the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Jun 2026
Simulating the permafrost thermal regime in the Northwestern Antarctic Peninsula from 1950 to 2100
Abstract. Permafrost underlies most of the Antarctic ice-free areas, being crucial for terrestrial ecosystems, influencing pedogenesis, hydrology, geomorphic dynamics, and the carbon biogeochemical cycle. However, uncertainty in the evolution of permafrost temperature is particularly relevant in the Antarctic Peninsula, a climatic hotspot of the continent where an increase in mean annual air temperature of 3.4 ± 1.2 °C has been recorded, together with a rise in both the frequency and intensity of warm-weather episodes. The impact on permafrost is difficult to foresee, due to a scarce monitoring network implemented following the International Polar Year in 2007–2008, with limited temporal coverage that constrains trends evaluation. In this study, we simulate past and future permafrost temperature evolution in the Northwestern Antarctic Peninsula using the CryoGrid Community Model at five permafrost observatories from the University of Lisbon’s network (PERMANTAR). Simulations forced with ERA5 reanalysis reconstruct ground temperature evolution since 1950, revealing a warming trend at all depths, with mean annual ground surface temperature, temperature at the top of permafrost, and mean annual ground temperature warming at rates between 0.19 and 0.29 °C dec⁻¹. Four distinct periods are identified since 1950: early sustained warming (1950–1975), highly variable warming and cooling (1975–2000), a short cooling period with increased snowfall (2000–2015), and intense warming after 2015 that accelerates permafrost temperature increases at 20 m depth, resulting in mean annual ground temperatures above -2 °C. Future projections using the ACCESS‑CM2 model under SSP1‑2.6, SSP2‑4.5, and SSP5‑8.5 reveal progressive permafrost degradation and a widespread transition to positive mean annual ground temperatures at 20 m depth during the 21st century, with timing and magnitude strongly influenced by elevation, snow cover duration, and local changes in ocean–atmosphere interactions. In the SSP1-2.6, warming rates ranging from 0.21 ± 0.01 to 0.23 ± 0.01 °C dec⁻¹ are predicted for the sites, resulting in extensive permafrost degradation at low altitude. Under SSP2‑4.5, all sites are projected to develop positive mean annual ground temperatures at 20 m between 2049 and 2087, with warming rates ranging from 0.31 ± 0.01 to 0.43 ± 0.01 °C dec⁻¹, implying increased permafrost degradation and increased susceptibility of coastal terrestrial ecosystems in the Northwestern Antarctic Peninsula to climate-driven change. Under the SSP5-8.5, with warming rates between 0.46 ± 0.01 and 0.62 ± 0.01 °C dec⁻¹, the degradation of permafrost is predicted to occur more rapidly, with positive mean annual ground temperature values at 20 m reached between 2045 and 2071. The projected warming and associated permafrost loss will change surface and subsurface hydrology, biochemical fluxes and geomorphological processes, impacting the sensitive terrestrial ecosystems.
- Preprint
(3018 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1914', Tomáš Uxa, 28 Jul 2026
-
AC1: 'Reply on RC1', Joana Baptista, 21 Sep 2026
We appreciate the time and effort spent on the revision of the manuscript “Simulating the permafrost thermal regime in the Northwestern Antarctic Peninsula from 1950 to 2100”. The comments and suggestions were very well received, and we have subsequently revised and improved the text and work presented. In the following, we provide point-by point replies to all issues raised. The reviewer comment appears in bold font, our replies in normal font and changes to the text which will be implemented in the revised version of the manuscript are in italics.
On behalf of all authors,
Joana BaptistaReply to the major comments:
Point 1: The simulation of recent past conditions was based on the bias-corrected ERA5 for 1950–2020 (with 10 spin-up loops in 1940–1949). By contrast, the simulation of future conditions was based on the bias-corrected CMIP6 models for 1850–2100, with 10 spin-up loops in 1850–1860 and bias-corrected ERA5 used in 1940–2022. Why did not you apply the modelling strategy used for future to recent past as well? The 100-year model initialization phase running since 1850 would likely be better than the 10 spin-up loops over 1940–1949. Moreover, this would mean better consistency and less work since you would have single time series covering the whole period 1850–2100 and both recent past and future scenarios.
Additionally, was the CMIP6 bias correction based on the bias-corrected ERA5? This is unclear to me.Reply: In the submitted manuscript, the simulation of past evolution followed the approach adopted in Baptista et al. (2025). The simulation of future evolution was implemented as an additional step in the model implementation to the Antarctic Peninsula region. However, as pointed out, a longer spin-up period would better represent the initial thermal conditions, particularly at greater depths. It would also provide a more consistent approach. Therefore, we revised the procedure running a long-term simulation from 1850 until 2100. The analysis of both the past and future evolutions are derived from this long-term simulation.
Regarding the question, “Was the CMIP6 bias correction based on the bias-corrected ERA5 data?”, the answer is yes.
Section 3.4 Long-term simulation of the past and future evolution of permafrost temperatures:
“For this reason, we synthesized a model forcing time series from the site-corrected ERA5 records (referred to as ERA5 in the following), from 1940 to 2020, and monthly CMIP6 output, spanning the entire CMIP6 period from 1850 to 2100.”Point 2: Considering that you did the bias correction only in the case of air temperature, while the rest of the variables retained their uncorrected values, is the resulting dataset internally consistent? For instance, if the air temperature was corrected, while the dew point temperature remained uncorrected, this may affect outputs calculated from these two variables. Likewise, precipitation retained its uncorrected value, but solid and liquid precipitation usually occurs at some specific temperatures, which can also be affected if the temperature is corrected. I have no idea how this might affect the simulation outputs, but the issue is here obvious.
Reply: Following the adaptation of air temperature based on in situ measurements, the dew point T was also adjusted so the relative humidity stays constant, and the total precipitation is partitioned according to bias-corrected air temperature. To clarify the procedure, the explanation was extended in the methods.
Section 3.3.1 ERA5:
“Here, the ERA5-derived air temperature at the four observatories (King Sejong, Papagal, Cierva and Amsler) was corrected following Westermann et al. (2016), in which a linear regression is used to correct the average bias and trend of model forcing (Table 2). Additionally, the 2 m dew point temperature was adapted to maintain constant the relative humidity, and total precipitation was partitioned according to bias-corrected air temperature.”Point 3: In the methods, explain why you modelled ground temperatures specifically at a depth of 20 m. Frequently, you also present ground temperatures from a depth of 10 m, which is much closer to a maximum depth of the boreholes and might therefore have a higher validity. Was this because that is at or below the depth of zero annual amplitude?
Reply: Yes, we aimed to estimate the MAGT at a depth of zero annual amplitude and in most observatories, this is not reached at 10 m depth. The estimates at 10 m depth complement the analysis and allowed a comparison to the records in some of the observatories. We extended the explanation in the methods.
Section 3.4 Long-term simulation of the past and future evolution of permafrost temperatures:
“In the simulation’s output we provide estimates of ground temperatures for a profile of 20 m depth, with vertical resolution varying with depth: 0.05 m from the surface to 1 m, 0.1 m from 1 to 2 m, 0.2 m from 2 to 5 m, and 0.5 m below 5 m depth. These are used to calculate the MAGST and MAGT at 10 and 20 m depth. MAGT at 10 m depth provides an overview of near-surface thermal changes and their potential implications for surface and subsurface hydrology, biogeochemical fluxes, and geomorphological processes. MAGT at 20 m depth provides insight into long-term ground temperature conditions near the depth of zero annual amplitude.”Point 4: For consistency, descriptions of both recent past and future simulations should present the evolution of mean annual air temperatures, mean annual ground surface temperatures, temperatures at the base of the freeze-layer layer (TTOP under permafrost conditions), and mean annual ground temperatures at a depth of 20 m. Some of these temperature metrics are frequently missing, which makes a little sense in terms of consistency and (not only) mutual context. If mean annual ground temperatures at a depth of 10 m are shown as well, I would present them before those from 20 m. Because of heat diffusion, the temperature patterns at 10 m precede those at 20 m, so it would have more logic to proceed in this order.
Reply: We removed the TTOP metric from the analysis and focused on MAGST and MAGT. The analysis of MAGT at 10 m is now presented before the MAGT at 20 m depth. Additionally, Figure 6 (previous Figure 5) now includes the estimates for MAGT at 10 m depth that were previous located in the Appendix E.
Point 5: The discussion is relatively short and in principle deals with results of this study alone, and the context of other studies is provided to a very limited extent. While I am aware that there is currently no analogous study published for Antarctica, there are several studies that assessed short-term trends in permafrost temperatures or active-layer thickness (e.g., Guglielmin and Cannone, 2012; Ramos et al., 2017; Biskaborn et al., 2019; Hrbáček and Uxa, 2020; Hrbáček et al., 2021, 2023, 2025; Baptista et al., 2025) or some several long-term meteorological series (e.g., Oliva et al., 2017). I do not understand why you do not mention your previous study in the discussion (Baptista et al., 2025).
Reply: We agree with the comment and further developed the discussion section with respect to the studies focusing on permafrost temperature on the Western Antarctic Peninsula, especially Biskaborn et al. (2019) and Hrbáček et al. (2023), presenting the rates estimated by the authors.
Section 5.2 Past evolution of permafrost temperature (1950 - 2020):
“At the surface, warming rates were very similar across the five observatories, ranging from 0.26 to 0.30 ± 0.04 ºC dec-1, close to the rate reported by Hrbáček et al. (2023) for the South Shetlands Islands of 0.28 ºC dec-1 (2006-2020).”
(…)
“The cooling recorded between 2000 and 2015, together with increased snowfall that led to persistent snow cover during the thawing season that limited surface warming by increasing albedo and reducing the heat loss, produced a slight reduction in MAGT (<1 ºC) and temperature trends at 20 m depth ranging from -0.05 ± 0.02 ºC dec-1 to -0.01 ± 0.02 ºC dec-1. Biskaborn et al. (2019), in the analysis of the global dataset of permafrost temperatures for the decade between 2007 and 2016, reported temperature trends between -0.4 to 0.8 ºC dec-1 for the Antarctic Peninsula region.”We mentioned the previous study (Baptista et al., 2025) along the methods since it was the first test to the approached followed. Additionally, we also include a reference in the section 5.1 Evaluation of the forcing data discussion the limitations previous found on ERA5 reanalysis data.
Section 5.1 Evaluation of the forcing data:
“In the case of ERA5, the comparison between reanalysis-derived air temperature and in situ measurements revealed a cold bias during the summer season, which limited surface warming and the development of the active-layer thickness observed in borehole records. A similar cold bias was also identified by Baptista et al. (2025) when modelling ground temperature at Barton Peninsula, King George Island. To reduce its impact on the simulations, the ERA5 air temperature data was bias-corrected following the approach proposed by Westermann et al. (2016).”What I also miss in the discussion is a section devoted to the evaluation of the ERA5 and CMIP6 models used as model forcings. Consequently, I would move the Appendix C into the discussion. You may also refer to Baptista et al. (2025) or Kaplan Pastíriková et al. (2025) when evaluating the suitability of ERA5 or reananalyses for permafrost studies in Antarctica in general.
Reply: We agree with the comment and divided the Appendix C between the methods and discussion sections. Regarding the references, Kaplan Pastíriková et al. (2025) and Baptista et al. (2025) are now mentioned in the discussion.
Section 5.1 Evaluation of the forcing data:
“In the case of ERA5, the comparison between reanalysis-derived air temperature and in situ measurements revealed a cold bias during the summer season, which limited surface warming and the development of the active-layer thickness observed in borehole records. A similar cold bias was also identified by Baptista et al. (2025) when modelling ground temperature at Barton Peninsula, King George Island. To reduce its impact on the simulations, the ERA5 air temperature data was bias-corrected following the approach proposed by Westermann et al. (2016).
Other studies have also evaluated the performance of ERA5 in Antarctica. Zhu et al. (2021) assessed near-surface air temperature against observations from 41 weather stations and identified both regional and seasonal biases, including a warm bias inland and difficulties in reproducing local scale variability in areas of complex terrain. Nevertheless, ERA5 performed better than ERA-Interim. Similarly, Jones and Lister (2015) highlighted that the extreme environmental conditions and sparse observational network of Antarctica pose significant challenges for data assimilation, contributing to uncertainties in surface temperature and precipitation estimates. Focusing on the Antarctic Peninsula, Tetzner et al. (2019) reported a seasonal cold bias in 2 m air temperature and an underestimation of wind speed in coastal areas, despite ERA5 providing a generally good representation of the prevailing wind regimes. On the eastern side of the Antarctic Peninsula, (Kaplan Pastíriková et al., 2025) validated ERA5-Land reconstructed air temperatures against observations from three automated weather stations (AWSs) on James Ross Island for the period 2004-2021. Their analysis showed that MAAT derived from ERA5-Land had lower interannual variability (σ=0.96 ºC) when compared to the observations (σ between 1.32 to 1.35 ºC). Moreover, the mean monthly temperatures during summer and autumn differed substantially from the observations, with ERA5-Land underestimating the air temperatures recorded, as observed at the PERMANTAR observatories. This resulted in RMSE values of approximately 3 ºC and biases ranging from -1.6 to -0.9 ºC. The authors also reported excessive snowfall in ERA5-Land, associated with an overestimation of snow accumulation that produced persistent snow depths of 3-24 m and affected ground temperatures during summer. This overestimation was also observed at the PERMANTAR observatories, requiring the adjustment of the snow factor.”Point 6: Consider carefully whether the content of the appendices really needs to placed in them. I think that most of the appendices could be involved in the main body of the manuscript without making it fragmented.
Reply: Following your suggestions, the previous Appendix C (Evaluation of CMIP6 models) was incorporated into the main structure of the manuscript. Information related to the site characteristics and observational period from the Appendix A were also incorporated into the main structure in a summarized way. The figure presented on Appendix E was moved to the Section 4.3 and incorporated on the Figure 6 (previous Figure 5).
Reply to the minor comments:
P1L13: State the period over which the mean annual air temperature increase of 3.4±1.2 °C was recorded.Reply: The sentence was rewritten for clarity.
“a climatic hotspot of the continent where an increase in mean annual air temperature of 3.4 ± 1.2 ºC since 1950 has been recorded”
P1L18: Specify the depth for the mean annual ground temperature. Is it 20 m?
Reply: Corrected.
“Simulations forced with ERA5 reanalysis were used to reconstruct ground temperature evolution since 1950, revealing a warming trend at all depths, with mean annual ground surface temperature, and mean annual ground temperature at 10 and 20 m depth warming at rates of 0.29 ± 0.04 ºC dec⁻¹, 0.24 ± 0.01 ºC dec⁻¹ and 0.18 ± 0.01 ºC dec⁻¹, respectively”
P1L20: In contrast to the above, it is unnecessary to specify the depth of 20 m here because it is a general statement.
Reply: Corrected.
“Four distinct periods are identified since 1950: early sustained warming (1950–1975), highly variable warming and cooling (1975–2000), a short cooling period (2000–2015), and intense warming after 2015 that accelerated permafrost temperature increase, resulting in mean annual ground temperatures above -2 ºC at Papagal, SKO Station and Amsler observatories.”
P2L42–43: Again, state the period over which the mean annual air temperature increase of 3.4±1.2 °C was recorded. Considering that mean annual air temperature can exhibit substantial variations from year to year, is it appropriate to use a single year as a baseline value?
Reply: The sentence was rewritten for clarity and the reference to a single year removed.
