Mette, T.; Brandl, S.; Kölling, C. Climate Analogues for Temperate European Forests to Raise Silvicultural Evidence Using Twin Regions. Sustainability2021, 13, 6522.
Mette, T.; Brandl, S.; Kölling, C. Climate Analogues for Temperate European Forests to Raise Silvicultural Evidence Using Twin Regions. Sustainability 2021, 13, 6522.
Mette, T.; Brandl, S.; Kölling, C. Climate Analogues for Temperate European Forests to Raise Silvicultural Evidence Using Twin Regions. Sustainability2021, 13, 6522.
Mette, T.; Brandl, S.; Kölling, C. Climate Analogues for Temperate European Forests to Raise Silvicultural Evidence Using Twin Regions. Sustainability 2021, 13, 6522.
Abstract
Climate analogues provide forestry practice empirical evidence of how forests are managed in “twin” regions, i.e. regions where the current climate is comparable to the expected future climate at a site of interest. But the uncertain future climate creates uncertainty in how to adapt the forests. We therefore investigate how the uncertainty in future climate affects tree species suitability and whether there is a common underlying pattern. Like most studies we employ different ensemble variants of RCP 4.5 and 8.5. But instead of focusing on a single point in future time, we resolve each variant in a climate trajectory from 2000 to 2100. We calculate climatic distances between the climate trajectories of our site of interest and the current climate in Europe, generating maps with twin regions from 2000 to 2100. Forest inventories from the twin regions allow us to trace the changes in the prevalence of 23 major tree species. We find that it is not the direction but rather the velocity of the change that differs between the scenarios. We use this pattern to propose a tree species suitability concept that integrates the uncertainty in future climate. Twin regions provide further information on silvicultural practices, pest management, product chains etc.
Keywords
climate analogue; climate change; model ensemble; twin region; analogue region; national forest inventories; species suitability; forest adaptation; forestry practice; Europe
Subject
Environmental and Earth Sciences, Atmospheric Science and Meteorology
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
4 April 2021
Commenter:
Patrick Grenier
The commenter has declared there is no conflict of interests.
Comment:
This paper is very interesting, at it strives to produce down-to-earth and useful reflexions for forestry under climate change. From a climate science perspective, the methodology considers most technical aspects involved in the computation of spatial analog trajectories. Some suggestions would be the following :
1) The authors worried about interpretation, notably by quoting "The appropriateness of a specific analogy in a specific situation […] does not concern the number of similarities two objects share but rather the significance of the similarities." (Glantz, 1988). This is interesting, and I also encourage the authors to have a look at the related work of P. Bartha (https://plato.stanford.edu/entries/reasoning-analogy/).
2) Traditionnaly, climate analogs have been divided into two categories : spatial and temporal (Mearns et al., 2001, IPCC chapter 13, "Climate scenario development"). It would be interesting to see the paper connecting with this terminology.
3) The authors have used already-bias-adjusted EURO-CORDEX simulations. Because this adjustement is not necessarily done at the same temporal resolution as that of the indices used for assessing similarity (three-month JJA T and Pr averages), it would be worth verifying whether the spatial analog trajectory starts from the target itself (Roth) in 1991-2010. In a recent paper ("The issue of properly ordering climate indices calculation and bias correction before identifying spatial analogs for agricultural applications"; https://www.sciencedirect.com/science/article/pii/S2405880719300639), myself and colleagues proposed the "self-analog test" as a standard to verify that the bias adjustement details do not prevent from having a fully meaningful spatial analog trajectory. The authors mention in the Results section that "The 2000 twin regions (grey) cover the vicinity of Roth itself", and this is well supported by their figures. I guess this can count as an equivalent of the self-analog test, or maybe as a less stricter variant of it. But it would be interesting to verify whether Roth is its own best analog for the recent-past period. And by the way, I think there is a small error in the descriptions of the indices, as at some point "June-July-August" is associated with "meteorological winter".
4) The authors show that uncertainty on future climate change is reflected into uncertainty on the closest spatial analogs identified. For the urban context, Hallegatte et al., 2007 ("Using Climate Analogues for Assessing Climate Change Economic Impacts in Urban Areas") showed how uncertainty on analogs illustrates the uncertainty on the best way to adapt to a specific problematic. Macro-level methodological links with this paper would be interesting.
