Development and evaluation of the interactive Model for Air Pollution and Land Ecosystems (iMAPLE) version 1.0

Yue, Xu; Zhou, Hao; Tian, Chenguang; Ma, Yimian; Hu, Yihan; Gong, Cheng; Zheng, Hui; Liao, Hong

doi:https://doi.org/10.5194/gmd-2023-144

Preprints

https://doi.org/10.5194/gmd-2023-144

Preprints

Submitted as: model evaluation paper

19 Oct 2023

Submitted as: model evaluation paper |

| 19 Oct 2023

Status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

Development and evaluation of the interactive Model for Air Pollution and Land Ecosystems (iMAPLE) version 1.0

Xu Yue, Hao Zhou, Chenguang Tian, Yimian Ma, Yihan Hu, Cheng Gong, Hui Zheng, and Hong Liao

Abstract. Land ecosystems are important sources and sinks of atmospheric components. In turn, air pollutants affect the exchange rates of carbon and water fluxes between ecosystems and atmosphere. However, these biogeochemical processes are usually not well presented in the Earth system models, limiting the explorations of interactions between land ecosystems and air pollutants from the regional to global scales. Here, we develop and validate the interactive Model for Air Pollution and Land Ecosystems (iMAPLE) by upgrading the Yale Interactive terrestrial Biosphere model with process-based water cycles, fire emissions, wetland methane (CH₄) emissions, and the trait-based ozone (O₃) damages. Within the iMAPLE, soil moisture and temperature are dynamically calculated based on the water and energy balance in soil layers. Fire emissions are dependent on dryness, lightning, population, and fuel load. Wetland CH₄ is produced but consumed through oxidation, ebullition, diffusion, and plant-mediated transport. The trait-based scheme unifies O₃ sensitivity of different plant functional types (PFTs) with the leaf mass per area. Validations show correlation coefficients (R) of 0.59–0.86 for gross primary productivity (GPP) and 0.57–0.84 for evapotranspiration (ET) across the six PFTs at 201 flux tower sites, and yield an average R of 0.68 for CH₄ emissions at 44 sites. Simulated soil moisture and temperature match reanalysis data with the high R above 0.86 and low normalized mean biases (NMB) within 7 %, leading to reasonable simulations of global GPP (R=0.92, NMB=1.3 %) and ET (R=0.93, NMB=-10.4 %) against satellite-based observations for 2001–2013. The model predicts an annual global area burned of 507.1 Mha, close to the observations of 475.4 Mha with a spatial R of 0.66 for 1997–2016. The wetland CH₄ emissions are estimated to be 153.45 Tg [CH₄] yr^-1 during 2000–2014, close to the multi-model mean of 148 Tg [CH₄] yr^-1. The model also shows reasonable responses of GPP and ET to the changes in diffuse radiation, and yields a mean O₃ damage of 2.9 % to global GPP. The iMAPLE provides an advanced tool for studying the interactions between land ecosystem and air pollutants.

Received: 30 Jun 2023 – Discussion started: 19 Oct 2023

Download & links

Preprint (PDF, 3534 KB)

Supplement (317 KB)

Download & links

Xu Yue, Hao Zhou, Chenguang Tian, Yimian Ma, Yihan Hu, Cheng Gong, Hui Zheng, and Hong Liao

Status: closed

RC1:
'Comment on gmd-2023-144', Anonymous Referee #1, 04 Jan 2024

General comments:

This manuscript describes the development and validation of the interactive Model for Air Pollution and Land Ecosystems (iMAPLE). This involves coupling the process-based water cycle module from Noah-MP to an updated version of the Yale Interactive terrestrial Biosphere (YIBs) model.

The manuscript is well written, provides a comprehensive documentation of the development work, and includes a substantial expansion of the observations used in the evaluation of earlier versions of the YIBs model.

Specific comments:

In Section 2.3, the simulations performed are described and called “BASE”, “BASE_NW”, “O3LMA” and “O3S2007” but they are not consistently referred to using these names during the rest of the manuscript. It would aid the reader if the simulation names were used to refer to them throughout, and in Figure captions.

Line 56: could you be more specific here than “the ecosystem”

Line 57 – 61: as the size of the estimated net carbon sink is not constant over time (which you mention later in the Introduction) can you state a time period for the 2 Pg C yr^-1 value quoted here?

Line 212: Ws is mentioned here but I don’t think it’s defined (apologies if I missed that) and it’s not currently clear how this relates to equation 7, could you clarify – perhaps it should be Wsoil?

