A novel robotic chamber system allowing to accurately and precisely determining spatio-temporal CO2 flux dynamics of heterogeneous croplands

https://doi.org/10.1016/j.agrformet.2020.108206Get rights and content

Highlights

  • Novel robotic chamber for high small-scale spatial resolution of GHG emissions

  • Reliability of robotic chamber is shown for CO2 emissions based on sample dataset

  • Temporal dynamics of CO2 emissions were influenced by phenology/weather and management

  • Spatial pattern of CO2 emissions were influenced mainly by soil types

Abstract

The precise and accurate assessment of carbon dioxide (CO2) exchange is crucial to identify terrestrial carbon (C) sources and sinks and for evaluating their role within the global C budget. The substantial uncertainty in disentangling the management and soil impact on measured CO2 fluxes are largely ignored especially in cropland. The reasons for this lies in the limitation of the widely used eddy covariance as well as manual and automatic chamber systems, which either account for short-term temporal variability or small-scale spatial heterogeneity, but barely both. To address this issue, we developed a novel robotic chamber system allowing for dozens of spatial measurement repetitions, thus enabling CO2 exchange measurements in a sufficient temporal and high small-scale spatial resolution. The system was tested from 08th July to 09th September 2019 at a heterogeneous field (100 m × 16 m), located within the hummocky ground moraine landscape of northeastern Germany (CarboZALF-D). The field is foreseen for a longer-term block trial manipulation experiment extending over three erosion induced soil types and was covered with spring barley. Measured fluxes of nighttime ecosystem respiration (Reco) and daytime net ecosystem exchange (NEE) showed distinct temporal patterns influenced by crop phenology, weather conditions and management practices. Similarly, we found clear small-scale spatial differences in cumulated (gap-filled) Reco, gross primary productivity (GPP) and NEE fluxes affected by the three distinct soil types. Additionally, spatial patterns induced by former management practices and characterized by differences in soil pH and nutrition status (P and K) were also revealed between plots within each of the three soil types, which allowed compensating for prior to the foreseen block trial manipulation experiment. The results underline the great potential of the novel robotic chamber system, which not only detects short-term temporal CO2 flux dynamics but also reflects the impact of small-scale spatial heterogeneity.

Introduction

The precise and accurate assessment of terrestrial sources and sinks for anthropogenic carbon dioxide (CO2) is crucial to understand the global carbon (C) cycle, support the development of climate policies, and project future climate change (Friedlingstein et al., 2019). However, current estimates for croplands are still subject to considerable uncertainties from field to global scale (Carlson et al., 2016; Luo et al., 2016; McDermid et al., 2017; Rosenzweig et al., 2020). Mbow et al. (2019) for example, estimated global cropland emission with an uncertainty of almost 34%. This is mainly due to insufficient knowledge regarding the effect of management practices in relation to specific soil conditions on net CO2 exchange (NEE) and the resulting CO2 source and sink function (Chenu et al., 2019).

In croplands, soil is often characterized by small-scale spatial patterns, for instance as a result of soil erosion caused inter alia by ploughing that leads to lateral redistribution of soil C (Berhe et al., 2018; Doetterl et al., 2016). This induces extreme differences in soil C, which alters soil fertility, biological activity and crop growth even within a distance of a few meters only. In addition to that, these small-scale spatial patterns might change rapidly due to altered management practices. Both together cause strong small-scale spatial differences in the C-dynamics and CO2 exchange of such croplands (Dialynas et al., 2016; Nadeu et al., 2015). However, precise information is hardly available so far, since current measurement techniques are strongly limited when it comes to evaluate the influence of management practices on the CO2 exchange of heterogeneous croplands.