5) In the Discussion section, RCP8.5 was subjectively linked with "pessimist" and RCP4.5 subjectively linked with "optimist". Such subjective labels look reasonable to me, but it would be interesting to open the discussion to some "utopian" RCP2.6. If adjusted simulations of RCP2.6 had been used, what would we have obtained ? Maybe the same spatial direction of change, but just with a slower pace ? Maybe even a turnover of the trajectory at the end of the century ? This could suggest some impacts will occur and some adaptation will be needed, even with the most drastic CO2 reductions conceivable.
6) In the Introduction section, we see the following segment "And of not reaching forestry at its basis? We do not share this point of view as [...]". I guess this responds to some debate within the forest science community. But from outside this science, it is not clear what this debate refers to, or what exactly is at stake (?) I think some expansion or clarification on that matter would be welcome.
The commenter has declared there is no conflict of interests.
Comment:
Dear Mr. Patrick Grenier,
thanks for sharing your expertise and providing new perspectives for our research! On behalf of my co-authors and me, I would like to use this platform for a reply to your comments appreciating it as an open science forum.
1. [P. Barta, analog and analog reasoning on Stanford Encyclopedia of Philosophy]: a true jewel and encouragement for any scientist to reflect her/his own research philosophy!
2. [differentiation temporal <=> spatial analogues]: We are aware of that differentiation, but not of that particular reference. We draw our understanding from the review by Ford et al. (2010, c.f. references): “Analogues can be spatial in nature, where insights about a problem, condition, and outcome are drawn from a comparable region or location to the target where they are well understood, or temporal, where analysis of past conditions is used to develop understanding of the present day and make inferences about the future.” The reasons why we have not specified temporal analogues, are (1) not to confuse “new” readers by a deviation not particuluarly relevant to our work, and (2) because in our understanding temporal analogues play a minor role in searching modern solutions to climate change problems. When searching “climate analogue” on google (c) only 2 out of the first 20 hits employ temporal analogues: one on the Pliocene/ Eocence climate era (Burk et al. 2018, PNAS), one using recent climate anomalies to predict future park visits in Canada (Hewer & Gough, 2019, AMS). All others were spatial analogues. The statement in Mearny et al. (2001, IPCC) that spatial analogues are “severely restricted by the frequent lack of correspondence be-tween other important features (both climatic and non-climatic) of a study region and its spatial analogue” appears strange to us. For most applications – for instance, snow conditions for skiing, an example mentioned by the very reference – spatial analogues can provide valuable input in the search of solutions, e.g. planning of water reservoirs for snow cannons (we are no experts on that though). Any temporal analogue would quickly have to go back at least to the Holocene climate optimum if not Pliocene/ Eocene with doubtable usefulness in mitigating today’s challenges of climate change.
3. [climate scenario data, bias adjustment, self-analog test]: Indeed, the decisions on the climate scenario data and methods were the hardest ones. Our main problem was the incongruency of climate data sets. The scenario data had been adjusted with MESAN reanalysis data at ~22 km resolution, resulting in relatively precise data sets but at a low resolution. Being adjusted on a monthly basis there is no error in the seasonal aggregated values (at least for the adjustment period). However, the data deviate from the official reference of the German Weather Service (which is only available over Germany, resolution 1 km), and also from the final reference data set we decided on, the CHELSA data (worldwide, resolution 1 km). As the adjusted climate scenario data were already relatively precise we could not decide for a while whether or not to correct the small remaining offset to the CHELSA data, and implemented both solutions. But we noticed that the offset often resembled a topographic effect, i.e. a negative temperature offset was related to a positive precipitation offset and vice versa. That’s plausible as the MESAN resolution of 22 km may integrate over a considerable altitude range. At this point we decided to give preference to solution that corrects the offset of the climate scenario data to the CHELSA data. The correction procedure makes a self-analog test unnecessary, although double-checking is possible by the grey-coloured 2000 twin regions, and a control summary table. Being one of the few studies dealing with the technical challenges in climate data for analogues we took the liberty of including your reference in our methods section of the revised manuscript.
