrical conductivity measurements. European Journal of Agronomy. 2015; 8–20.
    • 66. Katayama A, Kume T, Komatsu H, Ohashi M, Nakagawa M, Yamashita M, et al. Effect of forest structure on the spatial variation in soil respiration in a bornean tropical rainforest. Agricultural and Forest Meteorology. 2009; 149: 1666–1673.
    • 67. Prolingheuer N, Scharnagl B, Graf A, Vereecken H, Herbst M. Spatial and seasonal variability of heterotrophic and autotrophic soil respiration in a winter wheat stand. Biogeosciences. 2010; 12: 9137–9173.
    • 68. Ding W, Cai Y, Cai Z, Yagi K, Zheng X. Soil respiration under maize crops: effects of water, temperature, and nitrogen fertilization. Soil Science Society of America Journal. 2007; 71: 944–951.
    • 69. Huete A, Didan K, Miura T, Rodriguez EP, Gao X, Ferreira LG. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment. 2002; 83: 195–213.
    • 70. Gitelson AA, Vina A, Arkebauer TJ, Rundquist DC, Keydan G, Leavitt B. Remote estimation of leaf area index and green leaf biomass in maize canopies. Geophysical Research Letters. 2003; 30,
    • 71. Sims DA, Rahman AF, Cordova VD, El-Masri BZ, Baldocchi DD, Flanagan LB, Goldstein AH, Hollinger DY, Misson L, Monson RK, Oechel WC, Schmid HP, Wofsy SC, Xu LK. On the use of MODIS EVI to assess gross primary productivity of North American ecosystems. Journal of Geophysical Research: Biogeosciences. 2006; 111, G04015,
    • 72. Sims DA, Rahman AF, Cordova VD, et al. A new model of gross primary productivity for North American ecosystems based solely on the enhanced vegetation index and land surface temperature from MODIS. Remote Sensing of Environment. 2008; 112: 1633–1646.
    • 73. Jägermeyr J, Gerten D, Lucht W, Hostert P, Migliavacca M, Nemani R. A high‐resolution approach to estimating ecosystem respiration at continental scales using operational satellite data. Global Change Biology. 2014; 20: 1191–1210. pmid:24259306
    • 74. Hashimoto S, Carvalhais N, Ito A, Migliavacca M, Nishina K, Reichstein M. Global spatiotemporal distribution of soil respiration modeled using a global database. Biogeosciences. 2015; 12: 4331–4364.
    • 75. Vargas R, Baldocchi D, Bahn M, Hanson PJ, Hosman KP, Kulmala L. On the temporal correlation between photosynthesis and soil respiration: reconciling lags and observations. New Phytologist. 2011; 191: 1006–1017. pmid:21609333
    • 76. Levy-Varon JH, Schuster WSF, Griffin KL. The autotrophic contribution to soil respiration in a northern temperate deciduous forest and its response to stand disturbance. Oecologia. 2012; 169: 211–220. pmid:22076310