While the widely used eddy covariance (EC) technique provides precise high temporal resolution CO2 flux data, the spatial resolution is relatively coarse (Baldocchi, 2014). As a result, neither multifactorial treatment comparisons on field trials (Baldocchi, 2014) nor the detection of small-scale spatial differences are possible using EC (Paustian et al., 2019), since both would require spatially distinct measurements in two dimensions (2D). Compared to that, the manual chamber (MC) measurements have the advantage of being spatially distinct, which enables them to detect small-scale spatial differences and allow its use within field experiments (Hoffmann et al., 2017; Koskinen et al., 2013; Pavelka et al. 2018). Anyhow, these measurements are labor intensive and thus characterized by either a low temporal resolution, wherefore they are usually in need of extensive gap filling, commonly conducted using e.g., empirical modeling approaches (Moffat et al., 2007), or limited to shorter term studies (< 1 year; e.g., Srivastava et al., 2018). As a result, short-term changes in CO2 dynamics caused by e.g., weather conditions and crop development might be misrepresented in annual or multiyear CO2 exchange measurements using MC. Accordingly, it has been shown that measurements using MC are less reliable than EC measurements (Lucas-Moffat et al., 2018). In principle, automatic chambers (AC) can combine the advantages of EC (temporal resolution) and MC (spatial resolution) measurements (Baldocchi, 2014; Pumpanen et al., 2004). However, AC placed permanently above measurement plots, are substantially influencing the microclimate and crop development underneath (Hoffmann et al., 2017), which might bias CO2 flux estimates. In addition to that, they are restricted to the total number of measurement plots (Morin et al., 2017), which limits the amount of treatments to be compared and directly affects the number of repetitions being possible.

To overcome these disadvantages, we developed a novel robotic chamber system for conducting CO2 exchange measurements in a high temporal and spatial resolution. The developed robotic chamber system is mainly inspired by former developments described by Kiese et al. (2018) and Pütz et al. (2016), who used opaque chambers capable to move in two dimensions (2D) connected to a small gantry crane to monitor GHG emissions at grassland sites. This represents a further development of the 1D approaches presented by Keane et al. (2018, 2019), who used a single chamber moving along a line. We hypothesize that the presented novel robotic chamber system will enable us to record not only short-term temporal dynamics but also the impact of management practices and small-scale spatial differences in soil condition on measured NEE and derived flux components, ecosystem respiration (Reco) and gross primary production (GPP). To test this hypothesis, measurements were carried out from the beginning of July to the beginning of September 2019 on heterogeneous cropland, on which the establishment of a block trial manipulation experiment, dedicated to new solutions for permanent additional C and N storage is planned. The novel robotic system should help us to find out whether the spatial heterogeneity of C-dynamics is low within the areas of the individual soils, which is a prerequisite for obtaining trustworthy results in the planned field experiment.

Section snippets

Study area

The study area “CarboZALF-D” is located within the Uckermark region, an intensively agricultural used hummocky ground moraine landscape in the northeast of Germany (53° 23´ N, 13° 47´ E; ~50-60 m a.s.l; Fig. 1a). The soil types of the selected undulating field (110 × 16 m, height difference between center and edges 1m) at a summit of the study area represent the full gradient of erosion stages present in this type of landscape and usually occur in a very close proximity (<30m; Sommer et al.,

Data reliability

Using the described robotic chamber system a total of 14,099 measurements were conducted (34% during nighttime (Reco); 66% during daytime (NEE)) during the study period. Out of these measurements, 11 % did not meet linear regression assumptions when a linear regression was applied to the entire measurement length of 7 min, with a small difference between chamber 1 (8 %) and chamber 2 (13 %). Hence, measurements of chamber 2 seem to be slightly less reliable and more prone to potential chamber

Data availability statement

The data that support the findings of this study are openly available through the ZALF open-research-data repository at https://www.doi.org/10.4228/ZALF.DK.143

Declaration of Competing Interest

None.

Acknowledgements

This study was supported by the German Federal Ministry of Food and Agriculture (FNR Grant: 22404117). The authors are very grateful to the Pfannenstiel ProProject GmbH for the excellent collaboration in designing as well as constructing the gantry crane system and ATTEC Automation GmbH for programming the system control.

References (55)

  • J. Pumpanen et al.

    Comparison of different chamber techniques for measuring soil CO 2 efflux

    Agric. For. Meteorol.

    (2004)
  • M. Sommer et al.

