the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 May 2023
Effects of idealised land cover and land management changes on the atmospheric water cycle
Abstract. Land cover and land management changes (LCLMCs) play an important role in achieving low-end warming scenarios through land-based mitigation. However, their effects on moisture fluxes and recycling remain uncertain although they have important implications for the future viability of such strategies. Here, we analyse the impact of idealised LCLMC scenarios on atmospheric moisture transport in three different ESMs: the Community Earth System Model (CESM), the Max Planck Institute Earth System Model (MPI-ESM) and the European Consortium Earth System Model (EC-EARTH). The LCLMC scenarios comprise of a full cropland world, a fully afforested world, and a cropland world with unlimited irrigation expansion. The effects of these LCLMCs in the different ESMs are analysed for precipitation, evaporation and vertically integrated moisture flux convergence to understand the LCLMC-induced changes in the atmospheric moisture cycle. Then, a moisture tracking algorithm is applied to assess the effects of LCLMCs on moisture recycling at the local (grid cell level) and the global scale (continental moisture recycling). Our results indicate that LCLMCs are generally inducing consistent feedbacks on moisture fluxes over land in all ESMs. Cropland expansion causes drying and reduced local moisture recycling in all ESMs, while afforestation and irrigation expansion generally cause wetting and increased local moisture recycling. However, the strength of this influence varies in time and space and across the ESMs and shows a strong dependency on the dominant driver: Some ESMs show a dominance of large scale atmospheric circulation changes while other ESMs show a dominance of local to regional changes in the atmospheric water cycle only within the vicinity of the LCLMC. Overall, these results corroborate that LCLMCs can induce large effects on the atmospheric water cycle and moisture recycling, but more research is needed to constrain the uncertainty of these effects within ESMs and better evaluate land-based mitigation strategies.
- Preprint
(9892 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-953', Anonymous Referee #1, 20 Jun 2023
This paper presents results from three climate models forced by three different land use changes (a cropland expansion, a global afforestation, and an irrigation expansion) and details the changes to the global water cycle driven by these land use changes. The paper is well written and I believe the experiments are an interesting addition to the literature on this subject, which, as the authors note, is typically limited to one model performing a suite of land use change experiments.
The main part of the paper that I believe needs improvement is the methods section. I found the description of some of the analyses difficult to follow, which in turn made the paper’s results difficult to evaluate. I also feel like the authors presented a lot of results without attempting to understand them in detail, I’ll make some suggestions on this below.
One issue is the different length scales (the evaporation vs. precipitation length scale). I didn’t understand how these were calculated or how they’re different from one another. It seems like a major effort went into these calculations with the Eulerian tracking algorithm applied to the climate model output, but I didn’t fully understand i) the distinction between the two length scales, or ii) what these length scales actually correspond do in physical terms. I think a more detailed description (and maybe a schematic) is required for readers to understand the results in Figs. 5/6.
A second issue is the description of the checkerboard pattern, and how this particularity of the experiments is leveraged to understand the local versus non-local effect of the land use change. I’m confused about how the plots in Figs. 2-4 are different from those in Appendix A. Is the “non-local” change just the change in grid cells where the land use change was not applied? It seems like this conflates “non-local” with “downstream” or “teleconnections” which are two other areas of interest for land-atmosphere coupling people. I think a better description of why these checkerboard experiments were used is important for readers to understand the results.
Lines 225-226: The increase in evaporation over CESM seems confined to the southern hemisphere, which is an interesting result that the authors do bring up later. I think “mostly an increase in evaporation” in the CESM is a bit of a mischaracterization of the results from Fig. 2b.
Lines 246-248: I like this discussion, but the authors haven’t discussed the local vs. non-local difference between the two models, which seems really important for interpreting the results and (to me) is the most interesting thing about the paper. The fact that two models forced by very similar land use change scenarios produce very different local vs. non-local results seems like something that would be of very broad interest to the community and is worth a more thorough investigation that is presented here. I think a whole analysis could likely be written about this result -- it seems very important for understanding the distinctions between two models
The discussion of moisture flux convergence (MFC) is difficult to follow, and makes statements about causality that are difficult to prove in this modeling context (for example, the ITCZ shift that the authors discuss are likely driven by MFC changes in response to atmospheric circulation patterns, not the other way around as lines 266-268 claim). In the discussion of Indian MFC changes, the authors claim that reductions in temperature lead to a reduction in MFC, but I think an extra causal step is necessary because temperature doesn’t appear in the MFC equation.