4. [Hallegatte et al. 2007, multiple analogs = uncertainty?]: The idea is very good, and we are on the way to implement it. At the point where foresters seek new silvicultural options from analogue regions, they must study the peculiarities of each region closer. For instance, if they want to learn about pubescent oak silviculture they should visit a twin region where this tree species is very common. They might also choose other criteria like the prevailing geological substrate, management intensity/ type etc. In the methods section we wrote: “Once the twin regions are defined, they can be studied in more detail in terms of geology, soils, landscape etc. to find the most comprehensive possible match.” In the end, city analogues (Hallegatte et al. 2007) and forest analogues may be based on very different criteria and look at very different parameters but, unless ending a “novel climate”, each analogue region offers a new option to learn.
5. [RCP 2.6]: For the chosen parameters and climate models, the RCP 2.6 analogue regions have the furthest distance in 2080 where temperature peaks. Analogues tend to be dominated by the far less certain changes in precipitation and one must be careful not to underestimate the effect of decadal anomalies. For the RCP 2.6 the main challenge lies in catching up with the already accumulated “adaption debt”, “climate debt” (sensu Johnstone et al. 2016).
6. [has science reached the forestry basis?]: It is the constant reproach of the ivory tower of science – we have not found anything in scientific publications neither in newspapers … We guess it is a rather internal debate and deleted the phrase in the revised manuscript.
We encourage comments and feedback from a broad range of readers. See criteria for comments and our Diversity statement.
Commenter: Patrick Grenier
The commenter has declared there is no conflict of interests.
1) The authors worried about interpretation, notably by quoting "The appropriateness of a specific analogy in a specific situation […] does not concern the number of similarities two objects share but rather the significance of the similarities." (Glantz, 1988). This is interesting, and I also encourage the authors to have a look at the related work of P. Bartha (https://plato.stanford.edu/entries/reasoning-analogy/).
2) Traditionnaly, climate analogs have been divided into two categories : spatial and temporal (Mearns et al., 2001, IPCC chapter 13, "Climate scenario development"). It would be interesting to see the paper connecting with this terminology.
3) The authors have used already-bias-adjusted EURO-CORDEX simulations. Because this adjustement is not necessarily done at the same temporal resolution as that of the indices used for assessing similarity (three-month JJA T and Pr averages), it would be worth verifying whether the spatial analog trajectory starts from the target itself (Roth) in 1991-2010. In a recent paper ("The issue of properly ordering climate indices calculation and bias correction before identifying spatial analogs for agricultural applications"; https://www.sciencedirect.com/science/article/pii/S2405880719300639), myself and colleagues proposed the "self-analog test" as a standard to verify that the bias adjustement details do not prevent from having a fully meaningful spatial analog trajectory. The authors mention in the Results section that "The 2000 twin regions (grey) cover the vicinity of Roth itself", and this is well supported by their figures. I guess this can count as an equivalent of the self-analog test, or maybe as a less stricter variant of it. But it would be interesting to verify whether Roth is its own best analog for the recent-past period. And by the way, I think there is a small error in the descriptions of the indices, as at some point "June-July-August" is associated with "meteorological winter".
4) The authors show that uncertainty on future climate change is reflected into uncertainty on the closest spatial analogs identified. For the urban context, Hallegatte et al., 2007 ("Using Climate Analogues for Assessing Climate Change Economic Impacts in Urban Areas") showed how uncertainty on analogs illustrates the uncertainty on the best way to adapt to a specific problematic. Macro-level methodological links with this paper would be interesting.
5) In the Discussion section, RCP8.5 was subjectively linked with "pessimist" and RCP4.5 subjectively linked with "optimist". Such subjective labels look reasonable to me, but it would be interesting to open the discussion to some "utopian" RCP2.6. If adjusted simulations of RCP2.6 had been used, what would we have obtained ? Maybe the same spatial direction of change, but just with a slower pace ? Maybe even a turnover of the trajectory at the end of the century ? This could suggest some impacts will occur and some adaptation will be needed, even with the most drastic CO2 reductions conceivable.
6) In the Introduction section, we see the following segment "And of not reaching forestry at its basis? We do not share this point of view as [...]". I guess this responds to some debate within the forest science community. But from outside this science, it is not clear what this debate refers to, or what exactly is at stake (?) I think some expansion or clarification on that matter would be welcome.
Looking forward for an official publication.
Commenter:
The commenter has declared there is no conflict of interests.
thanks for sharing your expertise and providing new perspectives for our research! On behalf of my co-authors and me, I would like to use this platform for a reply to your comments appreciating it as an open science forum.