Line 261: this is slightly confusing because “U” is defined in the sentence previously but “UP” is included in equation 20 and not yet defined. Could you rearrange the text to clarify?

Line 266: I dont think f_RH and f_θ are defined

Line 328 – 330: could you expand on this, which plant related factors determine A_CH4 in the model, is it parameterised?

Line 399: where do the surface O₃ concentrations required for the parameterisations come from (in the absence of coupling to an atmospheric chemistry model)?

Lines 424 – 431: would this description of the observations be better placed in Section 2.4 above?

Line 436: 438: are you basing the point that iMAPLE improves GPP simulations as compared to YIBs on simulations presented here (i.e. BASE_NW) or referring to previously published evaluations of YIBs? If the former can you refer to any figures that demonstrate this, if the latter can you include any comparable statistics from previous work?

Line 462: should the second site mentioned here be US-Tw4 (as referenced in the next sentence)? Could you also include here what the simulated CH₄ flux is for the gridcell that contains these two sites, for the corresponding time period? It would be useful for the reader to understand whether the simulated value lies somewhere between the two observed values or not.

Lines 554 - 565: This section describes the impact of O₃ damage on GPP under 2 different schemes but it would benefit from some clarity around the level of O₃ damage being simulated. I think panel (a) must represent the difference between GPP in the O3LMA simulation and the BASE simulation, but this needs to be stated in the discussion and Figure 12 caption. This is important because you go on to compare the impact on GPP to the value from Ma et al 2023 but it is not currently clear if the two % values are really comparable.

Lines 568 - 571: is this based on separating the FLUXNET or MERRA-2 shortwave radiation into diffuse and direct? It would be useful to add a note here to clarify that.

Line 1086: specify in the caption for Figure 3 that this data is from the BASE simulation (if it is) - this suggestion applies to all Figures

Line 1097: refer to panels (a) and (b) in the caption. Can you label the axes in panel (b) to specify that these are observed / simulated CH₄ fluxes, with units.

Line 1104: please label the colour bars in Figure 6, or add the units to the title of each panel

Line 1136: add to this caption that the anthropogenic emissions are taken from CMIP6 input (rather than being generated by iMAPLE)

Line 1142: specify the time period that the emissions represent. Assuming these are annual totals, do they represent the entire simulation period?

Technical corrections:

Line 58: “these” should be “this” and “respirations” should be “respiration”

Line 100: “matters” should be “matter”

Line 105: “assimilations” should be “assimilation”

Line 122: “BVOCs” should be “BVOC”

Line 393: “lighting” should be “lightning”

Line 429: “much” should be “many”

Line 544: I think “we” should be “as”

Citation: https://doi.org/10.5194/gmd-2023-144-RC1
- AC1: 'Reply on RC1', Xu Yue, 03 Apr 2024
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-144/gmd-2023-144-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2023-144-AC1
RC2:
'Comment on gmd-2023-144', Anonymous Referee #2, 11 Mar 2024

