the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Apr 2024
Using a convection-permitting climate model to predict wine grape productivity: two case studies in Italy
Abstract. Viticulture is tied to climate, it influences the suitability of an area, yield and quality of wine grapes. Therefore, traditional wine-growing regions could be threatened by a changing climate. Italy is at-risk being part of the Mediterranean climatic hotspot and judged in 2022 the second-largest exporter of wine worldwide. The article explores the potential of climate models to predict wine grape productivity at local scale. To this end, both single and multi-regression approaches are used to link productivity data provided by two Italian wine consortia with bioclimatic indices. Temperature and precipitation-based bioclimatic indices are computed by using the observational dataset E-OBS, the high-resolution climate reanalysis product SPHERA, and both the Regional and the Convection-permitting Climate Model (RCM and CPM). The potential of CPMs to represent the impact of climate variability on wine grape productivity at local scale in Italy is evaluated. The results indicate high correlations between some bioclimatic indices and productivity. Climate models appear to be a useful tool to explain productivity variance, however, the added value of CPM, became evident only when precipitation-based indices are considered. This assessment opens the path for using climate models, especially at convection-permitting scale, to investigate future climate change impact on wine production.
- Preprint
(2120 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 21 May 2024)
-
RC1: 'Comment on egusphere-2024-941', Anonymous Referee #1, 21 Apr 2024
reply
General comments:
- This study investigates correlations between wine productivity and bioclimatic indices, calculated based on observations, reanalysis data, RCM, and CPM. Temperature- and precipitation-based indices explain the changes in wine productivity in two consortia. The simulation output from climate models is validated at a local scale and is expected to predict productivity using highly correlated indices. It provides a valuable contribution to understanding the variability impact of the climate on viticulture. For this reason, this work should be published after several improvements are made. The written text should be proof-checked. The presentation of the work can be much improved regarding the description of the dataset and interpretation of the results (see commnets below).Specific comments:
- The abstract can be sharp and more specific. Please specify "some bioclimate indices" in line 15. It should also include a discussion on two cases, highlighting why it is necessary to use CPM.
- a good explanation of frost risks on line 31.
- line 33, please explain "important shifts" in "important shifts in viticulture suitability".
- line 64, "the driving RCM simulation" what drives RCM simulation
- Figure 1 does not show clearly the scale, north direction, and regions (FRA, LOM, MON). The zooming-in parts are not clearly shown in the map.
- line 95, okay, it is a fair assumption
- line 98, "thus contextualise this work within the broader framework of previous studies", what specific framework? is it necessary to mention this?
- line 107, "the Welch's t-test proves that both consortium distributions are part of the regional population", why this test can prove this conclusion here. please simply explain. What are the variables used in the table. please explain and not directly use terms from the code.
- line 154, "more than 3000 degrees per day", please explain if the unit is correct. also the same unit used below
- line 160, please explain the unit
- line 173, by "occurrence", do you count the frequency of frost days per year?
- line 200, The remapping strategy is the key to the workflow. Please explain the remapping strategy (average, weighted average, (fraction-) interpolation). How are datasets with different resolutions (OBS: 11 km, reanalysis: 2.2 km, RCM: 12.5 km, CPM: 2.5 km) interpolated onto the same grid? What happens when grids from different datasets do not overlap?
- line 200-210, what are the temporal resolutions of the datasets? i expect the value of accumulative variables to depend on the temporal resolution.
- the CNI indices looks like an accumulative value, but the suitable class range is within 12 -18 degrees.
- line 121, what do you mean by "common period"
- line 215, "SPHERA and E-OBS time series together provide a range within which the CPM and the RCM time series are expected to fall, 215 similar to a ‘confidence interval’." I understand that observations and reanalysis data are taken as a reference. I don't understand why observations and reanalysis data "provide a range" since observations and reanalysis data are not statistical tests. i don't see why CPM and RCM should fall into this range. Please rephrase.
- line 233, please rephrase the sentence.
