the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Jul 2024
Combined effects of topography, soil moisture and snow cover regimes on growth responses of grasslands in a low mountain range (Vosges, France)
Abstract. Growth responses of low mountain grasslands to Climate Change are poorly understood despite very large surfaces covered in Central Europe. They are characterized by still present agricultural exploitation and complex topographical features that limit species migration and increase differences in snow regimes. This study examined MODIS surface reflectances between 2000 and 2020 across the Vosges mountain grasslands to investigate trends and drivers of spatial patterns in annual maximum NDVI (Normalized Difference Vegetation Index). We found a majority of no significant trends indicating several environmental and ecological compensatory effects to warming in the Vosges Mountains. We also noted hotspots of browning grasslands (a decrease of annual maximum NDVI), largely overrepresented compared to the greening ones (an increase of annual maximum of NDVI), a pattern in contradiction with most productivity signals highlighted in European high mountain grasslands. Spatial patterns of browning are enhanced on north-facing slopes and at low elevations (<1100 m) where high producing grasslands with dominant herbaceous communities prevail. A low soil water recharge also appears pivotal to explain the probability of browning in the study site. Through the use of Winter Habitat Indices, we noted high responsiveness of low mountain grasslands to differences in intra seasonal snow regimes, partly modulated by topographic features. Prolonged and time-continuous snow cover promote higher productivity and shortened green-up period. High number of frost events result in lower productivity and prolonged green-up period. We hypothesize that observed growth responses in the Vosges Mountains are indicative of long term future responses to Climate Change in high mountain ranges. With shorter and more discontinuous snow cover, we expect higher diversity of growth responses in European low mountain grasslands due to strong contextual effects and high terrain complexity.
- Preprint
(4727 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-1935', Anonymous Referee #1, 02 Aug 2024
Peer review for the manuscript:
Combined effects of topography, soil moisture and snow cover regimes on growth responses of grasslands in a low mountain range (Vosges, France) By Herrault et al.The manuscript under consideration reports the results of applying 20 years of remote sensing data on grassland growth dynamics in a mountain range in France that exposed to range of climatic conditions and grazing. The researchers found that the majority of pixels did not show significant change in color in these 20 years and therefore suggest that Vosges Mountain grasslands exhibited predominant stable productivity trends between 2000 and 2020. The major explanation suggested for this stability is compensatory effects of genetic adaptations, traits of local plant communities, plasticity in local populations.
This is a good paper reporting data on an increasingly relevant topic. The researchers explain carefully and clearly about the working progress of analyzing the data including the complexity of co-linearity and the time-space scales. However, there are a few points of concern regarding the manuscript.
Comments:
1. One issue I was concerned about is the lack of the environmental conditions in the years of interest (2000-2020), did the air temperature increased in this period in the study site? Were there differences in precipitation events and the intervals between the events? For me this is important for interpreting the data. I would consider adding an environmental conditions figure (air temp, precipitation, VPD etc.). Then, the discussion can be more relevant to the specific conditions that the low mountain range grassland experienced.
2. The second point I suggest is to make some terms clearer. First, “soil water content”, this parameter appears in the title, in the discussion, and in the conclusions, but I couldn’t find the explanation of how it was calculated in the methods section. Second, “growing season”, growing season can be varied widely in different locations on Earth and therefore it is better to define the specific dates of growing season and how did they determine. Lastly, the term “no-significant greening”, what was the specific threshold that was used to define the category of “non-significant greening”?
3. The last major point is the direct connection that the authors make between color reflectance-based indices (NDVI) to plant growth response. Plant growth response can be confusing in this context because usually it is the increase between the previous day biomass to the next day. I suggest adding a sentence or two in the introduction about how NDVI represents plant growth, does it represent higher productivity? Increase in the total pixel biomass? increase in a specific plant species biomass? Generally, to avoid the direct connection between NDVI to plant growth without explaining what NDVI directly represents.
Minor comments:
Line 5-6: rewrite this sentence “We found a majority of no significant trends indicating several environmental and ecological compensatory effects to warming in the Vosges Mountains”. It is not clear what did the authors found.
Line 44- remove question mark
Line 101- remove question mark
Line 101- “(almost) semi-continental”- should be better defined.
Line 113- all species names should be italic
Line 144 during the year or during the growing season? In the line above the authors write growing season.
Line 147- change “ff” to “of”
Line 148- “exceeds”? the value was higher than the NDVI amplitude maximum? Or lower? If lower need to change “exceeds”
Line 207- change the semi title to be more specific, maybe “growth response” instead of “phenometric”.
Citation: https://doi.org/10.5194/egusphere-2024-1935-RC1 -
AC1: 'Reply on RC1', Pierre-Alexis Herrault, 02 Oct 2024
Publisher’s note: the supplement to this comment was edited on 8 October 2024. The adjustments were minor without effect on the scientific meaning.
Dear RC1,
Please our responses in the files attached.
Thank you again for your work
Best Regards
Pierre-Alexis Herrault
-
AC1: 'Reply on RC1', Pierre-Alexis Herrault, 02 Oct 2024
-
RC2: 'Comment on egusphere-2024-1935', Anonymous Referee #2, 02 Sep 2024
This study investigates the greening/browning trends in low mountain ranges with a case study in the Vosges mountains in France. The study goes beyond that and tries to explain these trends with topographical variables and snow cover dynamics in a spatially-explicit way. Under climate change, I think this research is timely and important. However, my main concern is the choice of the NDVI-driven index. I strongly think that the maximum NDVI is a very limited index in relating to productivity especially given that there is strong evidence of the changes to the whole NDVI curve (ie. plant growth dynamics) under climate change due to advanced snow-melt and thus, earlier plant growth. Here, usage of the maximum NDVI is not ecologically justified and disentangled from the advanced phenology. I suggest replicating the study with the time-integrated NDVI. Moreover, the article sections, especially the introduction and the discussion are very long and need restructuring to reflect more on the study questions.