Review of Development and evaluation of the interactive Model for Air Pollution and Land Ecosystems (iMaple) version 1.0 by Xu Yue et al.
The paper describes a substantial upgrade to the YIBs model, the paper is well laid out and generally has an appropriate amount of detail on the model description and evaluation. I agree with the final conclusion that the new model is well suited for studying climate-chemistry-ecosystem interactions, either driven by atmospheric data or coupled to an atmospheric model.
There are several places where the text could be clearer or a little more detail could be added, I list these below.
General: Please can you add a table that lists all parameters, their values and units?
Line 143: In section 2.1, please can you state if each gridbox has a single PFT, or a mixture of PFTs? If it is a mixture of PFTs, do all PFTs share a single soil column (all draw from the same soil moisture store)?
Line 184: How are ECAN and EGRO calculated? TWS doesn’t include a canopy storage term, so what storage term is the ECAN flux taken from?
Line 200: “Here, we assume independent and exponential distributions of infiltration capacity and precipitation in each grid cell when considering soil infiltration processes and 𝑄_soil,in is the infiltration into the soil, following the approach by Schaake et al. (1996).”
I don’t understand this sentence, please can you expand on what you mean. Do you mean each grid cell is independent of all other grid cells? Or that infiltration capacity is independent of precipitation? Do you mean that there is an exponential relationship between infiltration capacity and precipitation?
Line 206: Does K₄ vary spatially? If it does vary spatially, what dataset is used to calculate K₄?
Line 215: The model’s soil column is only 2m deep, was that depth inherited from Noah-MP, or chosen by the authors? I think a deeper soil column would improve the ecosystem representation and interactions with climate, particularly during drier conditions. Perhaps, you could comment on the choice of soil of total soil depth in the discussion.
Line 240: In section 2.2.2, please state that simulated burnt area has not impact on vegetation, or feedback onto fuel load. I appreciate this is mentioned in the discussion, but I think it should be mentioned here too.
Line 254: What are the units of PD?
Line 261: This function is complicated, it is difficult to know how the burnt area is related to the atmospheric drivers. Please can you plot BA as a function of U for a fixed RH, and plot BA as a function of RH for a fixed U? If plots of these relationships exist in Pechony and Shindell (2009) or Li et al. (2012), you could refer to their figures, but it would still be good to add some description of the relationships, e.g. does BA depend strongly on U, is RH more important, what happens if U is zero?
Line 270: Why were these values of RH_low and RH_up chosen?
Line 298: Should the left hand side of equation 30 be “f_TS” and not “Q₁₀”?
Line 363: How do you solve equations 36 and 38? Do you use an iterative process?
Line 385: How is the spin-up run? What driving data is used?
Line 399: Is LMA PFT-specific? Are you using a different map of LMA for each PFT?
Line 418: Please state that CMIP6 anthropogenic CH4 emissions are used for context and not for validation purposes.
Line 440: In addition to biases in meteorological input, it would be good to acknowledge that 1x1 degree gridded simulations would not be expected to match site-level observations, because of differences in vegetation cover and soil properties.
Line 442: Why would the increase in site number and record length decrease R?
Line 482: In this paragraph, it would be good to include the global total GPP in Pg/yr. This would allow easy comparison to other estimates.
Line 512: This paragraph and figure 9 would be better if fire emissions were evaluated. The paper and the model have a focus on climate-chemistry-ecosystem interactions, and fire has been included in order to simulate emissions, not particularly to predict burnt area.
Line 624: “As a result, the interactions between fire and ecosystems are underestimated in the current model framework.” Can you speculate on how these interactions would affect your results? My guess is that the lack of feedback from burnt area to fuel load, means that the model will overestimate burnt area and fire emissions.
Line 629: Instead of “refrain”, perhaps “limits” or “prevents” would be better words to use.
Figure 6: Could consider using a colourblind friendly colour scale. I don’t know if the journal has a policy on this.
Figure 9: Would be improved by including a maps of fire emissions from iMAPLE and GFED. I’m not sure what emissions data GFED provides, but emissions of a single species would be sufficient.
Figure 10: Would be improved by adding maps of soil carbon and wetland area.
Figure 10: In the caption, please could you be clear that wetland emissions are simulated by iMAPLE and anthropogenic sources are taken from input4mips.

Citation: https://doi.org/10.5194/gmd-2023-144-RC2
- AC2: 'Reply on RC2', Xu Yue, 03 Apr 2024
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-144/gmd-2023-144-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2023-144-AC2

Status: closed

RC1:
'Comment on gmd-2023-144', Anonymous Referee #1, 04 Jan 2024

General comments:

This manuscript describes the development and validation of the interactive Model for Air Pollution and Land Ecosystems (iMAPLE). This involves coupling the process-based water cycle module from Noah-MP to an updated version of the Yale Interactive terrestrial Biosphere (YIBs) model.

The manuscript is well written, provides a comprehensive documentation of the development work, and includes a substantial expansion of the observations used in the evaluation of earlier versions of the YIBs model.

Specific comments:

In Section 2.3, the simulations performed are described and called “BASE”, “BASE_NW”, “O3LMA” and “O3S2007” but they are not consistently referred to using these names during the rest of the manuscript. It would aid the reader if the simulation names were used to refer to them throughout, and in Figure captions.

Line 56: could you be more specific here than “the ecosystem”

Line 57 – 61: as the size of the estimated net carbon sink is not constant over time (which you mention later in the Introduction) can you state a time period for the 2 Pg C yr^-1 value quoted here?

Line 212: Ws is mentioned here but I don’t think it’s defined (apologies if I missed that) and it’s not currently clear how this relates to equation 7, could you clarify – perhaps it should be Wsoil?

Line 261: this is slightly confusing because “U” is defined in the sentence previously but “UP” is included in equation 20 and not yet defined. Could you rearrange the text to clarify?

Line 266: I dont think f_RH and f_θ are defined

Line 328 – 330: could you expand on this, which plant related factors determine A_CH4 in the model, is it parameterised?