“An increase in mean annual air temperature (MAAT) of 3.4 ± 1.2 ºC since 1950 has been recorded at Esperanza Station”
P2L50: Change (Obu et al., 2020) to Obu et al. (2020).
Reply: Corrected.
P2L57–58: In the aims of the manuscript, be specific and clearly define the time frames for both the recent past and the future modelling periods.
Reply: The sentence was rewritten for clarity.
“In the present study, the CCM was implemented for five observatories of the PERMANTAR network (http://permantar.weebly.com) distributed along the Northwestern Antarctic Peninsula, specifically, to validate the models against observed permafrost temperature data, simulate past conditions from 1950 to 2020 using reanalysis data, and project the ground temperature evolution from 2020 until 2100 under three Shared Socioeconomic Pathways (SSPs) from the Coupled Model Intercomparison Project (CMIP) (IPCC, 2014; O’Neill et al., 2016).”
P3L61–62: This previous study (Babtista et al., 2025) should be briefly described before the aims of the manuscript, so that the objectives follow logically from it. I think that, in this case, it disrupts the structure of the paragraph devoted to the aims.
Reply: The sentences were rewritten for clarity.
“Our goal is to implement a modelling approach based on the CryoGrid Community Model (CCM) (Westermann et al., 2023), which integrates site-specific characteristics to estimate ground temperature over defined time periods. In Baptista et al. (2025), the performance of the CCM was evaluated for the King Sejong permafrost observatory on Barton Peninsula, King George Island. The results of the simulation from 2020–2022 show that the model successfully represented the site conditions with correlations above 0.8 and MAE ranging from 0.1 to 0.7 ºC. Following the validation of the model, a long-term simulation from 1950 to 2022 was run showing a ground warming trend at 20 m of 0.25 ºC dec⁻¹ followed by an increase in the active layer thickness from 1.5 m in 1950 to 3.5 m in 2022. In the present study, the CCM was implemented for five observatories of the PERMANTAR network (http://permantar.weebly.com) distributed along the Northwestern Antarctic Peninsula, specifically, to validate the models against observed permafrost temperature data, simulate past conditions from 1950 to 2020 using reanalysis data, and project the ground temperature evolution from 2020 until 2100 under three Shared Socioeconomic Pathways (SSPs) from the Coupled Model Intercomparison Project (CMIP) (IPCC, 2014; O’Neill et al., 2016).”
P3L76: Change (Thomas e Tetzner 2019) to (Thomas and Tetzner 2019).
Reply: Corrected.
P4L96: Considering the substantial warming trend, the mean annual air temperature could be reported for a shorter, more recent period to better represent current climatic conditions. Of course, this depends on your intentions.
Reply: In the revised manuscript, we now report the mean annual air temperature at Bellingshausen Station based on the climatological analysis of Turner et al. (2020), who report a MAAT of −2.1 °C for the 1981–2010 period.
“The climate is polar maritime with a MAAT of -2.1 ºC, recorded at Bellingshausen station (16 m a.s.l.) from 1981 to 2010, and annual precipitation of 534 mm, measured at King Sejong Station at 11 m a.s.l. from 2005 to 2019 (Baptista et al., 2024; Turner et al., 2020).”
P5L100–101: Rewrite the sentence “… the highest areas of Hurd Peninsula were deglaciated at 20 ka with most high interfluves becoming ice-free around 14-16 ka …” so that it is clear when these locations were actually deglaciated. Was it at 20 ka or around 16-14 ka?Reply: The sentence was rewritten for clarity.
“With an ice-free area of about 20 km2, the deglaciation of Hurd Peninsula accelerated during the LGM at 20-18 ka with most high interfluves becoming ice-free around 16-14 ka. and the lower valleys around 7.5 ka.”
P5L101: Change 14-16 ka to 16-14 ka.
Reply: Corrected.
P5L106: State the period for which the mean annual air temperature and mean annual precipitation are reported.
Reply: The sentence was changed based on the information published by Bañón et al. (2013).
“ The climate is polar maritime, with a MAAT of -1 ºC and annual precipitation between 900 to 1000 mm, recorded at BAEJCI station (12 m a.s.l.) from 2002 to 2010 (Bañón et al., 2013; Fernández-Fernández et al., 2021).”
P5L115–116: State the period for which the mean annual air temperature and mean annual precipitation are reported.
Reply: Corrected.
“The climate is polar maritime, with a MAAT of -3.2 ºC, recorded at 50 m a.s.l., and annual precipitation ranging from 400 to 1100 mm from 1995 to 2013 (Wilhelm et al., 2016).”
P5L124–125: State the period for which the mean annual air temperature and mean annual precipitation are reported.
Reply: Corrected.
“Under a polar maritime climate, the MAAT is of -1.8 ºC at 8 m a.s.l., with annual precipitation ranging from 714 to 1410 mm from 1995 to 2013 (Wilhelm and Bockheim, 2017).”
P6L127–140: Move the whole section 3.1 to the section 2.2. The details on permafrost temperature and active-layer thickness at individual sites can be presented in the sections 2.2.1 to 2.2.4.
Reply: We appreciate your suggestion. We agree that the values of permafrost temperature and active layer thickness are useful for the study area description. Therefore, these values have been added to the subsections of Section 2.2. Regarding the detailed description of the permafrost network, borehole setup, and measurement depths, we prefer to keep this information in Section 3.1, as it is part of the Materials and Methods and is necessary to understand the subsequent analyses and results.
Section 2.2.1 Barton Peninsula, King George Island:
“In 2020, the King Sejong permafrost observatory, located at 127 m a.s.l., had an active-layer thickness (ALT) of 3 m and a mean annual ground temperature (MAGT) of −1.5 °C at 13 m depth.”
Section 2.2.2 Hurd Peninsula, Livingston Island:
“In 2020, the Papagal permafrost observatory, located at 147 m a.s.l., had an ALT of more than 4 m and a MAGT of −0.7 °C at 4 m depth. The SKO Station, located at 25 m a.s.l., had an ALT of more than 8 m and a MAGT of 0 °C at 8 m depth.”
Section 2.2.3 Cierva Point, Antarctic Peninsula:
“In 2020, the Cierva permafrost observatory, located at 187 m a.s.l., had an ALT of 6 m and a MAGT of −1.2 °C at 15 m depth.”
Section 2.2.4 Amsler Island, Palmer Archipelago:
“In 2020, the Amsler permafrost observatory, located at 67 m a.s.l., had an ALT of more than 9 m and a MAGT of −0.2 °C at 9 m depth.”
P7L151–152. This sentence on the calibration strategy is difficult to understand. Rewrite it even though it is partly obvious from the following text.
Reply: The sentence was rewritten for clarity. Additionally, in response to comments from Reviewers 2 and 3 regarding the division of the observational period for calibration and evaluation, it was updated accordingly.
“Following Baptista et al. (2025), the model calibration for each borehole was performed in two phases: i. the simulations were run for 2021 forced with ground surface temperature, being subsurface parameters (i.e. mineral and water/ice content) adjusted to obtain the best configuration through a sensitive analysis, ii. the best configuration scheme was then used when forcing the simulations with ERA5 data of the same year, but using the CCM surface energy balance scheme, in order to define the snow factor. The model performance was evaluated for each borehole by comparing the results from the beginning of the observational period until 2020 (Table 1).”
P7L156: What is the geothermal heat flux of 50 mWm^-2 based on? Please cite the source.Reply: Thanks for pointing this out. In the initial simulations, we used a geothermal heat flux value of 50 mW m⁻², following the implementation described by Westermann et al. (2023). We have re-run the long-term simulations used to derive the past and future estimates and used the geothermal heat fluxes derived from An et al. (2015) for each observatory, which are now described on the methods. The final result changes were minimal, but it is an important point for improved accuracy.
Section 3.2.2 Model calibration and evaluation:
“Measured ground surface temperature from the PERMANTAR monitoring sites was used as upper boundary input while a geothermal heat flux value obtained from An et al. (2015) and applied as the lower boundary. Geothermal heat flux values for King Sejong, Papagal and SKO Meteo were 73 mWm-2, for Cierva were 61 mWm-2, and for Amsler were 58 mWm-2.”
P7L157: What does the phrase “With the structure set, …” mean?
Reply: The structure relates to the employed configuration of the CryoGrid model, but we very much agree that it is not a good term and have rephrased the sentence for clarity.
“1D simulations were run with different configurations of mineral and water/ice relative volumetric content for each of the five boreholes.”
P7L160: Is the supplement “… using a probe.” necessary?
Reply: The supplement was removed.
P7L164: What do you mean by “relative volume fraction” here?
Reply: The term “relative volume fraction” was removed from the manuscript and replaced simply by the term “volumetric mineral/water/ice content”. This value goes from 0 to 1, which is the same as 0-100%, expressing the percentage of the total volume occupied by each component.
“1D simulations were run with different configurations of mineral and water/ice volumetric content, that goes from 0 to 1, with correspondence to 0 and 100%, for each of the five boreholes.”
P7L171: Indicate that Fub(t) is the energy flux into the uppermost grid cell. The rest of the variables (radiation and heat fluxes) would be better to describe after Eq. (1).Reply: The sentence was rewritten following the suggestion.
“The energy flux into the uppermost grid cell (F_ub (t)) is calculated through Eq. (1):
(1)
F_ub (t)=S_in (t)-S_out (t)+L_in (t)-L_out (t)-Q_h (t)-Q_e (t)
where Sout and Lout are the outgoing short- and longwave radiation and Qh and Qe the sensible and latent heat flux.”P7L175: “The heat conduction based on Fourier’s law is the main mode of heat transport in the subsurface.” is too brief description of the subsurface heat transfer.
Reply: The sentence was further developed.
“The heat conduction based on Fourier’s law is the main mode of heat transport in the subsurface and is calculated as:
(2)
j_hc=〖-K〗_h (∂T(e))/∂z
where K_h [Wm-1K-1] is the thermal conductivity and temperature T is a function of enthalpy e.”P8L183: Change “phenomenologically” to “phenomenological”.
Reply: Corrected.
P8L186: State also that you calculated p-value.
Reply: Corrected.
“As for the first phase, the validation of the surface energy model and snow factor were performed through the comparison between the estimated and measured ground temperatures for 2021 at 4 depths and the analysis of the four statistical measures (bias, r, p-value, MAE and RMSE).”
P9L219: I guess there should be “SPPs” in “The RCPs are used to represent 4 pathways…”
Reply: The sentence was complemented for clarity.
“The RCPs are used to represent 4 pathways of GHG emissions and atmospheric concentrations driven by population size, economic activity, lifestyle, energy used, land use patterns, technology and climate policy associated to SSPs (IPCC, 2014).”
P10L230: Note that the SPP5-8.5 scenario is now considered an extreme case that is rather implausible. You should point this out in the manuscript (here and in the discussion).
Reply: A reference was added to the sentence.
“SSP5-8.5: predicts a trend of human development with substantial investments in education and health, rapid economic growth, and functioning institutions, but based on intensive energy consumption and a fossil-based economy, in which global warming will range from 3 to 5 ºC. It is worth notice that this scenario has been recently consider as implausible due to the trends in the costs of renewables, the emergence of climate policy and recent emission trends (Van Vuuren et al., 2026)”
P10L235: Remove the word “models”.
Reply: If it is the word “models” on line 232, we are referring to the models from CMIP6 designated as this on the documentation.
P10L243–245: Place this paragraph at the end of the whole section 3.3.2.
Reply: Corrected.
P10L253: I would remove “…in the selected PERMANTAR monitoring sites.” because it is redundant here.
Reply: Corrected.
P11L258: The vertical resolution of the model is unclear from this: “… from 0.05 to 0.5 m, below 5 m depth.” Revise it.
Reply: The sentence was rewritten for clarity.
“In the simulation’s output we provide estimates of ground temperatures for a profile of 20 m depth, with vertical resolution varying with depth: 0.05 m from the surface to 1 m, 0.1 m from 1 to 2 m, 0.2 m from 2 to 5 m, and 0.5 m below 5 m depth.”
P11L259: Specify the depths at which you calculated the mean annual ground temperature.
Reply: Corrected.
“These are used to calculate the MAGST and MAGT at 10 and 20 m depth. MAGT at 10 m depth provides an overview of near-surface thermal changes and their potential implications for surface and subsurface hydrology, biogeochemical fluxes, and geomorphological processes. MAGT at 20 m depth provides insight into long-term ground temperature conditions near the depth of zero annual amplitude.”
P11L266–269: Rewrite this so it is more clear that you bias-corrected the period 1850–2100 from CMIP6 using the statistical relationships with ERA5 for the sub-period 1940–2020.
Reply: The sentence was rewritten for clarity.
“For this reason, we synthesized a model forcing time series from the site-corrected ERA5 records (referred to as ERA5 in the following), from 1940 to 2020, and monthly CMIP6 output, spanning the entire CMIP6 period from 1850 to 2100.”
P12L300: Remove the sentence “In the figures, the x-axis spans the period 2000-2100.”
Reply: Corrected.
P13L326: In Table 3, you should indicate that the thermal conductivity relates to the rock.
Reply: Corrected.
P14L354: In Fig. 2, the range seems to be somewhat smaller than -5 to 5 °C.
Reply: The range was presented in more detail.
“Temperatures fluctuate between positive (4 ºC) and negative (-3 ºC) values during DJF and MAM, and remain negative (0 to -5 ºC) during JJA and SON.”
P16L364–380: In addition to the trends, present also the temperatures or their ranges. This is only done at three stations for TTOP and at all of them for MAGT. Report also the values and trends from the mean annual air temperature (and maybe snowfall if both are shown in Fig. 3).
Reply: The sentences were completed with estimated values for 1950 and 2020. References to the TTOP were removed.
“The simulation of ground temperatures since 1950 enabled the reconstruction of past permafrost temperature evolution, revealing a general warming trend in ground temperatures at all depths that follows the warming trend of air temperature (Fig. 4 and Table 7). Air temperature ranged from -5 to -3 ºC during 1950-1960, with a warming rate varying from 0.25 ± 0.03 ºC dec⁻¹ to 0.36 ± 0.05 ºC dec⁻¹, reaching values between -3 to -1 ºC during 2010-2020. At the surface, most observatories showed a warming rate of 0.30 ± 0.04 ºC dec⁻¹, except for SKO Station, where a warming rate of 0.26 ± 0.04 ºC dec⁻¹ was obtained. MAGST during 1950-1960 was estimated to range from -4 to -3 ºC, while during 2010-2020 it ranged from -3 to 1 ºC.”
P16L381: In Table 4, specify the depth for the mean annual ground temperature.
Reply: Corrected.
P17L385: Fig. 3 has a misleading caption because snowfall and air temperature cannot be referred to as ground temperatures.
Reply: The words “of ground temperatures” were removed. Previous Figure 3 is now Figure 4.
“Long-term simulation (January 1950 to December 2020) for the PERMANTAR observatories: annual snowfall (mm water equivalent), mean annual air temperature - MAAT (ºC); mean annual ground surface temperature - MAGST (ºC); mean annual ground temperature at 10 m - MAGT (ºC); mean annual ground temperature at 20 m - MAGT (ºC).”
P18L388: In Fig. 4, I think it is a bit misleading to present the temperature for a single year 1950 because it was exceptionally cold. Is it appropriate to use a single year here, at least for the mean annual ground surface temperature, which has an instantaneous response to air temperature?