Point-to-point comments:
“Climate Change” I do not think that it needs to be capital letters here.
In general, the introduction is very long and lacks structuring and flow. More importantly, the reader is not very well prepared for the concerned research questions, except the first one. For example, topographic effects are not very well hypothesized as well as different effects behind snow cover dynamics. Much more emphasis was put on the land-use changes, however we do not see the relevance of this emphasis to the research questions. I think the introduction needs to be re-written to better highlight the motivation in tackling these research questions and the hypotheses behind.
L35-55: this part is very long and hard to follow. Authors should consider significant shortening. If the purpose of this part is to emphasize contrasting hypotheses, authors might want to consider creating a table with expected responses and matching mechanisms, for example.
L48: Adaptation to what? Also it is not clear how functional traits relate here especially in the context of given example. Do you mean “specific leaf area”?
L66: The connection with the first part of the introduction and phenology is very loose. Authors might even consider mentioning productivity and phenology relationship earlier.
L74: This paragraph is also very long and loosely connected to the previous one. Instead of explaining MODIS, its products, and indices, I suggest that authors consider emphasizing more the reasons behind looking at these products and not the products per se. Products can be mentioned at the methods section.
L94: So far, the emphasis was on climate change but now the questions are based on “warming”. Climate change also involves drought, especially in the mountains. This question falsely sets expectations that the authors will be able to disentangle warming from drought effects.
Methods:
I highly appreciate the detail given for the methods (except the study area section). The study sounds reproducible, however I would like to see the code to assure this. However, the github link does not open.
2.1 Study area: This part can be significantly shortened.
L113: Species names should be written in italic and the first letter of the genus name in capital.
It is not clear why 900m was selected as a threshold.
L142: Usage of the maximum NDVI was never discussed or justified especially over the usage of time-integrated NDVI (ie. area under the NDVI curve) as a proxy of productivity. I think that for the research questions and phenology-productivity relationship time-integrated NDVI seem more suitable to me. Why did the authors choose using maxNDVI?
Figure 2: In this figure, I would also add an illustration of the NDVI curve and show the indices calculated.
L180: Is this threshold determined for this study only? Can you show the distribution on the appendix?
L188: Interpretation of the browning will be hard (I prefer browning than “negative greening”). The browning can be simply due to advanced phenology. (ie. plants grow more in the early season and thus, the growth does not produce the usual peak) The fact that the max greening is less would not mean that there is an overall browning trend to me. That is why in fact I strongly suggest the usage of the time-integrated NDVI.
Results:
Figure 3 is great! One small improvement could be indicating the percentages (or directly giving the percentages) of pixels for herbs vs shrubs comparison.
Figure 4: Colors are not needed. Instead of at the bottom/top indications, I suggest using letters (A/B).
Figure 5: Perhaps add on the y-axis “Probability of browning” and order the variables same as in the Fig.4
On the figures, sometimes abbreviations and sometimes full variable names were given. I suggest harmonizing them by giving the full names at all times.
L221: Where does the information on herbs vs. shrubs come from? Moreover, herbs vs. shrub comparison was not very highlighted in the introduction. What were the hypotheses behind?
L230: Important predictor of what? Browning? This needs to be clearer in the text.
Discussion:
The discussion can be shortened.
Overall, I am not convinced if the trends that we see are driven by the earlier start of the season. That is why, I would like the authors to first clarify that point before I provide an extended review on the discussion on the predictors of the trends. I have the intuition that if authors use time-integrated NDVI the results might change.
L265: Again, I think this is related to changes in earlier season/greening start.
L275: When methodological improvement suggested like here, reader immediately expects to be applied instead of the chosen method.
L277: Are they all looking at the same NDVI index? Ie. maxNDVI?
L293: “do not” instead of “don’t”
There are many results. I suggest that the reader would greatly benefit from a figure where the results are summarized like the one on the following article:(‘Figure 6)
Wang, H., Liu, H., Cao, G., Ma, Z., Li, Y., Zhang, F., Zhao, X., Zhao, X., Jiang, L., Sanders, N.J., Classen, A.T. and He, J.-S. (2020), Alpine grassland plants grow earlier and faster but biomass remains unchanged over 35 years of climate change. Ecol Lett, 23: 701-710. https://doi.org/10.1111/ele.13474
Citation: https://doi.org/10.5194/egusphere-2024-1935-RC2 -
AC2: 'Reply on RC2', Pierre-Alexis Herrault, 02 Oct 2024
Publisher’s note: the supplement to this comment was edited on 8 October 2024. The adjustments were minor without effect on the scientific meaning.
Dear RC2,
Please our responses in the files attached.
Thank you again for your work
Best Regards
Pierre-Alexis Herrault
-
AC2: 'Reply on RC2', Pierre-Alexis Herrault, 02 Oct 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 212 | 82 | 130 | 424 | 13 | 14 |
- HTML: 212
- PDF: 82
- XML: 130
- Total: 424
- BibTeX: 13
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1