In discussion of Figs. 7-8, I’m not sure how the fractional changes associated with local (land-atmosphere coupling) processes are evaluated, so it’s hard to interpret these results.
Lines 414-416: This also seems like an important finding, but since reanalyses do have interactive connections between the land surface and the atmosphere I’m not sure what the authors are implying here.
Citation: https://doi.org/10.5194/egusphere-2023-953-RC1 -
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-953/egusphere-2023-953-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
-
RC2: 'Comment on egusphere-2023-953', Arie Staal, 23 Jun 2023
This manuscript by De Hertog et al. presents an interesting analysis on the precipitation effects of idealized land cover and land management changes (LCLMCs) in three ESMs. The results are largely consistent, but also show some contrasts among regions and models, which shows that the effects of LCLMCs are not straightforward. Some results appear to be linked to artefacts related to the checkerboard pattern of LCLMCs, as the authors acknowledge when they mention that the assumptions behind the checkerboard approach are not met, but in my opinion this should not preclude publication of the manuscript. I do, however, have some (generally minor) remarks.
It needs to be better explained how the signal separation (lines 101-102) in the checkerboard procedure works exactly. For example, in lines 215-216, the results are definitely not as “clear” to me as they are to the authors. Section 4.3 mentions artefacts resulting from the checkerboard approach, but this could be expanded upon: how exactly do which artefacts come about?
Line 139 mentions that the evaporation and precipitation length scales represent the average distance that moisture travels, but I believe these length scales are not the same as averages. Please explain more carefully what these length scales are and how they should be interpreted.
It is not clear why the length scales differ so strongly among models (lines 281-282). Please elaborate on this and provide some quantifications. How do these length scales correspond to those in the literature?
The way in which Lagrangian tracking models are portrayed is not entirely accurate. Computational demand scales with number of parcels, not area (lines 152-153), and in contrast to what is claimed in lines 203-204, parcels can be released simultaneously and therefore all continental moisture can be tracked at the same time.
The manuscript is well-written overall, but please fix the following language-related issues:
- Throughout the manuscript, the authors mention “feedbacks” where these are not really feedbacks – often, if not always, “effect” seems to be the more appropriate term.
- Line 233: “boreal latitudes”. I believe “high latitudes” are meant, as boreal latitudes would mean the latitudes of the Northern Hemisphere.
- Lines 251 and 344: “is causing” should be “causes”.
Citation: https://doi.org/10.5194/egusphere-2023-953-RC2 -
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-953/egusphere-2023-953-AC1-supplement.pdf
Status: closed
-
RC1: 'Comment on egusphere-2023-953', Anonymous Referee #1, 20 Jun 2023
This paper presents results from three climate models forced by three different land use changes (a cropland expansion, a global afforestation, and an irrigation expansion) and details the changes to the global water cycle driven by these land use changes. The paper is well written and I believe the experiments are an interesting addition to the literature on this subject, which, as the authors note, is typically limited to one model performing a suite of land use change experiments.
The main part of the paper that I believe needs improvement is the methods section. I found the description of some of the analyses difficult to follow, which in turn made the paper’s results difficult to evaluate. I also feel like the authors presented a lot of results without attempting to understand them in detail, I’ll make some suggestions on this below.
One issue is the different length scales (the evaporation vs. precipitation length scale). I didn’t understand how these were calculated or how they’re different from one another. It seems like a major effort went into these calculations with the Eulerian tracking algorithm applied to the climate model output, but I didn’t fully understand i) the distinction between the two length scales, or ii) what these length scales actually correspond do in physical terms. I think a more detailed description (and maybe a schematic) is required for readers to understand the results in Figs. 5/6.