Line 399: where do the surface O₃ concentrations required for the parameterisations come from (in the absence of coupling to an atmospheric chemistry model)?

Lines 424 – 431: would this description of the observations be better placed in Section 2.4 above?

Line 436: 438: are you basing the point that iMAPLE improves GPP simulations as compared to YIBs on simulations presented here (i.e. BASE_NW) or referring to previously published evaluations of YIBs? If the former can you refer to any figures that demonstrate this, if the latter can you include any comparable statistics from previous work?

Line 462: should the second site mentioned here be US-Tw4 (as referenced in the next sentence)? Could you also include here what the simulated CH₄ flux is for the gridcell that contains these two sites, for the corresponding time period? It would be useful for the reader to understand whether the simulated value lies somewhere between the two observed values or not.

Lines 554 - 565: This section describes the impact of O₃ damage on GPP under 2 different schemes but it would benefit from some clarity around the level of O₃ damage being simulated. I think panel (a) must represent the difference between GPP in the O3LMA simulation and the BASE simulation, but this needs to be stated in the discussion and Figure 12 caption. This is important because you go on to compare the impact on GPP to the value from Ma et al 2023 but it is not currently clear if the two % values are really comparable.

Lines 568 - 571: is this based on separating the FLUXNET or MERRA-2 shortwave radiation into diffuse and direct? It would be useful to add a note here to clarify that.

Line 1086: specify in the caption for Figure 3 that this data is from the BASE simulation (if it is) - this suggestion applies to all Figures

Line 1097: refer to panels (a) and (b) in the caption. Can you label the axes in panel (b) to specify that these are observed / simulated CH₄ fluxes, with units.

Line 1104: please label the colour bars in Figure 6, or add the units to the title of each panel

Line 1136: add to this caption that the anthropogenic emissions are taken from CMIP6 input (rather than being generated by iMAPLE)

Line 1142: specify the time period that the emissions represent. Assuming these are annual totals, do they represent the entire simulation period?

Technical corrections:

Line 58: “these” should be “this” and “respirations” should be “respiration”

Line 100: “matters” should be “matter”

Line 105: “assimilations” should be “assimilation”

Line 122: “BVOCs” should be “BVOC”

Line 393: “lighting” should be “lightning”

Line 429: “much” should be “many”

Line 544: I think “we” should be “as”

Citation: https://doi.org/10.5194/gmd-2023-144-RC1
- AC1: 'Reply on RC1', Xu Yue, 03 Apr 2024
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-144/gmd-2023-144-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2023-144-AC1
RC2:
'Comment on gmd-2023-144', Anonymous Referee #2, 11 Mar 2024