Reply: We understand the reviewer’s concern. Figure 5 (previous figure 4) is intended as a spatial comparison. We agree that would be more appropriate to present the estimate for a period, as suggested. Therefore, we have calculated averages for the corresponding decades. This means that the value previously shown for 1950 now represents the average for the period 1950-1960, the value for 2020 represents the average for 2010-2020, and the value for 2100 refers to the period 2090-2100.
P18–19L391–416: Check that it is clearly stated everywhere what temperature the reported trends refer to.
Reply: The sentences were reviewed accordingly.
P18L409: Rewrite the sentence “Despite continued interannual air temperature variability, MAGST fluctuated between -4 to -1 °C.” which does not make complete sense.
Reply: The sentence was rewritten for clarity.
“During this period, MAGST ranged between -4 and 0 ºC, with a cooling rate around -0.47 ± 0.04 ºC dec-1. MAGT at 10 m depth, showed very small variations with values around -3 ºC at Cierva, -2ºC at Amsler, King Sejong and Papagal, and -1 ºC at SKO Station, experiencing a cooling rate of around -0.21 ± 0.06 ºC dec-1. At 20 m depth, the colling rate was minimal (-0.02 ± 0.02 ºC dec-1) with MAGT remaining constant.”
P19L412: It is unclear what temperature the trend of 2.10±0.60 °C refers to.
Reply: Corrected.
“The final years of the series were marked by intense and rapid warming of air temperature (2.14 ± 0.12 ºC dec-1) and a general decrease of snowfall, when compared to the previous period.”
P19L423: You should rather write for 2020–2100. Trends should always refer to a period, not a single year. Additionally, it is unclear what the temperature trend of 0.59±0.22 °C refers to.
Reply: The sentence has been clarified, and the trends are now presented per decade for consistency.
“Under the SSP1-2.6 scenario, an average warming on air temperature of 0.07 ± 0.03 ºC dec-1 is projected for the Northwestern Antarctic Peninsula for the period 2020-2100.”
P20L427: Change “… being the warming more pronounced at the …” to “… with more pronounced warming at the…”
Reply: Corrected.
P20L429: Remove “For most of the observatories, “ Then, the sentence will make more sense.
Reply: Corrected.
P20L434–437: Maybe move this paragraph before the previous one. Definitely, it is better to start with the statement that all stations host permafrost at present, and then describe that it will warm and likely degrade or disappear at some of them.
Reply: Done as suggested.
P20L439: Again, it is unclear what the first trend of 2.59±0.20 °C refers to.
Reply: Corrected. All trends are now presented for the decade for consistency.
“Following the historical trend, the SSP2-4.5 scenario estimates a much stronger MAAT warming rate of 0.31 ± 0.02 ºC dec⁻¹.”
P21L469–470: Change “At Cierva despite a warming rate of 0.51 ± 0.01 °C.dec-1 positive MAGT is projected…” to “At Cierva, despite a warming rate of 0.51 ± 0.01 °C.dec-1, positive MAGT is projected…”
Reply: Corrected.
P21L476: Table 5 has a misleading caption because it presents other temperature variables in addition to the mean annual ground temperature at 20 m.
Reply: The caption was changed.
“Change rates from the simulations of future temperature evolution on the PERMANTAR observatories from 2020 to 2100.”
P22L480: Fig. 5 has a misleading caption because air temperature cannot be referred to as ground temperature.
Reply: The caption was changed.
“Long-term simulation of air and ground temperature (2020 - 2100) on PERMANTAR observatories: mean annual air temperature - MAAT (ºC); mean annual ground surface temperature - MAGST (ºC); mean annual ground temperature at 10 m depth - MAGT (ºC); mean annual ground temperature at 20 m depth - MAGT (ºC).”
P24L505–509: The interpretation that ground temperature warming in the periods 1950–1975 and 2015–2020 was favoured by decreased snowfall and shorter duration of the snow cover contradicts former explanations that thicker and longer snow cover tends to warm the ground. Likewise, cooling simulated for 2000–2015 is difficult to explain by reduced snowfall.Reply: Thicker and longer snow cover are well-known to be key controls in warmer ground conditions in the Arctic, as you point out. However, in very moist and cloudy environments, with low annual thermal amplitude (not very cold winters – mean monthly air temperature ~-5ºC and cool summers air temperatures ~+1ºC), such as the Maritime Antarctic, where permafrost is close to 0 ºC, the effect of snow insulation becomes key in the shoulder seasons and summer. It has been shown by various authors that the ground cooling period between 2000 and 2015 was associated with a progressive development and longer persistence of the snow cover, including during the thawing season (De Pablo et al., 2016; Ramos et al., 2017, 2020). As reported by De Pablo et al. (2016) for Livingston Island, persistent snow cover during summer can limit surface warming by increasing albedo and reducing the heat loss. In contrast, snow accumulation occurring early in the freezing season may favour warmer ground conditions because the snowpack acts as an insulating layer, reducing heat loss from the ground after the thawing season. Therefore, the thermal effect of snow cover depends not only on its thickness, but also on its timing, duration, and seasonal persistence.
We have revised the manuscript to emphasize that the observed ground temperature response is associated with the persistence and seasonal timing of snow cover.“The cooling recorded between 2000 and 2015, together with increased snowfall that led to persistent snow cover during the thawing season that limited surface warming by increasing albedo and reducing the heat loss, produced a slight reduction in MAGT (<1 ºC) and temperature trends at 20 m depth ranging from -0.05 ± 0.02 ºC dec-1 to -0.01 ± 0.02 ºC dec-1.”
P24L513–514: Remove the first sentence in the section 5.2; it is redundant.
Reply: Corrected.
P24L516: Why are the periods 2000–2015 and 2016–2022 discussed in the section entitled “Future evolution of permafrost temperature (2022 - 2100)”?
Reply: They were there to better frame the results. However, we agree and removed the first period and kept the period 2016 to 2022.
“In the three CMIP6 scenarios, the evolution of the MAAT, MAGST and MAGT exhibit one similar period prior to the divergence into distinct warming trends. This period, from 2016 to 2040, is characterized by a rapid increase in temperature. The initial years of this period (2016-2022) correspond to a phase of intensified warming marked by several warm weather episodes documented by (González-Herrero et al., 2022; Gorodetskaya et al., 2023).”
P25L532: Change “Under the more optimistic SSP1-2.6 scenario, …” to “Under the SSP1-2.6 scenario, …”
Reply: Corrected.
P25L539: Add “with values ranging from 1 to 2 °C in 2100.”
Reply: Corrected.
P26L557: Specify to what depth the bias refers to.
Reply: Corrected.
“The bias calculated from measured and simulated ground temperatures at 4 depths (0.4, 1.2, 3.0, and maximum depth of borehole) was below 1 ºC”
P26L574: Change “In the worst-case scenario (SSP5-8.5), …” to “Under the SSP5-8.5 scenario, …”
Reply: Corrected.
P27L582: Change “thermokarst and landslide activity should increase” to “thermokarst and landslide activity will likely increase”
Reply: Corrected.
P31L666: Change “An historical period …” to “A historical period …”
Reply: Corrected.
P31L674: Change “improved coupled between” to “improved coupling between”
Reply: Corrected.
P31L679: Change “a RMSE of 1.33 °C. and a standard deviation” to “a RMSE of 1.33 °C and a standard deviation”
Reply: Corrected.
References:
An, M., Wiens, D. A., Zhao, Y., Feng, M., Nyblade, A., Kanao, M., Li, Y., Maggi, A., and Lévêque, J.: Temperature, lithosphere‐asthenosphere boundary, and heat flux beneath the Antarctic Plate inferred from seismic velocities, J. Geophys. Res. Solid Earth, 120, 8720–8742, https://doi.org/10.1002/2015JB011917, 2015.Bañón, M., Justel, A., Velázquez, D., and Quesada, A.: Regional weather survey on Byers Peninsula, Livingston Island, South Shetland Islands, Antarctica, Antarct. Sci., 25, 146–156, https://doi.org/10.1017/S0954102012001046, 2013.
Baptista, J., Vieira, G., and Lee, H.: Ground surface temperature regimes are controlled by the topography and snow cover in the ice-free areas of Maritime Antarctica, CATENA, 240, 107947, https://doi.org/10.1016/j.catena.2024.107947, 2024.
Baptista, J., Brito Guapo Teles Vieira, G., Manuel De Carvalho Soares Correia, A., Lee, H., and Westermann, S.: Modelling the evolution of permafrost temperatures and active layer thickness in King George Island, Antarctica, since 1950, The Cryosphere, 19, 3459–3476, https://doi.org/10.5194/tc-19-3459-2025, 2025.
Biskaborn, B. K., Smith, S. L., Noetzli, J., Matthes, H., Vieira, G., Streletskiy, D. A., Schoeneich, P., Romanovsky, V. E., Lewkowicz, A. G., Abramov, A., Allard, M., Boike, J., Cable, W. L., Christiansen, H. H., Delaloye, R., Diekmann, B., Drozdov, D., Etzelmüller, B., Grosse, G., Guglielmin, M., Ingeman-Nielsen, T., Isaksen, K., Ishikawa, M., Johansson, M., Johannsson, H., Joo, A., Kaverin, D., Kholodov, A., Konstantinov, P., Kröger, T., Lambiel, C., Lanckman, J.-P., Luo, D., Malkova, G., Meiklejohn, I., Moskalenko, N., Oliva, M., Phillips, M., Ramos, M., Sannel, A. B. K., Sergeev, D., Seybold, C., Skryabin, P., Vasiliev, A., Wu, Q., Yoshikawa, K., Zheleznyak, M., and Lantuit, H.: Permafrost is warming at a global scale, Nat. Commun., 10, 264, https://doi.org/10.1038/s41467-018-08240-4, 2019.
De Pablo, M. A., Ramos, M., Molina, A., Vieira, G., Hidalgo, M. A., Prieto, M., Jiménez, J. J., Fernández, S., Recondo, C., Calleja, J. F., Peón, J. J., and Mora, C.: Frozen ground and snow cover monitoring in the South Shetland Islands, Antarctica: Instrumentation, effects on ground thermal behaviour and future research, Cuad. Investig. Geográfica, 42, 475–495, https://doi.org/10.18172/cig.2917, 2016.
Fernández-Fernández, J. M., Oliva, M., Palacios, D., Garcia-Oteyza, J., Navarro, F. J., Schimmelpfennig, I., Léanni, L., and Team, A.: Ice thinning on nunataks during the glacial to interglacial transition in the Antarctic Peninsula region according to Cosmic-Ray Exposure dating: Evidence and uncertainties, Quat. Sci. Rev., 264, 107029, https://doi.org/10.1016/j.quascirev.2021.107029, 2021.
González-Herrero, S., Barriopedro, D., Trigo, R. M., López-Bustins, J. A., and Oliva, M.: Climate warming amplified the 2020 record-breaking heatwave in the Antarctic Peninsula, Commun. Earth Environ., 3, 122, https://doi.org/10.1038/s43247-022-00450-5, 2022.
Gorodetskaya, I. V., Durán-Alarcón, C., González-Herrero, S., Clem, K. R., Zou, X., Rowe, P., Rodriguez Imazio, P., Campos, D., Leroy-Dos Santos, C., Dutrievoz, N., Wille, J. D., Chyhareva, A., Favier, V., Blanchet, J., Pohl, B., Cordero, R. R., Park, S.-J., Colwell, S., Lazzara, M. A., Carrasco, J., Gulisano, A. M., Krakovska, S., Ralph, F. M., Dethinne, T., and Picard, G.: Record-high Antarctic Peninsula temperatures and surface melt in February 2022: a compound event with an intense atmospheric river, Npj Clim. Atmos. Sci., 6, 202, https://doi.org/10.1038/s41612-023-00529-6, 2023.
Hrbáček, F., Oliva, M., Hansen, C., Balks, M., O’Neill, T. A., De Pablo, M. A., Ponti, S., Ramos, M., Vieira, G., Abramov, A., Kaplan Pastíriková, L., Guglielmin, M., Goyanes, G., Francelino, M. R., Schaefer, C., and Lacelle, D.: Active layer and permafrost thermal regimes in the ice-free areas of Antarctica, Earth-Sci. Rev., 242, 104458, https://doi.org/10.1016/j.earscirev.2023.104458, 2023.
IPCC: Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)], IPCC, 2014.
Jones, P. D. and Lister, D. H.: Antarctic near-surface air temperatures compared with ERA-Interim values since 1979, Int. J. Climatol., 35, 1354–1366, https://doi.org/10.1002/joc.4061, 2015.
Kaplan Pastíriková, L., Hrbáček, F., and Matějka, M.: Validation of ERA5-Land-based reconstructed air temperature and near-surface ground temperature on James Ross Island, Polar Geogr., 48, 95–115, https://doi.org/10.1080/1088937X.2024.2434744, 2025.
O’Neill, B. C., Tebaldi, C., Van Vuuren, D. P., Eyring, V., Friedlingstein, P., Hurtt, G., Knutti, R., Kriegler, E., Lamarque, J.-F., Lowe, J., Meehl, G. A., Moss, R., Riahi, K., and Sanderson, B. M.: The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6, Geosci. Model Dev., 9, 3461–3482, https://doi.org/10.5194/gmd-9-3461-2016, 2016.
Ramos, M., Vieira, G., De Pablo, M. A., Molina, A., Abramov, A., and Goyanes, G.: Recent shallowing of the thaw depth at Crater Lake, Deception Island, Antarctica (2006–2014), CATENA, 149, 519–528, https://doi.org/10.1016/j.catena.2016.07.019, 2017.
Ramos, M., Vieira, G., De Pablo, M. A., Molina, A., and Jimenez, J. J.: Transition from a Subaerial to a Subnival Permafrost Temperature Regime Following Increased Snow Cover (Livingston Island, Maritime Antarctic), Atmosphere, 11, 1332, https://doi.org/10.3390/atmos11121332, 2020.
Tetzner, D., Thomas, E., and Allen, C.: A Validation of ERA5 Reanalysis Data in the Southern Antarctic Peninsula—Ellsworth Land Region, and Its Implications for Ice Core Studies, Geosciences, 9, 289, https://doi.org/10.3390/geosciences9070289, 2019.
Turner, J., Marshall, G. J., Clem, K., Colwell, S., Phillips, T., and Lu, H.: Antarctic temperature variability and change from station data, Int. J. Climatol., 40, 2986–3007, https://doi.org/10.1002/joc.6378, 2020.
Van Vuuren, D. P., O’Neill, B. C., Tebaldi, C., Sanderson, B. M., Chini, L. P., Friedlingstein, P., Hasegawa, T., Riahi, K., Govindasamy, B., Bauer, N., Eyring, V., Fall, C. M. N., Frieler, K., Gidden, M. J., Gohar, L. K., Högner, A., Jones, A. D., Kikstra, J., King, A., Knutti, R., Kriegler, E., Lawrence, P., Lennard, C., Lowe, J., Mathison, C., Mehmood, S., Nicholls, Z., Prado, L. F., Zhang, Q., Rose, S. K., Ruane, A. C., Sandstad, M., Schleussner, C.-F., Seferian, R., Sillmann, J., Smith, C., Sörensson, A. A., Panickal, S., Tachiiri, K., Vaughan, N., Vishwanathan, S. S., Yokohata, T., Zecchetto, M., and Ziehn, T.: The Scenario Model Intercomparison Project for CMIP7 (ScenarioMIP-CMIP7), Geosci. Model Dev., 19, 2627–2656, https://doi.org/10.5194/gmd-19-2627-2026, 2026.