- line 236, "The best subsets regression technique" what do you mean by subsets regression technique, why it is the best
- the multi-regression part can be improved by specifically describing how the bioclimatic data is being processed.
- line 249, could you explain "the variance explained by the MR model"? variance of what, what does "explain" the variance mean? you mean the model is able to predict the variance of the productivity data to some degree (percentage). The terms here are vague. This applies to similar terms used later on.
- line 253- 255, sorry i couldn't understand what you mean here, "evaluate biases"; "could lead to biases"
- line 255- 257, what do you mean "common period", are these variables different from your "ten indices"? if it is the statistics over the whole period, please specify that.
- line 255 - 257, since the validation part is important to show the validity of the climate model, why not put it in the main text?
- line 259, "Further investigations highlighted that this temperature fall affects the entire TOS" what do you mean this temperature fall affects the entire TOS? does this decreasing temperature have an influence on the TOS region? please rephrase, make it clear that the indices in the TOS region are influenced by this.
- line 264-269, I suggest moving this part before the underestimated temperature part.
- line 265, "reproduce the same variability", high correlations does not mean "variability". please rephrase.
- line 266, Table A3, again What are the variables used in the table. please explain and not directly used the term from the code.
- line 268, "The Welch's t-test confirmed that E-OBS is closer in mean value to RCM than CPM simulations." please rephrase.
- line 297, i don't understand what you mean by "the long-term component."
- Table A5, Sen's slope needs explanation to the reader. how is the slope calculated? In Table A6, why is only productivity from FRA shown?
- line 298-299, "RCM presents a statistically significant outcome also for TnRest". This statement seems to be a bit vague.
- line 302, could you specify what resolution you used for "the regional scale", just for the comparison.
- In FRA (Table 3, line 342), could you explain why MR picks up GSP, a precipitation-based indice, in RCM instead of CPM?
- line 379, "in a small number of cases". do you mean a small number of indices?
- It would be beneficial if the authors could include a discussion on future studies concerning how precipitation-based indices become essential with climate change.
Technical corrections:
- line 7, "..., it influences the suitability...", please clarify "it" and rephrase the sentence.
- line 8, "at risk" no hyphen
- line 12, delete "by"
- line 16, "..., however, .." split the sentence.
- line 50, remove "the" "the wine consortia"
- line 53, "Thanks to" informal
- line 53, "without the need for parameterisation," this is inaccurate. most of the parameterizations are turned off in CPM models.
- line 64, "added-value" no hyphen
- line 71, "as well as the hectares devoted to viticulture" can be concise: "planting area"
- line 73, are "LOM" and "TOS" used a lot later in the text, if it is not, you may consider not use these acronyms since there are already many.
- line 77, "thanks to" same as the above
- line 79, same as the comment in line 73
- line 94, "the grape yield", remove "the"
- line 98, , after e.g. "e.g.,"
- line 99, "the productivity", remove "the"
- line 98, remove "the" before productivity
- line 99, remove "the"
- line 100, provide data of ...
- line 100, "the area devoted to vines" please rephrase this.
- line 100 -104, missing "the"
- line 103, and at the national scale.
- I will stop commenting on the usage of articles.
- line 689, whit? Do you mean "with"
- line 129, delete "the period"
- line 218, "RSME" - RMSE, also, explain what it is when you mention it the first time.
- line 218, "the percentage differences of RMSE with the mean of the reference" this is the normalized RMSE
- line 219, remove "the". I will stop commenting on the usage of articles.
- In all line plots, please change your legends indicating the right datasets (dashed, solid)
- The ticks and legends are barely visible in the plots of the appendix.
- line 295, "for some of the temperature-based indices"
- line 295, "few cases"
- line 346, "in other datasets"Citation: https://doi.org/10.5194/egusphere-2024-941-RC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 131 | 33 | 14 | 178 | 9 | 7 |
- HTML: 131
- PDF: 33
- XML: 14
- Total: 178
- BibTeX: 9
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1