Review of Development and evaluation of the interactive Model for Air Pollution and Land Ecosystems (iMaple) version 1.0 by Xu Yue et al.
The paper describes a substantial upgrade to the YIBs model, the paper is well laid out and generally has an appropriate amount of detail on the model description and evaluation. I agree with the final conclusion that the new model is well suited for studying climate-chemistry-ecosystem interactions, either driven by atmospheric data or coupled to an atmospheric model.
There are several places where the text could be clearer or a little more detail could be added, I list these below.
General: Please can you add a table that lists all parameters, their values and units?
Line 143: In section 2.1, please can you state if each gridbox has a single PFT, or a mixture of PFTs? If it is a mixture of PFTs, do all PFTs share a single soil column (all draw from the same soil moisture store)?
Line 184: How are ECAN and EGRO calculated? TWS doesn’t include a canopy storage term, so what storage term is the ECAN flux taken from?
Line 200: “Here, we assume independent and exponential distributions of infiltration capacity and precipitation in each grid cell when considering soil infiltration processes and 𝑄_soil,in is the infiltration into the soil, following the approach by Schaake et al. (1996).”
I don’t understand this sentence, please can you expand on what you mean. Do you mean each grid cell is independent of all other grid cells? Or that infiltration capacity is independent of precipitation? Do you mean that there is an exponential relationship between infiltration capacity and precipitation?
Line 206: Does K₄ vary spatially? If it does vary spatially, what dataset is used to calculate K₄?
Line 215: The model’s soil column is only 2m deep, was that depth inherited from Noah-MP, or chosen by the authors? I think a deeper soil column would improve the ecosystem representation and interactions with climate, particularly during drier conditions. Perhaps, you could comment on the choice of soil of total soil depth in the discussion.
Line 240: In section 2.2.2, please state that simulated burnt area has not impact on vegetation, or feedback onto fuel load. I appreciate this is mentioned in the discussion, but I think it should be mentioned here too.
Line 254: What are the units of PD?
Line 261: This function is complicated, it is difficult to know how the burnt area is related to the atmospheric drivers. Please can you plot BA as a function of U for a fixed RH, and plot BA as a function of RH for a fixed U? If plots of these relationships exist in Pechony and Shindell (2009) or Li et al. (2012), you could refer to their figures, but it would still be good to add some description of the relationships, e.g. does BA depend strongly on U, is RH more important, what happens if U is zero?
Line 270: Why were these values of RH_low and RH_up chosen?
Line 298: Should the left hand side of equation 30 be “f_TS” and not “Q₁₀”?
Line 363: How do you solve equations 36 and 38? Do you use an iterative process?
Line 385: How is the spin-up run? What driving data is used?
Line 399: Is LMA PFT-specific? Are you using a different map of LMA for each PFT?
Line 418: Please state that CMIP6 anthropogenic CH4 emissions are used for context and not for validation purposes.
Line 440: In addition to biases in meteorological input, it would be good to acknowledge that 1x1 degree gridded simulations would not be expected to match site-level observations, because of differences in vegetation cover and soil properties.
Line 442: Why would the increase in site number and record length decrease R?
Line 482: In this paragraph, it would be good to include the global total GPP in Pg/yr. This would allow easy comparison to other estimates.
Line 512: This paragraph and figure 9 would be better if fire emissions were evaluated. The paper and the model have a focus on climate-chemistry-ecosystem interactions, and fire has been included in order to simulate emissions, not particularly to predict burnt area.
Line 624: “As a result, the interactions between fire and ecosystems are underestimated in the current model framework.” Can you speculate on how these interactions would affect your results? My guess is that the lack of feedback from burnt area to fuel load, means that the model will overestimate burnt area and fire emissions.
Line 629: Instead of “refrain”, perhaps “limits” or “prevents” would be better words to use.
Figure 6: Could consider using a colourblind friendly colour scale. I don’t know if the journal has a policy on this.
Figure 9: Would be improved by including a maps of fire emissions from iMAPLE and GFED. I’m not sure what emissions data GFED provides, but emissions of a single species would be sufficient.
Figure 10: Would be improved by adding maps of soil carbon and wetland area.
Figure 10: In the caption, please could you be clear that wetland emissions are simulated by iMAPLE and anthropogenic sources are taken from input4mips.

Citation: https://doi.org/10.5194/gmd-2023-144-RC2
- AC2: 'Reply on RC2', Xu Yue, 03 Apr 2024
  
  The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2023-144/gmd-2023-144-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/gmd-2023-144-AC2

Xu Yue, Hao Zhou, Chenguang Tian, Yimian Ma, Yihan Hu, Cheng Gong, Hui Zheng, and Hong Liao

Supplement

https://doi.org/10.5194/gmd-2023-144-supplement

Xu Yue, Hao Zhou, Chenguang Tian, Yimian Ma, Yihan Hu, Cheng Gong, Hui Zheng, and Hong Liao

Viewed

Total article views: 576 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
452	100	24	576	40	10	19

HTML: 452
PDF: 100
XML: 24
Total: 576
Supplement: 40
BibTeX: 10
EndNote: 19

Views and downloads (calculated since 19 Oct 2023)

Month	HTML	PDF	XML	Total
Oct 2023	163	39	3	205
Nov 2023	43	3	0	46
Dec 2023	40	11	3	54
Jan 2024	63	11	1	75
Feb 2024	41	7	3	51
Mar 2024	55	19	6	80
Apr 2024	47	10	8	65

Cumulative views and downloads (calculated since 19 Oct 2023)

Month	HTML	PDF	XML	Total
Oct 2023	163	39	3	205
Nov 2023	43	3	0	46
Dec 2023	40	11	3	54
Jan 2024	63	11	1	75
Feb 2024	41	7	3	51
Mar 2024	55	19	6	80
Apr 2024	47	10	8	65

Viewed (geographical distribution)

Total article views: 569 (including HTML, PDF, and XML) Thereof 569 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 26 Apr 2024

Short summary

We develop the interactive Model for Air Pollution and Land Ecosystems (iMAPLE). The model considers the full coupling between carbon and water cycles, dynamic fire emissions, wetland methane emissions, biogenic volatile organic compounds emissions, and the trait-based ozone vegetation damage. Evaluations show that the iMAPLE is a useful tool for the study of the interactions among climate, chemistry, and ecosystems.


Total:	0
HTML:	0
PDF:	0
XML:	0