Westermann, S., Langer, M., Boike, J., Heikenfeld, M., Peter, M., Etzelmüller, B., and Krinner, G.: Simulating the thermal regime and thaw processes of ice-rich permafrost ground with the land-surface model CryoGrid 3, Geosci. Model Dev., 9, 523–546, https://doi.org/10.5194/gmd-9-523-2016, 2016.
Westermann, S., Ingeman-Nielsen, T., Scheer, J., Aalstad, K., Aga, J., Chaudhary, N., Etzelmüller, B., Filhol, S., Kääb, A., Renette, C., Schmidt, L. S., Schuler, T. V., Zweigel, R. B., Martin, L., Morard, S., Ben-Asher, M., Angelopoulos, M., Boike, J., Groenke, B., Miesner, F., Nitzbon, J., Overduin, P., Stuenzi, S. M., and Langer, M.: The CryoGrid community model (version 1.0) – a multi-physics toolbox for climate-driven simulations in the terrestrial cryosphere, Geosci. Model Dev., 16, 2607–2647, https://doi.org/10.5194/gmd-16-2607-2023, 2023.
Wilhelm, K. R. and Bockheim, J. G.: Climatic controls on active layer dynamics: Amsler Island, Antarctica, Antarct. Sci., 29, 173–182, https://doi.org/10.1017/S0954102016000511, 2017.
Wilhelm, K. R., Bockheim, J. G., and Haus, N. W.: Properties and processes of recently established soils from deglaciation of Cierva Point, Western Antarctic Peninsula, Geoderma, 277, 10–22, https://doi.org/10.1016/j.geoderma.2016.05.001, 2016.
Zhu, J., Xie, A., Qin, X., Wang, Y., Xu, B., and Wang, Y.: An Assessment of ERA5 Reanalysis for Antarctic Near-Surface Air Temperature, Atmosphere, 12, 217, https://doi.org/10.3390/atmos12020217, 2021.
-
AC1: 'Reply on RC1', Joana Baptista, 21 Sep 2026
-
RC2: 'Comment on egusphere-2026-1914', Tomáš Uxa, 28 Jul 2026
Of course, the list of useful references in my review omits Kaplan Pastíriková et al. (2023), so I am adding it separately.
Kaplan Pastíriková, L., Hrbáček, F., Uxa, T., Láska, K., 2023. Permafrost table temperature and active layer thickness variability on James Ross Island, Antarctic Peninsula, in 2004–2021. Science of the Total Environment, 869, 161690. https://doi.org/10.1016/j.scitotenv.2023.161690
All the best,
Tomáš Uxa
Citation: https://doi.org/10.5194/egusphere-2026-1914-RC2 -
AC4: 'Reply on RC2', Joana Baptista, 21 Sep 2026
Reply provided on Reply on RC1
Citation: https://doi.org/10.5194/egusphere-2026-1914-AC4
-
AC4: 'Reply on RC2', Joana Baptista, 21 Sep 2026
-
RC3: 'Comment on egusphere-2026-1914', Anonymous Referee #2, 30 Jul 2026
This manuscript applies CryoGrid to reconstruct and project ground temperatures at five observatories in the northwestern Antarctic Peninsula. The study is timely, the observational network is valuable, and the comparison among sites could make a useful contribution. However, several aspects of the model evaluation and future projections should be clarified before the main conclusions can be assessed with confidence. I therefore recommend major revision.
1) It appears that the snowfall multipliers and some subsurface parameters were adjusted using the same ground-temperature observations later used to evaluate model performance. If so, Table 3 represents calibration performance rather than independent validation. Please specify the calibration and evaluation periods, observations, and depths used at each site. I would recommend a split-sample or leave-one-year-out evaluation, together with the parameter ranges and selection criterion. Agreement at 0.4 m should not be presented as independent model skill if that depth was used to determine the snowfall multiplier.
2) The projections depend on one CMIP6 model and apparently one random realization of the monthly carrier sequence. ACCESS-CM2 was selected partly because it reproduces the ERA5 trend, although another candidate has smaller MAE and RMSE. This does not necessarily demonstrate that ACCESS-CM2 provides the most credible future regional warming. Please show how model selection and carrier sampling affect the main results. Ideally, the authors should include projections from the other candidate models and several carrier realizations. If this is not feasible, the results should be described as ACCESS-CM2-driven sensitivity experiments, and threshold crossings should be reported by approximate decade or as ranges rather than as exact years.
Section 3.5 also needs clarification. Please verify Equation (2) against the implemented code and explain whether the same sampled month was used for all forcing variables. Otherwise, covariance among temperature, humidity, precipitation, wind, and radiation may not be preserved.
3) The manuscript frequently interprets positive mean annual ground temperature at 20 m as the disappearance of permafrost. I would avoid treating these as equivalent. A positive annual mean at one depth does not establish the thermal state of the full profile or whether permafrost persists above or below that level. Please define the criterion used to diagnose permafrost degradation and distinguish it from the year when MAGT at 20 m becomes positive. If possible, provide active-layer thickness and talik diagnostics from the complete simulated profile. Otherwise, “permafrost loss” should be replaced by the more precise phrase “positive MAGT at 20 m”. Active-layer thickness should also be removed from the title of Section 4.3 unless the corresponding results are presented.
Other comments:
Abstract, lines 14-15: Replace “constrains trends evaluation” with “constrains the evaluation of trends.”
Abstract: Please state clearly that the final projections are based on one CMIP6 model. Also avoid wording that equates positive MAGT at 20 m directly with complete permafrost loss.
Line 50: Remove the unnecessary parentheses around “Obu et al. (2020).”
Lines 58-60: “Recommended Concentration Pathway” should be “Representative Concentration Pathway.” More importantly, CMIP6 scenarios are SSP-based and should not be presented as predictions.
Section 2.1: The heading uses “Western Antarctic Peninsula,” whereas the rest of the manuscript refers to the “northwestern Antarctic Peninsula.” Please standardize the geographical terminology and capitalization.
Line 73: Replace “divided in two” with “divided into two.”
Line 76 and elsewhere: Replace “Thomas e Tetzner 2019” with “Thomas and Tetzner (2019).”
Line 97: “measured at King Sejong Station.”
Table 1: “Measurement height/depth” combines height above the surface and depth below it. Please separate these quantities or use a clear signed convention.
Lines 193-196: Air temperature appears to be missing from the ERA5 variable list. Dew-point temperature is listed, whereas CryoGrid uses specific humidity; please explain the conversion.
Table 2: Please provide units for MAE and RMSE and specify the sampling frequency and evaluation period. The meaning of the reported p-value should also be defined, considering temporal autocorrelation.
Lines 246-250: Please clarify whether monthly CMIP6 data were used and standardize the variable names and units. Specific humidity should be reported in kg kg⁻¹ rather than as dimensionless.
Line 257: The description of the vertical model resolution is unclear. Please state explicitly the grid spacing above and below 5 m depth.
Section 3.5: In addition to the issues raised above, all symbols in Equations (2) and (3) should be defined consistently, and the ERA5 and CMIP6 terms should be distinguishable.
Table 3: Please define whether mineral and water/ice contents are volume fractions and provide units, sample sizes, and evaluation periods for the performance statistics.
Figure 2: The caption should define the box, whisker, and outlier conventions and give the observation period for each site.
Section 4.2 and Figure 3: These give different historical periods. More generally, please check all analysis periods, as the manuscript alternates among 1950–2020, 1950–2022, 2020–2100, and 2022–2100.
Figure 3: Please define the SWE statistic. For example, whether it represents an annual mean, maximum, or another quantity. The caption should also be revised because it groups air temperature and snowfall under “ground temperatures.”
Figure 4: This appears to show five point simulations rather than a spatially continuous estimate. Please revise the figure or caption accordingly. The use of the single year 1950 as a baseline should also be justified, particularly if it was exceptionally cold.
Results and Discussion: Please distinguish clearly between total temperature changes in °C and trends in °C decade⁻¹. Each value should have a clearly defined reference period.
Results, Discussion, and Conclusions: Several threshold years differ among the abstract, Results, Discussion, and Conclusions. I suggest generating all threshold years from one summary table and using them consistently throughout.
Section 4.3: Active-layer thickness should be removed from the section title unless the corresponding results are presented.
Figure 8: Figure 8 is cited, but no Figure 8 is included. Please identify the intended figure.
Table 5: The caption should reflect that the table includes MAAT and MAGST as well as MAGT at 20 m. Please also use uppercase “SSP” consistently.
Figure 5: Please identify the transition from ERA5 to synthesized ACCESS-CM2 forcing and define the line colors in the caption.
Figure 6: Please use the complete scenario names SSP1-2.6, SSP2-4.5, and SSP5-8.5.
Discussion: Some explanations involving the Amundsen Sea Low, sea-ice changes, and snow effects are presented causally without direct supporting analysis. Please provide the relevant diagnostics or present these mechanisms as possible explanations.
Conclusions: Please verify statements such as “most observatories” against the site-specific results. Wider implications involving thermokarst and slope instability should also be presented as possible consequences rather than direct model results, particularly because most sites are represented as bedrock with little ground ice.
Appendix C: The text refers to Table A1, but this appears to be Table C1.
Table C1: Please use the official CMIP6 model identifiers consistently. The IPSL model appears in different forms, including IPSL-CM6A, IPSL-CM6-A, and IPSL-CM6A-LR.
Appendix E: Its title gives 2022–2100, whereas the associated analysis also uses 2020–2100. Please standardize the period.
Code and data availability: Please provide the exact CryoGrid version, release, or commit used. The processed forcing, parameter files, carrier realization or ensemble, and relevant scripts should also be made available in a persistent repository to ensure reproducibility.
Citation: https://doi.org/10.5194/egusphere-2026-1914-RC3 -
AC2: 'Reply on RC3', Joana Baptista, 21 Sep 2026
We appreciate the time and effort spent on the revision of the manuscript “Simulating the permafrost thermal regime in the Northwestern Antarctic Peninsula from 1950 to 2100”. The comments and suggestions were very well received, and we have subsequently revised and improved the text and work presented. In the following, we provide point-by point replies to all issues raised. The reviewer comment appears in bold font, our replies in normal font and changes to the text which will be implemented in the revised version of the manuscript are in italics.
On behalf of all authors,
Joana BaptistaReply to the major comments:
Point 1: It appears that the snowfall multipliers and some subsurface parameters were adjusted using the same ground-temperature observations later used to evaluate model performance. If so, Table 3 represents calibration performance rather than independent validation. Please specify the calibration and evaluation periods, observations, and depths used at each site. I would recommend a split-sample or leave-one-year-out evaluation, together with the parameter ranges and selection criterion. Agreement at 0.4 m should not be presented as independent model skill if that depth was used to determine the snowfall multiplier.
Reply: In our implementation of CryoGrid, subsurface parameters (mineral, water/ice and organic contents) were defined based on-site characteristics and only minor adjustments were made during model calibration through the comparison with observed ground temperatures at different depths (0.4, 1.2, 3.0 m, and the maximum depth for each borehole). MAGST (0.02 m depth) was used as the upper boundary forcing. This approach aimed at reducing potential biases associated with ERA5 forcing and at maximizing the use of the relatively short observational records available at several sites.
The snowfall multiplier was determined in a second phase using ERA5 forcing and the surface energy balance scheme. The multiplier was adjusted to reproduce the observed insulating effect of snow cover observed on ground temperature at 0.4 which was not used to force the model.
We acknowledge that the separation of the calibration and evaluation periods can improve the interpretation of the procedure. To address this, the year 2021 is now used for calibration of the subsurface parameters and snowfall multiplier, while the period from the beginning of the observational records through 2020 was used for independent model evaluation (Table 1). We have also specified in the revised manuscript (Table 1 and Section 3.2.2 Model calibration and evaluation), for each observatory, the calibration and evaluation periods, the observations used, and the depths considered. Separate tables reporting calibration and evaluation statistics have been added accordingly (Section 4.1 Sensitivity analysis to subsurface parameterization and validation of the surface energy model).Point 2: The projections depend on one CMIP6 model and apparently one random realization of the monthly carrier sequence. ACCESS-CM2 was selected partly because it reproduces the ERA5 trend, although another candidate has smaller MAE and RMSE. This does not necessarily demonstrate that ACCESS-CM2 provides the most credible future regional warming. Please show how model selection and carrier sampling affect the main results. Ideally, the authors should include projections from the other candidate models and several carrier realizations. If this is not feasible, the results should be described as ACCESS-CM2-driven sensitivity experiments, and threshold crossings should be reported by approximate decade or as ranges rather than as exact years.
Section 3.5 also needs clarification. Please verify Equation (2) against the implemented code and explain whether the same sampled month was used for all forcing variables. Otherwise, covariance among temperature, humidity, precipitation, wind, and radiation may not be preserved.
Reply: The selection of the ACCESS-CM2 model based on the reproduction of the ERA5 trend is especially important considering that this is the SSP2-4.5 scenario that simulates a central pathway in which the trend continues the historical pattern without substantial deviation which is observed on the INM CM4 8 model, despite the lower MAE and RMSE values. This deviation contradicts also the projections of Bracegirdle et al. (2020) for Antarctic. The INM CM4 8 model also misrepresents the interannual variability. In the revised manuscript we complemented the evaluation of the CMIP6 models. Additionally, we are aware of the limitation impose by using one model but running a multi-GCM ensemble is currently not possible on the CCM. However, this is a development planned in a future study.
Section 3.3.2 CMIP scenarios and models:
“To evaluate the ability of the downscaled CMIP6 models to represent climate conditions in the northwestern Antarctic Peninsula, we compared the simulated MAATs at the Papagal observatory location with ERA5 reanalysis data for the reference period 1950–2020 (Fig. 2 and Table 3). Papagal observatory was selected as the reference site because it is located in an exposed location that is representative of regional climate conditions and provides a long-term, high-quality observational record. Among the models, ACCESS-CM2 simulated slightly warmer temperatures than ERA5, with a MAE of 1.16 ºC and a RMSE of 1.33 ºC. Its standard deviation (σ = 0.98) was close to that of ERA5 (σ = 1.01), indicating a similar representation of interannual variability. INM-CM4-8 generally reproduced the ERA5 MAATs but misrepresented the interannual variability, as indicated by its lower standard deviation (σ = 0.78). However, it showed the lowest errors among the three models, with a MAE of 0.74 ºC and a RMSE of 0.84 ºC. In contrast, IPSL-CM6A-LR overestimated temperatures by up to 5 ºC in some years (e.g. 2011-2013) and had the highest MAE and RMSE values, at 1.37 ºC and 2.81 ºC. Its standard deviation was the highest (σ = 1.49.), suggesting that it captured the magnitude of interannual variability, despite its substantial temperature bias.
For the period 1950- 2020 linear regression showed that ACCESS-CM2 simulates a warming trajectory of 0.36 ºC dec-1, closer to ERA5 (0.33 ºC dec-1), while IPSL-CM6A-LR estimates a stronger warming of 0.44 ºC dec-1. In contrast, INM-CM4-8 results in a much weaker warming of 0.15 ºC dec-1. Additionally, INM-CM4-8 projects for the future (2020-2100) a cooling trend of -0.05 ºC dec-1 that contradicts the projections of Bracegirdle et al. (2020) for Antarctic and the Southern Ocean surface climate.