1. [P. Barta, analog and analog reasoning on Stanford Encyclopedia of Philosophy]: a true jewel and encouragement for any scientist to reflect her/his own research philosophy!
2. [differentiation temporal <=> spatial analogues]: We are aware of that differentiation, but not of that particular reference. We draw our understanding from the review by Ford et al. (2010, c.f. references): “Analogues can be spatial in nature, where insights about a problem, condition, and outcome are drawn from a comparable region or location to the target where they are well understood, or temporal, where analysis of past conditions is used to develop understanding of the present day and make inferences about the future.” The reasons why we have not specified temporal analogues, are (1) not to confuse “new” readers by a deviation not particuluarly relevant to our work, and (2) because in our understanding temporal analogues play a minor role in searching modern solutions to climate change problems. When searching “climate analogue” on google (c) only 2 out of the first 20 hits employ temporal analogues: one on the Pliocene/ Eocence climate era (Burk et al. 2018, PNAS), one using recent climate anomalies to predict future park visits in Canada (Hewer & Gough, 2019, AMS). All others were spatial analogues. The statement in Mearny et al. (2001, IPCC) that spatial analogues are “severely restricted by the frequent lack of correspondence be-tween other important features (both climatic and non-climatic) of a study region and its spatial analogue” appears strange to us. For most applications – for instance, snow conditions for skiing, an example mentioned by the very reference – spatial analogues can provide valuable input in the search of solutions, e.g. planning of water reservoirs for snow cannons (we are no experts on that though). Any temporal analogue would quickly have to go back at least to the Holocene climate optimum if not Pliocene/ Eocene with doubtable usefulness in mitigating today’s challenges of climate change.
3. [climate scenario data, bias adjustment, self-analog test]: Indeed, the decisions on the climate scenario data and methods were the hardest ones. Our main problem was the incongruency of climate data sets. The scenario data had been adjusted with MESAN reanalysis data at ~22 km resolution, resulting in relatively precise data sets but at a low resolution. Being adjusted on a monthly basis there is no error in the seasonal aggregated values (at least for the adjustment period). However, the data deviate from the official reference of the German Weather Service (which is only available over Germany, resolution 1 km), and also from the final reference data set we decided on, the CHELSA data (worldwide, resolution 1 km). As the adjusted climate scenario data were already relatively precise we could not decide for a while whether or not to correct the small remaining offset to the CHELSA data, and implemented both solutions. But we noticed that the offset often resembled a topographic effect, i.e. a negative temperature offset was related to a positive precipitation offset and vice versa. That’s plausible as the MESAN resolution of 22 km may integrate over a considerable altitude range. At this point we decided to give preference to solution that corrects the offset of the climate scenario data to the CHELSA data. The correction procedure makes a self-analog test unnecessary, although double-checking is possible by the grey-coloured 2000 twin regions, and a control summary table. Being one of the few studies dealing with the technical challenges in climate data for analogues we took the liberty of including your reference in our methods section of the revised manuscript.
4. [Hallegatte et al. 2007, multiple analogs = uncertainty?]: The idea is very good, and we are on the way to implement it. At the point where foresters seek new silvicultural options from analogue regions, they must study the peculiarities of each region closer. For instance, if they want to learn about pubescent oak silviculture they should visit a twin region where this tree species is very common. They might also choose other criteria like the prevailing geological substrate, management intensity/ type etc. In the methods section we wrote: “Once the twin regions are defined, they can be studied in more detail in terms of geology, soils, landscape etc. to find the most comprehensive possible match.” In the end, city analogues (Hallegatte et al. 2007) and forest analogues may be based on very different criteria and look at very different parameters but, unless ending a “novel climate”, each analogue region offers a new option to learn.
5. [RCP 2.6]: For the chosen parameters and climate models, the RCP 2.6 analogue regions have the furthest distance in 2080 where temperature peaks. Analogues tend to be dominated by the far less certain changes in precipitation and one must be careful not to underestimate the effect of decadal anomalies. For the RCP 2.6 the main challenge lies in catching up with the already accumulated “adaption debt”, “climate debt” (sensu Johnstone et al. 2016).
6. [has science reached the forestry basis?]: It is the constant reproach of the ivory tower of science – we have not found anything in scientific publications neither in newspapers … We guess it is a rather internal debate and deleted the phrase in the revised manuscript.