A second issue is the description of the checkerboard pattern, and how this particularity of the experiments is leveraged to understand the local versus non-local effect of the land use change. I’m confused about how the plots in Figs. 2-4 are different from those in Appendix A. Is the “non-local” change just the change in grid cells where the land use change was not applied? It seems like this conflates “non-local” with “downstream” or “teleconnections” which are two other areas of interest for land-atmosphere coupling people. I think a better description of why these checkerboard experiments were used is important for readers to understand the results.
Lines 225-226: The increase in evaporation over CESM seems confined to the southern hemisphere, which is an interesting result that the authors do bring up later. I think “mostly an increase in evaporation” in the CESM is a bit of a mischaracterization of the results from Fig. 2b.
Lines 246-248: I like this discussion, but the authors haven’t discussed the local vs. non-local difference between the two models, which seems really important for interpreting the results and (to me) is the most interesting thing about the paper. The fact that two models forced by very similar land use change scenarios produce very different local vs. non-local results seems like something that would be of very broad interest to the community and is worth a more thorough investigation that is presented here. I think a whole analysis could likely be written about this result -- it seems very important for understanding the distinctions between two models
The discussion of moisture flux convergence (MFC) is difficult to follow, and makes statements about causality that are difficult to prove in this modeling context (for example, the ITCZ shift that the authors discuss are likely driven by MFC changes in response to atmospheric circulation patterns, not the other way around as lines 266-268 claim). In the discussion of Indian MFC changes, the authors claim that reductions in temperature lead to a reduction in MFC, but I think an extra causal step is necessary because temperature doesn’t appear in the MFC equation.
In discussion of Figs. 7-8, I’m not sure how the fractional changes associated with local (land-atmosphere coupling) processes are evaluated, so it’s hard to interpret these results.
Lines 414-416: This also seems like an important finding, but since reanalyses do have interactive connections between the land surface and the atmosphere I’m not sure what the authors are implying here.
Citation: https://doi.org/10.5194/egusphere-2023-953-RC1 -
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-953/egusphere-2023-953-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
-
RC2: 'Comment on egusphere-2023-953', Arie Staal, 23 Jun 2023
This manuscript by De Hertog et al. presents an interesting analysis on the precipitation effects of idealized land cover and land management changes (LCLMCs) in three ESMs. The results are largely consistent, but also show some contrasts among regions and models, which shows that the effects of LCLMCs are not straightforward. Some results appear to be linked to artefacts related to the checkerboard pattern of LCLMCs, as the authors acknowledge when they mention that the assumptions behind the checkerboard approach are not met, but in my opinion this should not preclude publication of the manuscript. I do, however, have some (generally minor) remarks.
It needs to be better explained how the signal separation (lines 101-102) in the checkerboard procedure works exactly. For example, in lines 215-216, the results are definitely not as “clear” to me as they are to the authors. Section 4.3 mentions artefacts resulting from the checkerboard approach, but this could be expanded upon: how exactly do which artefacts come about?
Line 139 mentions that the evaporation and precipitation length scales represent the average distance that moisture travels, but I believe these length scales are not the same as averages. Please explain more carefully what these length scales are and how they should be interpreted.
It is not clear why the length scales differ so strongly among models (lines 281-282). Please elaborate on this and provide some quantifications. How do these length scales correspond to those in the literature?
The way in which Lagrangian tracking models are portrayed is not entirely accurate. Computational demand scales with number of parcels, not area (lines 152-153), and in contrast to what is claimed in lines 203-204, parcels can be released simultaneously and therefore all continental moisture can be tracked at the same time.
The manuscript is well-written overall, but please fix the following language-related issues:
- Throughout the manuscript, the authors mention “feedbacks” where these are not really feedbacks – often, if not always, “effect” seems to be the more appropriate term.
- Line 233: “boreal latitudes”. I believe “high latitudes” are meant, as boreal latitudes would mean the latitudes of the Northern Hemisphere.
- Lines 251 and 344: “is causing” should be “causes”.
Citation: https://doi.org/10.5194/egusphere-2023-953-RC2 -
AC1: 'Reply on RC1', Steven De Hertog, 23 Jul 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-953/egusphere-2023-953-AC1-supplement.pdf
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 499 | 234 | 27 | 760 | 23 | 21 |
- HTML: 499
- PDF: 234
- XML: 27
- Total: 760
- BibTeX: 23
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1