The difference in the air temperature simulated by each model strongly impacts the simulation of ground temperatures, leading to markedly different permafrost thermal regimes and future trajectories. ACCESS CM2 was selected to force the simulations of future permafrost evolution as it shows the best overall performance against ERA5, combining a similar linear trend and a realistic representation of interannual variability.”Along the manuscript was stated that the future projections are based on the ACCESS-CM2 model (abstract, section 3.3.2 CMIP6 scenarios and models and 6. Conclusions) and we reinforced that this is an ACCESS-CM2-driven sensitivity experiment in section 4.3 Future evolution of permafrost temperature under CMIP6 scenarios (2020 - 2100). Additionally, the Appendix on the evaluation of the CMIP6 models was moved into the main structure and divided into the sections 3.3.2 CMIP6 scenarios and models and 5.1 Evaluation of the forcing data.
Regarding the “threshold crossings” we also agree that the projected threshold crossings should not be interpreted as exact dates. Accordingly, the revised manuscript reports these transitions as approximate decades or time ranges rather than as specific years.Section 4.3.1 SSP1-2.6:
“At 10 m depth, MAGT exhibits an average warming trend of 0.27 ± 0.01 ºC dec-1, accompanied by pronounced interannual variability (Table 8 and Fig. 6). This warming is projected to result in permafrost thawing at the low-elevation SKO Station and Amsler observatory during the 2040s.”
Concerning the equation 3 (previous equation 2), the same month was used for all forcing variables to conserve the covariance. The sentence was rewritten for clarity.
“miCMIP is equal to MiCMIP during the overlap period, while it is set to the running mean of 10-year periods for the remainder of the time series, so that short-term fluctuations can be separated from multi-annual trends (which are accounted for by the first three terms on the right-hand side). To conserve the covariance between the different model forcing variables, they were all derived from the same month. With this procedure, we ensure that the corrected CMIP6 data features the same monthly mean and standard deviation as the ERA5 data for the overlap period.”
Point 3: The manuscript frequently interprets positive mean annual ground temperature at 20 m as the disappearance of permafrost. I would avoid treating these as equivalent. A positive annual mean at one depth does not establish the thermal state of the full profile or whether permafrost persists above or below that level. Please define the criterion used to diagnose permafrost degradation and distinguish it from the year when MAGT at 20 m becomes positive. If possible, provide active-layer thickness and talik diagnostics from the complete simulated profile. Otherwise, “permafrost loss” should be replaced by the more precise phrase “positive MAGT at 20 m”. Active-layer thickness should also be removed from the title of Section 4.3 unless the corresponding results are presented.
Reply: In the manuscript we associated positive MAGT at 20 m depth with permafrost degradation. In the revised version, we now present permafrost degradation as a possible outcome rather than as a definitive conclusion and have removed statements implying a direct relationship. Moreover, we replace some references to degradation by the word thaw. Additionally, we have added two figures with the complete simulated profile to support the analysis in the Appendix D.
Reply to the minor comments:
Abstract, lines 14-15: Replace “constrains trends evaluation” with “constrains the evaluation of trends.”
Reply: Corrected.
Abstract: Please state clearly that the final projections are based on one CMIP6 model. Also avoid wording that equates positive MAGT at 20 m directly with complete permafrost loss.
Reply: When referring to future projections the model (ACCESS CM2) used is identified. Regarding the permafrost loss, we adjusted the text to prevent the direct relation.
Line 50: Remove the unnecessary parentheses around “Obu et al. (2020).”Reply: Corrected.
Lines 58-60: “Recommended Concentration Pathway” should be “Representative Concentration Pathway.” More importantly, CMIP6 scenarios are SSP-based and should not be presented as predictions.
Reply: The sentences were rewritten for clarity.
Previous “Our goal is to implement a modelling approach based on the CryoGrid Community Model (CCM) (Westermann et al., 2023), which integrates site-specific characteristics to estimate ground temperature over defined time periods. Specifically, we simulate recent past conditions using reanalysis data, validate the models against observed permafrost temperature data and project the ground temperature evolution until 2100 using selected Coupled Model Intercomparison Project (CMIP) models in different Recommended Concentration Pathway scenarios (IPCC, 2014; O’Neill et al., 2016).”
New section “Our goal is to implement a modelling approach based on the CryoGrid Community Model (CCM) (Westermann et al., 2023), which integrates site-specific characteristics to estimate ground temperature over defined time periods. In the present study, the CCM was implemented for five observatories of the PERMANTAR network (http://permantar.weebly.com) distributed along the Northwestern Antarctic Peninsula, specifically, to validate the models against observed permafrost temperature data, simulate past conditions from 1950 to 2020 using reanalysis data, and project the ground temperature evolution from 2020 until 2100 under three Shared Socioeconomic Pathways (SSPs) from the Coupled Model Intercomparison Project (CMIP) (IPCC, 2014; O’Neill et al., 2016).
In Baptista et al. (2025), the performance of the CCM was evaluated for the King Sejong permafrost observatory on Barton Peninsula, King George Island. The results of the simulation from 2020-2022 show that the model successfully represented the site conditions with correlations above 0.8 and MAE ranging from 0.1 to 0.7 ºC. Following the validation of the model, a long-term simulation from 1950 to 2022 was run showing a ground warming trend at 20 m of 0.25 ºC dec⁻¹, followed by an increase in the active layer thickness from 1.5 m in 1950 to 3.5 m in 2022.”Section 2.1: The heading uses “Western Antarctic Peninsula,” whereas the rest of the manuscript refers to the “northwestern Antarctic Peninsula.” Please standardize the geographical terminology and capitalization.
Reply: You are correct and it was something pending from an earlier version. Actually, the section provides a general description of the Antarctic Peninsula- Hence in the heading the word “Western” was removed.
Line 73: Replace “divided in two” with “divided into two.”
Reply: Corrected.
Line 76 and elsewhere: Replace “Thomas e Tetzner 2019” with “Thomas and Tetzner (2019).”
Reply: Corrected.
Line 97: “measured at King Sejong Station.”
Reply: Corrected.
Table 1: “Measurement height/depth” combines height above the surface and depth below it. Please separate these quantities or use a clear signed convention.
Reply: The table was updated. Now exists one column for the height and another for the depth.
Lines 193-196: Air temperature appears to be missing from the ERA5 variable list. Dew-point temperature is listed, whereas CryoGrid uses specific humidity; please explain the conversion.
Reply: The 2 m temperature was missing from the previous version, being now identified on the variables listed.
Table 2: Please provide units for MAE and RMSE and specify the sampling frequency and evaluation period. The meaning of the reported p-value should also be defined, considering temporal autocorrelation.
Reply: The table was updated with the required information.
Lines 246-250: Please clarify whether monthly CMIP6 data were used and standardize the variable names and units. Specific humidity should be reported in kg kg⁻¹ rather than as dimensionless.
Reply: The sentence was rewritten for clarity and the unit added.
Section 3.3.2 CMIP6 scenarios and models:
“Data was downloaded from the Climate Data Store (Copernicus Climate Change Service, 2021) for the historical, the SSP1-2.6, the SSP2-4.5 and the SSP5-8.5 scenarios. The variables at monthly resolution used to force the simulation were:”
Line 257: The description of the vertical model resolution is unclear. Please state explicitly the grid spacing above and below 5 m depth.
Reply: The sentence was rewritten for clarity.
Section 3.4 Long-term simulation of the past and future evolution of permafrost temperatures:
“In the simulation’s output we provide estimates of ground temperatures for a profile of 20 m depth, with vertical resolution varying with depth: 0.05 m from the surface to 1 m, 0.1 m from 1 to 2 m, 0.2 m from 2 to 5 m, and 0.5 m below 5 m depth.”
Section 3.5: In addition to the issues raised above, all symbols in Equations (2) and (3) should be defined consistently, and the ERA5 and CMIP6 terms should be distinguishable.
Reply: We reviewed the terms to ensure the distinction.
Table 3: Please define whether mineral and water/ice contents are volume fractions and provide units, sample sizes, and evaluation periods for the performance statistics.
Reply: The table was updated with the information required. Additionally, information on the mineral and water/ice contents type of unit is provided on the methods section. As previously mentioned, we will split the calibration and evaluation periods and update the table accordingly.
Section 3.2.2 Model calibration and evaluation:
“Following Baptista et al. (2025), the model calibration for each borehole was performed in two phases: i. the simulations were run for 2021 forced with ground surface temperature, being subsurface parameters (i.e. mineral and water/ice content) adjusted to obtain the best configuration through a sensitive analysis, ii. the best configuration scheme was then used when forcing the simulations with ERA5 data of the same year, but using the CCM surface energy balance scheme, in order to define the snowfall multiplication factor. The model performance was evaluated for each borehole by comparing the results from the beginning of the observational period until 2020 (Table 1).
For the adjustment of the subsurface parameters, a class representing a ground column with temperature boundary condition at the surface (GROUND_freeW_ubT, Westermann et al., 2023) was selected, for which water phase change occurs at 0 ºC and the sum of water and ice contents remain constant (Baptista et al., 2025). Measured ground surface temperature from the PERMANTAR monitoring sites was used as upper boundary input while a geothermal heat flux value obtained from An et al. (2015) and applied as the lower boundary. Geothermal heat flux values for King Sejong, Papagal and SKO Meteo were 73 mWm-2, for Cierva were 61 mWm-2, and for Amsler were 58 mWm-2. The thermal conductivity was obtained from thermophysical analysis performed on the rock cores. For this, the cores were sliced, and the conductivity was measured with an ISOMET 2104. 1D simulations were run with different configurations of mineral and water/ice volumetric content, that goes from 0 to 1, for each of the five boreholes. From each simulation, estimated daily average ground temperatures at 4 depths (0.4, 1.2, 3.0, and maximum depth of borehole) were extracted and compared with the observational records. The metrics used for the assessment of model performance were bias, correlation coefficient (r), the p-value associated with the correlation coefficient which was considered statistically significant when p < 0.05, mean absolute error (MAE) and root mean square error (RMSE).”Figure 2: The caption should define the box, whisker, and outlier conventions and give the observation period for each site.
Reply: Corrected. Previous Figure 2 is now Figure 3.
“Seasonal variation (DJF, MAM, JJA and SON) of simulated and observed ground temperatures on the PERMANTAR observatories at 0.4 m, 3 m and maximum borehole depth. Boxes indicate the interquartile range, the central line indicates the median, whiskers extend to 1.5 times the interquartile range. The record covers 2020 for King Sejong, 2011-2020 for Papagal, 2012-2020 for SKO Station, 2014-2020 for Cierva and 2018-2020 for Amsler.”
Section 4.2 and Figure 3: These give different historical periods. More generally, please check all analysis periods, as the manuscript alternates among 1950–2020, 1950–2022, 2020–2100, and 2022–2100.
Reply: The analysis periods were standardized.
Figure 3: Please define the SWE statistic. For example, whether it represents an annual mean, maximum, or another quantity. The caption should also be revised because it groups air temperature and snowfall under “ground temperatures.”
Reply: Corrected. Previous Figure 3 is now Figure 4.
“Long-term simulation (January 1950 to December 2020) for the PERMANTAR observatories: annual snowfall (mm water equivalent), mean annual air temperature - MAAT (ºC); mean annual ground surface temperature - MAGST (ºC); mean annual ground temperature at 10 m - MAGT (ºC); mean annual ground temperature at 20 m - MAGT (ºC).”
Figure 4: This appears to show five point simulations rather than a spatially continuous estimate. Please revise the figure or caption accordingly. The use of the single year 1950 as a baseline should also be justified, particularly if it was exceptionally cold.
Reply: We understand the reviewer’s concern. Figure 5 (previous figure 4) is intended as a spatial comparison. We agree that would be more appropriate to present the estimate for a period, as suggested. Therefore, we have calculated averages for the corresponding decades. This means that the value previously shown for 1950 now represents the average for the period 1950-1960, the value for 2020 represents the average for 2010-2020, and the value for 2100 refers to the period 2090-2100.
Results and Discussion: Please distinguish clearly between total temperature changes in °C and trends in °C decade⁻¹. Each value should have a clearly defined reference period.
Reply: To uniformize, we replaced the temperature change by trend per decade.
Results, Discussion, and Conclusions: Several threshold years differ among the abstract, Results, Discussion, and Conclusions. I suggest generating all threshold years from one summary table and using them consistently throughout.
Reply: The thresholds were uniformized among the sections.
Section 4.3: Active-layer thickness should be removed from the section title unless the corresponding results are presented.
Reply: It was removed from the section title.
Figure 8: Figure 8 is cited, but no Figure 8 is included. Please identify the intended figure.
Reply: Corrected.
Table 5: The caption should reflect that the table includes MAAT and MAGST as well as MAGT at 20 m. Please also use uppercase “SSP” consistently.
Reply: Previous Table 5 is now Table 6. The caption was updated with the missing information and the SSP acronym corrected.
“Change rates from the simulations of future temperature evolution on the PERMANTAR observatories from 2020 to 2100: mean annual air temperature - MAAT (ºC/ decade); mean annual ground surface temperature - MAGST (ºC/ decade); mean annual ground temperature at 10 m depth - MAGT (ºC/ decade); mean annual ground temperature at 20 m depth - MAGT (ºC/ decade).”
Figure 5: Please identify the transition from ERA5 to synthesized ACCESS-CM2 forcing and define the line colors in the caption.
Reply: Previous Figure 5 is now Figure 6. The description of the dotted line was added to the legend of the figure.
“Long-term simulation of air and ground temperature (2020 - 2100) on PERMANTAR observatories: mean annual air temperature - MAAT (ºC); mean annual ground surface temperature - MAGST (ºC); mean annual ground temperature at 10 m depth - MAGT (ºC); mean annual ground temperature at 20 m depth - MAGT (ºC). Vertical line represents the transition from the ERA5 series to the synthesized CMIP6 series as forcing of the long-term simulation.”
Figure 6: Please use the complete scenario names SSP1-2.6, SSP2-4.5, and SSP5-8.5.
Reply: The complete scenario name is now used on the figures.
Discussion: Some explanations involving the Amundsen Sea Low, sea-ice changes, and snow effects are presented causally without direct supporting analysis. Please provide the relevant diagnostics or present these mechanisms as possible explanations.
Reply: We have developed further the discussion section to incorporate the explanations associated with the ASL and sea-ice changes.
“Following the historical trend, the SSP2-4.5 scenario estimates a much stronger MAAT warming rate of 0.31 ± 0.02 ºC dec⁻¹. Consequently, MAGT at 20 m depth exhibits a more accelerated warming, leading to positive MAGTs at 20 m depth at all five observatories in the 2080s, with values ranging from 1 to 2 ºC in 2100. In contrast to the SSP1-2.6 scenario, SSP2-4.5 displays a distinct spatial pattern in MAAT warming rates, with increased warming at the more southerly observatories which may be related to the sea ice loss on the Bellingshausen Sea and consequent deepening of the ASL that strengthens north-northwesterly flow towards the Antarctic Peninsula, favouring advection of warm, maritime air and raising surface air temperatures, especially during the winter (Abram et al., 2010; Clem et al., 2017; Hosking et al., 2013; Turner et al., 2013).”
Conclusions: Please verify statements such as “most observatories” against the site-specific results. Wider implications involving thermokarst and slope instability should also be presented as possible consequences rather than direct model results, particularly because most sites are represented as bedrock with little ground ice.
Reply: The sentences were rewritten to remove the direct relationship.