    Feedbacks of soil erosion on SOC patterns and carbon dynamics in agricultural landscapes-The CarboZALF experiment

    Soil Tillage Res

    (2016)
  • M. Sommer et al.

    Modelling soil landscape genesis — A “ time split ” approach for hummocky agricultural landscapes

    Geoderma

    (2008)
  • B. Tan et al.

    Possible effect of soil organic carbon on its own turnover: A negative feedback

    Soil Biol. Biochem.

    (2014)
  • H. Yan et al.

    Modeling gross primary productivity for winter wheat-maize double cropping system using MODIS time series and CO2 eddy flux tower data

    Agric. Ecosyst. Environ.

    (2009)
  • X.M. Zhao et al.

    Simulation of accumulation and mineralization (CO2 release) of organic carbon in chernozem under different straw return ways after corn harvesting

    Soil Tillage Res

    (2016)
  • Y. Zhou et al.

    Examining the short-term impacts of diverse management practices on plant phenology and carbon fluxes of Old World bluestems pasture

    Agric. For. Meteorol.

    (2017)
  • D. Baldocchi

    Measuring fluxes of trace gases and energy between ecosystems and the atmosphere – the state and future of the eddy covariance method

    Glob. Chang. Biol.

    (2014)
  • A.A. Berhe et al.

    Role of Soil Erosion in Biogeochemical Cycling of Essential Elements: Carbon, Nitrogen, and Phosphorus

    Annu. Rev. Earth Planet. Sci.

    (2018)
  • A.C. Burnett et al.

    Nutrient sink limitation constrains growth in two barley species with contrasting growth strategies

    Plant Direct

    (2018)
  • K.M. Carlson et al.

    Greenhouse gas emissions intensity of global croplands

    Nat. Clim. Chang.

    (2016)
  • Y.G. Dialynas et al.

    Global Biogeochemical Cycles of soil erosion on the carbon cycle

    Global Biogeochem. Cycles

    (2016)
  • DIN ISO 10390

    Bodenbeschaffenheit

    (1997)
  • DIN ISO` 10694

    Bodenbeschaffenheit

    (1996)
  • P. Friedlingstein et al.

    Global Carbon Budget 2019 1

    Earth Syst. Sci. Data

    (2019)
  • T.S. Gahoonia et al.

    Phosphorus (P) acquisition of cereal cultivars in the field at three levels of P fertilization

    Plant Soil

    (1999)
  • P. Groenendijk et al.

    Prediction of Nitrogen and Phosphorus leaching to groundwater and surface waters

    Process descriptions of the ANIMO4.0 model

    (2005)
  • Cited by (4)

    • Similar strong impact of N fertilizer form and soil erosion state on N<inf>2</inf>O emissions from croplands

      2023, Geoderma
      Citation Excerpt :

      Information on erosion impact on the importance of cropland as a source and sink of climate-relevant trace gases is still very incomplete. So far, the focus has been mainly on the influence the soil erosion status might have on carbon dioxide (CO2) fluxes (Doetterl et al., 2016; Hoffmann et al., 2017; Lal, 2019; Vaidya et al., 2021). For instance, Hoffmann et al. (2018) reported that the C dynamics are driven by soil erosion states and plant phenology rather than N fertilizer form.

    • Modeling carbon storage in urban vegetation: Progress, challenges, and opportunities

      2022, International Journal of Applied Earth Observation and Geoinformation
      Citation Excerpt :

      Mainstream flux networks (including FLUXNET, AmeriFLUX, ILTER, and ChinaFLUX, etc.) (Baldocchi et al., 2018; Houghton 2020) suggest that flux towers are distributed widely in East Asia, North America and Australia, while sparsely in Africa, South America, Middle East, and Central Asia (Baldocchi et al., 2018). In particular, flux towers in urban environments are rare and with limited accessibility (Vaidya et al., 2021; Yuan et al., 2013). The unevenness, sparseness, and limited accessibility of observational data greatly restrain the modeling of UV carbon storage (Wu et al., 2021).

    View full text