“Our results for bedrock sites indicate that the future projected warming across the Northwestern Antarctic Peninsula, independently of the CMIP6 scenario will very likely lead to the thawing of permafrost from coastal areas and low elevation sites. Although extrapolation to non-bedrock settings is subject to limitations, our results provide an important indication of regional warming trends that are likely to influence permafrost conditions across a range of ice-free environments. Such substantial changes in the ground thermal regime may modify surface and subsurface hydrology, biogeochemical fluxes and geomorphological processes, with implications for the sensitive terrestrial (and potentially also to nearshore) ecosystems. They may also increase the vulnerability of human infrastructure (research stations, roads, and airstrips) to degradation. Similarly to what is recorded in comparable regions in the Arctic, in non-consolidated deposits, thermokarst and landslide activity will likely increase. In the steep rocky slopes, so typical of the coastal areas of the Northwestern Antarctic Peninsula, as bedrock warms and thaw propagates along joints, rockfalls and landslides may become more frequent, increasing hazards in areas that are experiencing growing tourist visitation during the austral summer.”
Appendix C: The text refers to Table A1, but this appears to be Table C1.
Reply: Corrected.
Table C1: Please use the official CMIP6 model identifiers consistently. The IPSL model appears in different forms, including IPSL-CM6A, IPSL-CM6-A, and IPSL-CM6A-LR.
Reply: Corrected to IPSL-CM6A-LR.
Appendix E: Its title gives 2022–2100, whereas the associated analysis also uses 2020–2100. Please standardize the period.
Reply: The figure presented on Appendix E was incorporated in Figure 6.
Code and data availability: Please provide the exact CryoGrid version, release, or commit used. The processed forcing, parameter files, carrier realization or ensemble, and relevant scripts should also be made available in a persistent repository to ensure reproducibility.
Reply: The CryoGrid version was provided in the section “Code availability” (Westermann et al., 2026). The parameter files, forcing data and simulation outputs are available at Zenodo (Pedro Baptista et al., 2026b, a).
References:
Abram, N. J., Thomas, E. R., McConnell, J. R., Mulvaney, R., Bracegirdle, T. J., Sime, L. C., and Aristarain, A. J.: Ice core evidence for a 20th century decline of sea ice in the Bellingshausen Sea, Antarctica, J. Geophys. Res., 115, 2010JD014644, https://doi.org/10.1029/2010JD014644, 2010.
An, M., Wiens, D. A., Zhao, Y., Feng, M., Nyblade, A., Kanao, M., Li, Y., Maggi, A., and Lévêque, J.: Temperature, lithosphere‐asthenosphere boundary, and heat flux beneath the Antarctic Plate inferred from seismic velocities, JGR Solid Earth, 120, 8720–8742, https://doi.org/10.1002/2015JB011917, 2015.
Baptista, J., Brito Guapo Teles Vieira, G., Manuel De Carvalho Soares Correia, A., Lee, H., and Westermann, S.: Modelling the evolution of permafrost temperatures and active layer thickness in King George Island, Antarctica, since 1950, The Cryosphere, 19, 3459–3476, https://doi.org/10.5194/tc-19-3459-2025, 2025.
Bracegirdle, T. J., Krinner, G., Tonelli, M., Haumann, F. A., Naughten, K. A., Rackow, T., Roach, L. A., and Wainer, I.: Twenty first century changes in Antarctic and Southern Ocean surface climate in CMIP6, Atmospheric Science Letters, 21, e984, https://doi.org/10.1002/asl.984, 2020.
Clem, K. R., Renwick, J. A., and McGregor, J.: Large-Scale Forcing of the Amundsen Sea Low and Its Influence on Sea Ice and West Antarctic Temperature, Journal of Climate, 30, 8405–8424, https://doi.org/10.1175/JCLI-D-16-0891.1, 2017.
Copernicus Climate Change Service: CMIP6 predictions underpinning the C3S decadal prediction prototypes, https://doi.org/10.24381/CDS.C866074C, 2021.
Hosking, J. S., Orr, A., Marshall, G. J., Turner, J., and Phillips, T.: The Influence of the Amundsen–Bellingshausen Seas Low on the Climate of West Antarctica and Its Representation in Coupled Climate Model Simulations, Journal of Climate, 26, 6633–6648, https://doi.org/10.1175/JCLI-D-12-00813.1, 2013.
IPCC: Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)], IPCC, 2014.
O’Neill, B. C., Tebaldi, C., Van Vuuren, D. P., Eyring, V., Friedlingstein, P., Hurtt, G., Knutti, R., Kriegler, E., Lamarque, J.-F., Lowe, J., Meehl, G. A., Moss, R., Riahi, K., and Sanderson, B. M.: The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6, Geosci. Model Dev., 9, 3461–3482, https://doi.org/10.5194/gmd-9-3461-2016, 2016.
Pedro Baptista, J., Westermann, S., and Vieira, G.: Outputs from the long-term simulations (1950-2100) of permafrost temperature evolution in the Northwestern Antarctic Peninsula, https://doi.org/10.5281/ZENODO.22872400, 2026a.
Pedro Baptista, J., Westermann, S., and Vieira, G.: Parameters files and forcing data used for the long-term simulations (1950-2100) of permafrost temperature evolution in the Northwestern Antarctic Peninsula, https://doi.org/10.5281/ZENODO.22872867, 2026b.
Turner, J., Phillips, T., Hosking, J. S., Marshall, G. J., and Orr, A.: The Amundsen Sea low, Intl Journal of Climatology, 33, 1818–1829, https://doi.org/10.1002/joc.3558, 2013.
Westermann, S., Ingeman-Nielsen, T., Scheer, J., Aalstad, K., Aga, J., Chaudhary, N., Etzelmüller, B., Filhol, S., Kääb, A., Renette, C., Schmidt, L. S., Schuler, T. V., Zweigel, R. B., Martin, L., Morard, S., Ben-Asher, M., Angelopoulos, M., Boike, J., Groenke, B., Miesner, F., Nitzbon, J., Overduin, P., Stuenzi, S. M., and Langer, M.: The CryoGrid community model (version 1.0) – a multi-physics toolbox for climate-driven simulations in the terrestrial cryosphere, Geoscientific Model Development, 16, 2607–2647, https://doi.org/10.5194/gmd-16-2607-2023, 2023.
Westermann, S., Pedro Baptista, J., and Vieira, G.: CryoGridCommunityModel - Version for long-term simulations with CMIP6, https://doi.org/10.5281/ZENODO.22870043, 2026.
-
AC2: 'Reply on RC3', Joana Baptista, 21 Sep 2026
-
RC4: 'Comment on egusphere-2026-1914', Anonymous Referee #3, 05 Aug 2026
General comments
I found this manuscript timely and potentially valuable. It uses the CryoGrid Community Model to reconstruct and project ground thermal conditions at five PERMANTAR observatories in the northwestern Antarctic Peninsula, combining rare borehole observations with ERA5 and ACCESS-CM2 forcing. The observational basis is important for a region where long ground-temperature records remain scarce, and the comparison among sites could make a useful contribution to Antarctic permafrost research.
At the same time, I am not yet persuaded that the present analysis supports the precision and spatial scope of several of the main conclusions. Principal concerns relate to the distinction between calibration and independent model evaluation, the treatment of uncertainty in snow and subsurface parameters, and the reliance on a single CMIP6 model. A positive mean annual ground temperature at 20 m is sometimes discussed as if it demonstrated complete permafrost loss, even though most boreholes are shallower than 20 m and observed permafrost was not reached in three of them. The extent to which five bedrock sites represent the wider range of ice-free coastal terrain also needs to be considered more explicitly.
On this basis, I recommend major revision. In my view, the paper would be considerably stronger if the authors documented the observational and modelling workflow more fully, evaluated the principal sources of uncertainty, and narrowed some of the claims concerning permafrost degradation and regional impacts.
Specific comments
1. I understand the practical reason for selecting relatively deep boreholes in bedrock and thereby reducing site heterogeneity. Nevertheless, this choice also limits the environments represented by the study. I would like the authors to explain how far these five sites can reasonably represent the ice-free areas of the northwestern Antarctic Peninsula. In particular, are unconsolidated marine terraces, glacial deposits, colluvium, or ice-rich sediments largely absent from the selected network? If so, this should be stated clearly before the point simulations are extended to coastal landscapes or used to discuss thermokarst, slope instability, and infrastructure exposure.
2. It’s difficult to reconstruct the observational basis for each simulation from the information distributed between the main text and the appendices. A single site-characteristics table would be helpful. It should include logger type and accuracy, monitored depths, observation periods and gaps, slope aspect and inclination, surface and subsurface material, vegetation or organic cover, and available information on local snow conditions.
It is also unclear which air-temperature observations were used to correct ERA5 at Amsler, where the local air-temperature logger appears to have been installed only in 2023.
3. As I understand the workflow, at least some subsurface parameters and the snowfall multipliers were selected by comparing simulations with the same ground-temperature records that were later used to report model performance. If it is correct, those statistics describe goodness of fit after calibration rather than performance against independent data, and they are likely to be optimistic.
Please separate calibration and validation periods, or apply a suitable temporal cross-validation procedure. Performance metrics should be presented independently for both periods. The duration and number of observations used at each site and depth should also be reported.
4. I could not determine the full parameter space explored during calibration. Please report the tested ranges, increments, and selection criterion for mineral, organic, water, and ice fractions and for thermal conductivity. At line 155, the manuscript states that the sum of water and ice contents remains constant, but the value at which it is held constant is not defined.
It would be useful to distinguish parameters supported by direct observations from parameters inferred through fitting. Were porosity, fracture density, water or ice content, and related properties documented during drilling or measured on cores? If not, the adopted values should be treated as uncertain parameters rather than physically validated site properties. A sensitivity or ensemble analysis is required to assess parameter equifinality.
5. What is the basis for applying the same geothermal heat flux of 50 mW m⁻² at all five sites, and how deep is the lower model boundary? Is this value supported by regional measurements or previous studies? Since the principal results concern temperature at 20 m, sensitivity to geothermal heat flux, domain depth and initial temperature profile should be evaluated.
6. The boreholes are only 4–15 m deep, whereas the main projections concern MAGT at 20 m. At several sites, the calculated temperature at 20 m is therefore well outside the observed depth interval. The associated uncertainty should be explicitly quantified.
Positive MAGT at 20 m should not automatically be equated with complete permafrost loss. Please define the diagnostic criterion used to determine the presence or absence of permafrost. The manuscript should distinguish between: positive MAGT at a specified depth; loss of permafrost at that depth; active-layer deepening; and complete disappearance of the permafrost body.
Regional statements about widespread permafrost loss should be revised accordingly.
7. Snow is interpreted as a major control on differences among sites, but the snowfall multiplier was calibrated indirectly against ground temperature. Are any field observations of snow depth, snow-cover duration, or snow water equivalent available? If not, independent remote-sensing or photographic evidence should be considered.
The assumption of a constant multiplier of 0.3 or 0.5 throughout the future simulation also requires justification. Future changes in precipitation phase, wind redistribution, and snow duration may alter the relationship between large-scale precipitation and local accumulation. At minimum, the sensitivity of the projections to the snowfall multiplier should be shown.
8. The phrase “adequate data to force the CCM” should be defined more precisely. Please state which required variables, temporal resolutions, ensemble members, and experiments were available for each CMIP6 model.
ACCESS-CM2 is selected mainly because its MAAT trend and variability resemble ERA5 at Papagal. The evaluation should include all five sites and the forcing variables most relevant to the surface energy balance, especially precipitation, radiation, and wind.
Projections based on one GCM cannot represent the full climate-model uncertainty.
9. I found the construction of the submonthly forcing difficult to follow. When a historical ERA5 month is sampled as the carrier, are all meteorological variables taken together from the same month and year, or are variables sampled separately? The former is important for retaining covariance among temperature, humidity, precipitation, wind, and radiation.
The random carrier sequence is itself a source of uncertainty. I would expect different admissible sequences to produce some spread in annual ground temperatures and in the estimated timing of threshold crossings. Several realisations should be generated and propagated through the analysis, with the random seed and sampling procedure documented.
10. I could not find a description of the statistical method used to estimate temperature trends or the uncertainty reported after the ± sign. Please state the regression method, confidence level and significance criterion.
The four historical periods appear to have been selected visually. Either apply an objective change-point or segmented-regression method, or describe these intervals explicitly as qualitative periods. Decadal rates calculated for the short 2015–2022 interval should be interpreted cautiously.
11. I agree that elevation, snow persistence, wind exposure, sea-ice decline, and changes in the Amundsen Sea Low are plausible explanations for the simulated contrasts. However, the present experiments do not isolate their individual effects. Unless additional factorial simulations or atmospheric indices are analysed, the language should be changed from causal statements to cautious interpretations such as “may be associated with” or “is consistent with”.
12. Please cross-check the abstract, Results, Discussion, Table 5, and Conclusions. There are inconsistencies in the reported years when MAGT becomes positive and in the use of “warming” versus “warming rate”. For example, values such as 0.59, 2.59, and 5.05 °C appear to represent total changes by 2100 in some places but are described as rates elsewhere.
The statement that positive MAGT occurs at most sites by 2045 under SSP5-8.5 is also not consistent with the site-specific results. The title of Sect. 4.3 refers to active-layer thickness, but active-layer results are not presented.
Technical corrections
At first use, spell out CryoGrid Community Model (CCM), and then use the abbreviation consistently.
The manuscript uses “Recommended Concentration Pathways”; this should be corrected.
Please standardise units and notation, including °C decade⁻¹, W m⁻¹ K⁻¹, and m a.s.l.
TTOP needs a precise definition, including how it is diagnosed from the model output and how it is interpreted at sites without confirmed permafrost within the borehole.
Replace “biochemical fluxes” with “biogeochemical fluxes”.
Increase the font and panel size in Figs. 2, 3, 5, and D1. Several time series and legends are difficult to read.Citation: https://doi.org/10.5194/egusphere-2026-1914-RC4 -
AC3: 'Reply on RC4', Joana Baptista, 21 Sep 2026
We appreciate the time and effort spent on the revision of the manuscript “Simulating the permafrost thermal regime in the Northwestern Antarctic Peninsula from 1950 to 2100”. The comments and suggestions were very well received, and we have subsequently revised and improved the text and work presented. In the following, we provide point-by point replies to all issues raised. The reviewer comment appears in bold font, our replies in normal font and changes to the text which will be implemented in the revised version of the manuscript are in italics.
On behalf of all authors,
Joana BaptistaReply to the major comments:
Point 1: I understand the practical reason for selecting relatively deep boreholes in bedrock and thereby reducing site heterogeneity. Nevertheless, this choice also limits the environments represented by the study. I would like the authors to explain how far these five sites can reasonably represent the ice-free areas of the northwestern Antarctic Peninsula. In particular, are unconsolidated marine terraces, glacial deposits, colluvium, or ice-rich sediments largely absent from the selected network? If so, this should be stated clearly before the point simulations are extended to coastal landscapes or used to discuss thermokarst, slope instability, and infrastructure exposure.
Reply: With the exception of the very shallow boreholes in Deception Island (not analysed here), all PERMANTAR boreholes are on bedrock, and the network does not include unconsolidated marine terraces, glacial deposits, colluvium, or ice-rich sediments. This was a design option aiming at standardizing the setting and assessing the regional variability, rather than the effect of site specific conditions. While our boreholes represent a significant part of the regional setting, the results have to be interpreted considering that they are for bedrock and cannot be directly applied to unconsolidated materials. Regarding the characteristics of observatories, a description is provided both on the methods and appendix A also with photographs.
Concerning the statement that the simulations were run for bedrock sites, we clarified it in the conclusions section.Conclusions “Moreover, the application of reanalysis data and CMIP6 projections from ACCESS-CM2 as forcing, allowed modelling the evolution of the recent past and future permafrost conditions on bedrock, providing an overview of the long-term ground thermal regime in the Northwestern Antarctic Peninsula.
(…)
Our results for bedrock sites indicate that the future projected warming across the Northwestern Antarctic Peninsula, independently of the CMIP6 scenario will very likely lead to the thawing of permafrost from coastal areas and low elevation sites. Although extrapolation to non-bedrock settings is subject to limitations, our results provide an important indication of regional warming trends that are likely to influence permafrost conditions across a range of ice-free environments. Such substantial changes in the ground thermal regime may modify surface and subsurface hydrology, biogeochemical fluxes and geomorphological processes, with implications for the sensitive terrestrial (and potentially also to nearshore) ecosystems. They may also increase the vulnerability of human infrastructure (research stations, roads, and airstrips) to degradation. Similarly to what is recorded in comparable regions in the Arctic, in non-consolidated deposits, thermokarst and landslide activity will likely increase. In the steep rocky slopes, so typical of the coastal areas of the Northwestern Antarctic Peninsula, as bedrock warms and thaw propagates along joints, rockfalls and landslides may become more frequent, increasing hazards in areas that are experiencing growing tourist visitation during the austral summer.”Point 2: It’s difficult to reconstruct the observational basis for each simulation from the information distributed between the main text and the appendices. A single site-characteristics table would be helpful. It should include logger type and accuracy, monitored depths, observation periods and gaps, slope aspect and inclination, surface and subsurface material, vegetation or organic cover, and available information on local snow conditions.
It is also unclear which air-temperature observations were used to correct ERA5 at Amsler, where the local air-temperature logger appears to have been installed only in 2023.Reply: We complemented the Table 1 with information on the site description and observational period. Information on the dataloggers will remain available on the appendix as their inclusion in Table 1 would compromise the readability due to the various modifications made over time. Additionally, the origin of the air temperature used to corrected ERA5 is also explained on the section Appendix A: Description of PERMANTAR monitoring sites.
“Air-temperature observations from 2016 to 2017 are available from a meteorological station that worked intermittently from 2012 to 2017.”
Point 3: As I understand the workflow, at least some subsurface parameters and the snowfall multipliers were selected by comparing simulations with the same ground-temperature records that were later used to report model performance. If it is correct, those statistics describe goodness of fit after calibration rather than performance against independent data, and they are likely to be optimistic.
Please separate calibration and validation periods, or apply a suitable temporal cross-validation procedure. Performance metrics should be presented independently for both periods. The duration and number of observations used at each site and depth should also be reported.Reply: We acknowledge that the separation of the calibration and evaluation periods can improve the interpretation of the procedure. To address this, the year 2021 was now used for calibration of the subsurface parameters and snowfall multiplier, while the period from the beginning of the observational records through 2020 was used for independent model evaluation. We have also specified in the revised manuscript (Table 1 and Section 3.2.2 Model calibration and evaluation), for each observatory, the calibration and evaluation periods, the observations used, and the depths considered. Separate tables reporting calibration and evaluation statistics have been added accordingly (Section 4.1 Sensitivity analysis to subsurface parameterization and validation of the surface energy model).
Point 4: I could not determine the full parameter space explored during calibration. Please report the tested ranges, increments, and selection criterion for mineral, organic, water, and ice fractions and for thermal conductivity. At line 155, the manuscript states that the sum of water and ice contents remains constant, but the value at which it is held constant is not defined.
It would be useful to distinguish parameters supported by direct observations from parameters inferred through fitting. Were porosity, fracture density, water or ice content, and related properties documented during drilling or measured on cores? If not, the adopted values should be treated as uncertain parameters rather than physically validated site properties. A sensitivity or ensemble analysis is required to assess parameter equifinality.Reply: With the changes implemented following previous comments, the methods were rewritten explaining the parameters adjusted during the calibration phase (mineral, water and ice content and snowfall multiplication factor). Thermal conductivity, as stated was defined based on the measurements performed on the rock cores collected from the boreholes. The sum of water and ice was derived from the volumetric content. Fracture density was observed during the core analysis.
Point 5: What is the basis for applying the same geothermal heat flux of 50 mW m⁻² at all five sites, and how deep is the lower model boundary? Is this value supported by regional measurements or previous studies? Since the principal results concern temperature at 20 m, sensitivity to geothermal heat flux, domain depth and initial temperature profile should be evaluated.
Reply: We re-run the simulations using geothermal heat flux values from An et al. (2015) for each observatory. These is now described on the methods section. The results showed minor differences.
“Measured ground surface temperature from the PERMANTAR monitoring sites was used as upper boundary input while a geothermal heat flux value obtained from (An et al., 2015) and applied as the lower boundary. Geothermal heat flux values for King Sejong, Papagal and SKO Meteo were 73 mWm-2, for Cierva were 61 mWm-2, and for Amsler were 58 mWm-2.”
Point 6: The boreholes are only 4–15 m deep, whereas the main projections concern MAGT at 20 m. At several sites, the calculated temperature at 20 m is therefore well outside the observed depth interval. The associated uncertainty should be explicitly quantified.
Positive MAGT at 20 m should not automatically be equated with complete permafrost loss. Please define the diagnostic criterion used to determine the presence or absence of permafrost. The manuscript should distinguish between: positive MAGT at a specified depth; loss of permafrost at that depth; active-layer deepening; and complete disappearance of the permafrost body.
Regional statements about widespread permafrost loss should be revised accordingly.Reply: We aimed to estimate the MAGT at a depth of zero annual amplitude and in most observatories, this is not reached at 10 m depth. The estimates at 10 m depth complement the analysis and allowed a comparison to the records in some of the observatories. We extended the explanation in the methods.
“In the simulation’s output we provide estimates of ground temperatures for a profile of 20 m depth, with vertical resolution varying with depth: 0.05 m from the surface to 1 m, 0.1 m from 1 to 2 m, 0.2 m from 2 to 5 m, and 0.5 m below 5 m depth. These are used to calculate the mean annual ground surface temperature (MAGST) and MAGT at 10 and 20 m depth. MAGT at 10 m depth provides an overview of near-surface thermal changes and their potential implications for surface and subsurface hydrology, biogeochemical fluxes, and geomorphological processes. MAGT at 20 m depth provides insight into long-term ground temperature conditions near the depth of zero annual amplitude.”
Sentences establishing a direct relationship between positive MAGT and permafrost degradation were removed.
Point 7: Snow is interpreted as a major control on differences among sites, but the snowfall multiplier was calibrated indirectly against ground temperature. Are any field observations of snow depth, snow-cover duration, or snow water equivalent available? If not, independent remote-sensing or photographic evidence should be considered.
The assumption of a constant multiplier of 0.3 or 0.5 throughout the future simulation also requires justification. Future changes in precipitation phase, wind redistribution, and snow duration may alter the relationship between large-scale precipitation and local accumulation. At minimum, the sensitivity of the projections to the snowfall multiplier should be shown.Reply: The snowfall multiplier was calibrated by comparing estimated and measured ground surface temperatures and assessing the resulting snow insulating effect. The selected value therefore corresponds to the multiplier that provided the best agreement between modelled and observed ground surface temperatures.
Section 3.2.2 Model calibration and evaluation:
“In the second phase of the calibration, the best configuration of volumetric content was used in a ground class (GROUND_freeW_seb_snow, Westermann et al., 2023) in which the phase change of water again occurs at 0 ºC and the water and ice contents remain constant. The surface energy balance scheme is applied to simulate energy exchange processes between the atmosphere and the model’s first grid cell (Baptista et al. 2025; Westermann et al. 2023). For this, the simulations were forced with bias-corrected ERA5 reanalysis (see Sect. 3.3.1)
(…)
When the ground is snow-covered, the ground class is stacked below a snow class (SNOW_crocus_bucketW_seb, (Westermann et al., 2023) that represents the seasonal snow microphysics scheme following the Crocus model (Vionnet et al., 2012; Zweigel et al., 2021). The snow model represents transient snow density changes due to compaction and wind drift, as well as meltwater infiltration and refreezing. Due to the micro topography at each monitoring site and to the spatial resolution of ERA5 variables, an over- or underestimation of snow depth can be produced by the model. Previous studies reported that ERA5 often overestimates snow cover in coastal Antarctic areas (Obu et al., 2020), due to unrepresented topography and ice-free vs glacier surfaces and wind redistribution effects. Therefore, ERA5-derived snowfall Ps was corrected for each site using a snowfall multiplication factor which allows a phenomenological increase or reduction of the simulated snow depth based on the intensity of the insulating effect and duration of the snow cover (Baptista et al., 2025; Martin et al., 2019). As for the first phase, the validation of the surface energy model and snowfall multiplication factor were performed through the comparison between the estimated and measured ground temperatures for 2021 at 4 depths and the analysis of the four statistical measures (bias, r, p-value, MAE and RMSE).”Section 4.1 Sensitivity analysis to subsurface parameterization and validation of the surface energy model:
“In the sensitivity analysis of surface parameters, on the second phase of calibration when the simulations were forced ERA5 reanalysis data, snowfall multiplication factor values were determined by comparing measured and estimated ground temperatures at a depth of 0.4 m (Table 5). Initial simulations using ERA5 snowfall data resulted in an overestimation of snow accumulation resulting in excess insulation. Consequently, snow multiplication fractions were reduced across all observatories, ranging from 30% to 50% of the original ERA5 snowfall input (Table 5).”
Snow depth measurements are not available at the PERMANTAR observatories. Remote-sensing approaches were previously explored within work conducted for the PERMANTAR network. However, they present substantial limitations in the Northwestern Antarctic Peninsula. Multispectral imagery is frequently affected by persistent cloud cover, resulting in limited temporal coverage. SAR data provide information on wet snow being their applicability largely restricted to the summer season, when most of the study areas are snow free. Furthermore, the spatial resolution of available satellite products would introduce considerable uncertainty at the observatory scale, where microtopography and wind redistribution strongly control snow distribution and, consequently, its insulating effect.
Concerning the future projections, as stated in the manuscript, we applied the model configuration defined through the sensitivity analysis, while acknowledging the limitations associated with its application under future climatic conditions.Section 3.2.2. Model calibration and evaluation
“The selected configuration was subsequently applied in the long-term simulation of past and future evolution of permafrost temperatures, forced by CMIP6 model and ERA5. However, its application under future climatic conditions involves uncertainty since CMIP6-projected climate in the Antarctic Peninsula may differ from the historical conditions used for calibration. Therefore, the model configuration should be understood as the best available representation of the present-day site conditions rather than as a fully constrained description of future conditions.”
Point 8: The phrase “adequate data to force the CCM” should be defined more precisely. Please state which required variables, temporal resolutions, ensemble members, and experiments were available for each CMIP6 model.
ACCESS-CM2 is selected mainly because its MAAT trend and variability resemble ERA5 at Papagal. The evaluation should include all five sites and the forcing variables most relevant to the surface energy balance, especially precipitation, radiation, and wind.
Projections based on one GCM cannot represent the full climate-model uncertainty.Reply: To clarify, a description of the variables required, and their temporal resolution was added to the manuscript.
Section 3.3.2 CMIP6 scenarios and models:
“The INM-CM4-8, the IPSL-CM6A-LR and the ACCESS-CM2 models from CMIP6 were evaluated, as they provide the required monthly variables to force the CCM: (a) near surface air temperature, (b) surface downwelling longwave radiation, (c) surface downwelling shortwave radiation, (d) precipitation, (e) near surface specific humidity, (f) near surface wind speed, (g) toa incident shortwave radiation, and (h) surface air pressure.”
Detailing the variables, temporal resolutions, ensemble members, and experiments available for all CMIP6 models considered during the screening process is not feasible considering the existence of least 58 CMIP6 models.
Regarding the evaluation of ACCESS-CM2, Papagal was selected because it is representative of the regional climatic conditions of the study area and provides a suitable observational record for model evaluation. Observations of precipitation, radiation, and wind are not available, and extending the evaluation would not necessarily result in a more robust model selection and could introduce inconsistencies in the comparison.“To evaluate the ability of the downscaled CMIP6 models to represent climate conditions in the northwestern Antarctic Peninsula, we compared the simulated MAATs at the Papagal observatory location with ERA5 reanalysis data for the reference period 1950–2020 (Fig. 2 and Table 3). Papagal observatory was selected as the reference site because it is located in an exposed location that is representative of regional climate conditions and provides a long-term, high-quality observational record.”
We acknowledge that projections based on a single GCM do not capture the full range of climate-model uncertainty. The objective of this study, however, was to investigate the sensitivity of permafrost using a climate forcing that best reproduces the observed temperature characteristics at the study area. A section acknowledging the limitation associated to the forcing has been added to the discussion.
Section 5.1 Evaluation of the forcing data:
“In the present study, following the evaluation of each CMIP6 model against the bias-corrected ERA5 air temperature for the period 1950-2020, ACCESS-CM2 was ultimately selected as the most suitable forcing model, as it showed the best overall agreement with ERA5, combining a realistic long-term temperature trend with a good representation of interannual variability. Future studies should simulate permafrost evolution for an ensemble of different CMIP models to account for the considerable inter-model spread in future climate conditions.”
Point 9: I found the construction of the submonthly forcing difficult to follow. When a historical ERA5 month is sampled as the carrier, are all meteorological variables taken together from the same month and year, or are variables sampled separately? The former is important for retaining covariance among temperature, humidity, precipitation, wind, and radiation.
The random carrier sequence is itself a source of uncertainty. I would expect different admissible sequences to produce some spread in annual ground temperatures and in the estimated timing of threshold crossings. Several realisations should be generated and propagated through the analysis, with the random seed and sampling procedure documented.Reply: The same month was used for all forcing variables to conserve the covariance. The sentence was rewritten for clarity. Additionally, in the discussion a new section was added referring the limitations associated with the forcing data.
“miCMIP is equal to MiCMIP during the overlap period, while it is set to the running mean of 10-year periods for the remainder of the time series, so that short-term fluctuations can be separated from multi-annual trends (which are accounted for by the first three terms on the right-hand side). To conserve the covariance between the different model forcing variables, they were all derived from the same month. With this procedure, we ensure that the corrected CMIP6 data features the same monthly mean and standard deviation as the ERA5 data for the overlap period. As “carrier” for the synthesized time series, we randomly sample individual months from the ERA5 time series for each month of the year, which provides somewhat realistic weather conditions on below-monthly timescale. Finally, the carrier data are again corrected on the basis of monthly averages with the corrected CMIP6 time series by adding the offset.”
Point 10: I could not find a description of the statistical method used to estimate temperature trends or the uncertainty reported after the ± sign. Please state the regression method, confidence level and significance criterion.
The four historical periods appear to have been selected visually. Either apply an objective change-point or segmented-regression method, or describe these intervals explicitly as qualitative periods. Decadal rates calculated for the short 2015–2022 interval should be interpreted cautiously.Reply: In the methods section, an explanation of the statistical method used to calculate temperature trends and uncertainty was added. Regarding the intervals, they are now described as qualitative periods.
Section 3.4 Simulation Long-term simulation of the past and future evolution of permafrost temperatures:
“The analysis of past permafrost temperature evolution focuses on the period 1950-2020, whereas the analysis of future projections begins in 2020. The temperature trends were estimated though the calculation of a least-squares linear regression. The associated uncertainty was obtained from the standard error of the slope.”
Point 11: I agree that elevation, snow persistence, wind exposure, sea-ice decline, and changes in the Amundsen Sea Low are plausible explanations for the simulated contrasts. However, the present experiments do not isolate their individual effects. Unless additional factorial simulations or atmospheric indices are analysed, the language should be changed from causal statements to cautious interpretations such as “may be associated with” or “is consistent with”.
Reply: Corrected.
Example in Section 5.3 Future evolution of permafrost temperature (2020 - 2100):
“In contrast to the SSP1-2.6 scenario, SSP2-4.5 displays a distinct spatial pattern in MAAT warming rates, with increased warming at the more southerly observatories which may be related to the sea ice loss on the Bellingshausen Sea and consequent deepening of the ASL that strengthens north-northwesterly flow towards the Antarctic Peninsula, favoring advection of warm, maritime air and raising surface air temperatures, especially during the winter (Abram et al., 2010; Clem et al., 2017; Hosking et al., 2013; Turner et al., 2013)”
Point 12: Please cross-check the abstract, Results, Discussion, Table 5, and Conclusions. There are inconsistencies in the reported years when MAGT becomes positive and in the use of “warming” versus “warming rate”. For example, values such as 0.59, 2.59, and 5.05 °C appear to represent total changes by 2100 in some places but are described as rates elsewhere. The statement that positive MAGT occurs at most sites by 2045 under SSP5-8.5 is also not consistent with the site-specific results. The title of Sect. 4.3 refers to active-layer thickness, but active-layer results are not presented.
Reply: The inconsistencies were corrected as the warming rates now refer as decadal.
Reply to the minor comments:
At first use, spell out CryoGrid Community Model (CCM), and then use the abbreviation consistently.
Reply: Corrected.
The manuscript uses “Recommended Concentration Pathways”; this should be corrected.
Reply: Corrected.
Please standardise units and notation, including °C decade⁻¹, W m⁻¹ K⁻¹, and m a.s.l.
Reply: Corrected.
TTOP needs a precise definition, including how it is diagnosed from the model output and how it is interpreted at sites without confirmed permafrost within the borehole.
Reply: TTOP was removed from the analysis.
Replace “biochemical fluxes” with “biogeochemical fluxes”.
Reply: Corrected.
Increase the font and panel size in Figs. 2, 3, 5, and D1. Several time series and legends are difficult to read.
Reply: Corrected.
References
Abram, N. J., Thomas, E. R., McConnell, J. R., Mulvaney, R., Bracegirdle, T. J., Sime, L. C., and Aristarain, A. J.: Ice core evidence for a 20th century decline of sea ice in the Bellingshausen Sea, Antarctica, J. Geophys. Res. Atmos., 115, 2010JD014644, https://doi.org/10.1029/2010JD014644, 2010.
An, M., Wiens, D. A., Zhao, Y., Feng, M., Nyblade, A., Kanao, M., Li, Y., Maggi, A., and Lévêque, J.: Temperature, lithosphere‐asthenosphere boundary, and heat flux beneath the Antarctic Plate inferred from seismic velocities, J. Geophys. Res. Solid Earth, 120, 8720–8742, https://doi.org/10.1002/2015JB011917, 2015.
Baptista, J., Brito Guapo Teles Vieira, G., Manuel De Carvalho Soares Correia, A., Lee, H., and Westermann, S.: Modelling the evolution of permafrost temperatures and active layer thickness in King George Island, Antarctica, since 1950, The Cryosphere, 19, 3459–3476, https://doi.org/10.5194/tc-19-3459-2025, 2025.
Clem, K. R., Renwick, J. A., and McGregor, J.: Large-Scale Forcing of the Amundsen Sea Low and Its Influence on Sea Ice and West Antarctic Temperature, J. Clim., 30, 8405–8424, https://doi.org/10.1175/JCLI-D-16-0891.1, 2017.
Hosking, J. S., Orr, A., Marshall, G. J., Turner, J., and Phillips, T.: The Influence of the Amundsen–Bellingshausen Seas Low on the Climate of West Antarctica and Its Representation in Coupled Climate Model Simulations, J. Clim., 26, 6633–6648, https://doi.org/10.1175/JCLI-D-12-00813.1, 2013.
Martin, L., Nitzbon, J., Aas, K. S., Etzelmüller, B., Kristiansen, H., and Westermann, S.: Stability Conditions of Peat Plateaus and Palsas in Northern Norway, J. Geophys. Res. Earth Surf., 124, 705–719, https://doi.org/10.1029/2018JF004945, 2019.
Obu, J., Westermann, S., Vieira, G., Abramov, A., Balks, M., Bartsch, A., Hrbáček, F., Kääb, A., and Ramos, M.: Pan-Antarctic map of near-surface permafrost temperatures at 1 km2 scale, The Cryosphere, https://doi.org/10.5194/tc-2019-148, 2020.
Turner, J., Phillips, T., Hosking, J. S., Marshall, G. J., and Orr, A.: The Amundsen Sea low, Int. J. Climatol., 33, 1818–1829, https://doi.org/10.1002/joc.3558, 2013.
Vionnet, V., Brun, E., Morin, S., Boone, A., Faroux, S., Le Moigne, P., Martin, E., and Willemet, J.-M.: The detailed snowpack scheme Crocus and its implementation in SURFEX v7.2, Geosci. Model Dev., 5, 773–791, https://doi.org/10.5194/gmd-5-773-2012, 2012.
Westermann, S., Ingeman-Nielsen, T., Scheer, J., Aalstad, K., Aga, J., Chaudhary, N., Etzelmüller, B., Filhol, S., Kääb, A., Renette, C., Schmidt, L. S., Schuler, T. V., Zweigel, R. B., Martin, L., Morard, S., Ben-Asher, M., Angelopoulos, M., Boike, J., Groenke, B., Miesner, F., Nitzbon, J., Overduin, P., Stuenzi, S. M., and Langer, M.: The CryoGrid community model (version 1.0) – a multi-physics toolbox for climate-driven simulations in the terrestrial cryosphere, Geosci. Model Dev., 16, 2607–2647, https://doi.org/10.5194/gmd-16-2607-2023, 2023.
Zweigel, R. B., Westermann, S., Nitzbon, J., Langer, M., Boike, J., Etzelmüller, B., and Vikhamar Schuler, T.: Simulating Snow Redistribution and its Effect on Ground Surface Temperature at a High‐Arctic Site on Svalbard, J. Geophys. Res. Earth Surf., 126, e2020JF005673, https://doi.org/10.1029/2020JF005673, 2021.
-
AC3: 'Reply on RC4', Joana Baptista, 21 Sep 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 152 | 79 | 29 | 260 | 19 | 18 |
- HTML: 152
- PDF: 79
- XML: 29
- Total: 260
- BibTeX: 19
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This is an important and timely contribution on the recent past and future ground temperature dynamics in the Northwestern Antarctic Peninsula. Not only is there no similar study to date, but the numerical modelling approach used is novel in Antarctica. However, I have several comments and recommendations that should be addressed before the manuscript can be considered for publication.
(1) The simulation of recent past conditions was based on the bias-corrected ERA5 for 1950–2020 (with 10 spin-up loops in 1940–1949). By contrast, the simulation of future conditions was based on the bias-corrected CMIP6 models for 1850–2100, with 10 spin-up loops in 1850–1860 and bias-corrected ERA5 used in 1940–2022. Why did not you apply the modelling strategy used for future to recent past as well? The 100-year model initialization phase running since 1850 would likely be better than the 10 spin-up loops over 1940–1949. Moreover, this would mean better consistency and less work since you would have single time series covering the whole period 1850–2100 and both recent past and future scenarios.
Additionally, was the CMIP6 bias correction based on the bias-corrected ERA5? This is unclear to me.
(2) Considering that you did the bias correction only in the case of air temperature, while the rest of the variables retained their uncorrected values, is the resulting dataset internally consistent? For instance, if the air temperature was corrected, while the dew point temperature remained uncorrected, this may affect outputs calculated from these two variables. Likewise, precipitation retained its uncorrected value, but solid and liquid precipitation usually occurs at some specific temperatures, which can also be affected if the temperature is corrected. I have no idea how this might affect the simulation outputs, but the issue is here obvious.
(3) In the methods, explain why you modelled ground temperatures specifically at a depth of 20 m. Frequently, you also present ground temperatures from a depth of 10 m, which is much closer to a maximum depth of the boreholes and might therefore have a higher validity. Was this because that is at or below the depth of zero annual amplitude?
(4) For consistency, descriptions of both recent past and future simulations should present the evolution of mean annual air temperatures, mean annual ground surface temperatures, temperatures at the base of the freeze-layer layer (TTOP under permafrost conditions), and mean annual ground temperatures at a depth of 20 m. Some of these temperature metrics are frequently missing, which makes a little sense in terms of consistency and (not only) mutual context. If mean annual ground temperatures at a depth of 10 m are shown as well, I would present them before those from 20 m. Because of heat diffusion, the temperature patterns at 10 m precede those at 20 m, so it would have more logic to proceed in this order.
(5) The discussion is relatively short and in principle deals with results of this study alone, and the context of other studies is provided to a very limited extent. While I am aware that there is currently no analogous study published for Antarctica, there are several studies that assessed short-term trends in permafrost temperatures or active-layer thickness (e.g., Guglielmin and Cannone, 2012; Ramos et al., 2017; Biskaborn et al., 2019; Hrbáček and Uxa, 2020; Hrbáček et al., 2021, 2023, 2025; Baptista et al., 2025) or some several long-term meteorological series (e.g., Oliva et al., 2017). I do not understand why you do not mention your previous study in the discussion (Baptista et al., 2025).
What I also miss in the discussion is a section devoted to the evaluation of the ERA5 and CMIP6 models used as model forcings. Consequently, I would move the Appendix C into the discussion. You may also refer to Baptista et al. (2025) or Kaplan Pastíriková et al. (2025) when evaluating the suitability of ERA5 or reananalyses for permafrost studies in Antarctica in general.
(6) Consider carefully whether the content of the appendices really needs to placed in them. I think that most of the appendices could be involved in the main body of the manuscript without making it fragmented.
Specific comments:
P1L13: State the period over which the mean annual air temperature increase of 3.4±1.2 °C was recorded.
P1L18: Specify the depth for the mean annual ground temperature. Is it 20 m?
P1L20: In contrast to the above, it is unnecessary to specify the depth of 20 m here because it is a general statement.
P2L42–43: Again, state the period over which the mean annual air temperature increase of 3.4±1.2 °C was recorded. Considering that mean annual air temperature can exhibit substantial variations from year to year, is it appropriate to use a single year as a baseline value?
P2L50: Change (Obu et al., 2020) to Obu et al. (2020).
P2L57–58: In the aims of the manuscript, be specific and clearly define the time frames for both the recent past and the future modelling periods.
P3L61–62: This previous study (Babtista et al., 2025) should be briefly described before the aims of the manuscript, so that the objectives follow logically from it. I think that, in this case, it disrupts the structure of the paragraph devoted to the aims.
P3L76: Change (Thomas e Tetzner 2019) to (Thomas and Tetzner 2019).
P4L96: Considering the substantial warming trend, the mean annual air temperature could be reported for a shorter, more recent period to better represent current climatic conditions. Of course, this depends on your intentions.
P5L100–101: Rewrite the sentence “… the highest areas of Hurd Peninsula were deglaciated at 20 ka with most high interfluves becoming ice-free around 14-16 ka …” so that it is clear when these locations were actually deglaciated. Was it at 20 ka or around 16-14 ka?
P5L101: Change 14-16 ka to 16-14 ka.
P5L106: State the period for which the mean annual air temperature and mean annual precipitation are reported.
P5L115–116: State the period for which the mean annual air temperature and mean annual precipitation are reported.
P5L124–125: State the period for which the mean annual air temperature and mean annual precipitation are reported.
P6L127–140: Move the whole section 3.1 to the section 2.2. The details on permafrost temperature and active-layer thickness at individual sites can be presented in the sections 2.2.1 to 2.2.4.
P7L151–152. This sentence on the calibration strategy is difficult to understand. Rewrite it even though it is partly obvious from the following text.
P7L156: What is the geothermal heat flux of 50 mWm^-2 based on? Please cite the source.
P7L157: What does the phrase “With the structure set, …” mean?
P7L160: Is the supplement “… using a probe.” necessary?
P7L164: What do you mean by “relative volume fraction” here?
P7L171: Indicate that Fub(t) is the energy flux into the uppermost grid cell. The rest of the variables (radiation and heat fluxes) would be better to describe after Eq. (1).
P7L175: “The heat conduction based on Fourier’s law is the main mode of heat transport in the subsurface.” is too brief description of the subsurface heat transfer.
P8L183: Change “phenomenologically” to “phenomenological”.
P8L186: State also that you calculated p-value.
P9L219: I guess there should be “SPPs” in “The RCPs are used to represent 4 pathways…”
P10L230: Note that the SPP5-8.5 scenario is now considered an extreme case that is rather implausible. You should point this out in the manuscript (here and in the discussion).
P10L235: Remove the word “models”.
P10L243–245: Place this paragraph at the end of the whole section 3.3.2.
P10L253: I would remove “…in the selected PERMANTAR monitoring sites.” because it is redundant here.
P11L258: The vertical resolution of the model is unclear from this: “… from 0.05 to 0.5 m, below 5 m depth.” Revise it.
P11L259: Specify the depths at which you calculated the mean annual ground temperature.
P11L266–269: Rewrite this so it is more clear that you bias-corrected the period 1850–2100 from CMIP6 using the statistical relationships with ERA5 for the sub-period 1940–2020.
P12L300: Remove the sentence “In the figures, the x-axis spans the period 2000-2100.”
P13L326: In Table 3, you should indicate that the thermal conductivity relates to the rock.
P14L354: In Fig. 2, the range seems to be somewhat smaller than -5 to 5 °C.
P16L364–380: In addition to the trends, present also the temperatures or their ranges. This is only done at three stations for TTOP and at all of them for MAGT. Report also the values and trends from the mean annual air temperature (and maybe snowfall if both are shown in Fig. 3).
P16L381: In Table 4, specify the depth for the mean annual ground temperature.
P17L385: Fig. 3 has a misleading caption because snowfall and air temperature cannot be referred to as ground temperatures.
P18L388: In Fig. 4, I think it is a bit misleading to present the temperature for a single year 1950 because it was exceptionally cold. Is it appropriate to use a single year here, at least for the mean annual ground surface temperature, which has an instantaneous response to air temperature?
P18–19L391–416: Check that it is clearly stated everywhere what temperature the reported trends refer to.
P18L409: Rewrite the sentence “Despite continued interannual air temperature variability, MAGST fluctuated between -4 to -1 °C.” which does not make complete sense.
P19L412: It is unclear what temperature the trend of 2.10±0.60 °C refers to.
P19L423: You should rather write for 2020–2100. Trends should always refer to a period, not a single year. Additionally, it is unclear what the temperature trend of 0.59±0.22 °C refers to.
P20L427: Change “… being the warming more pronounced at the …” to “… with more pronounced warming at the…”
P20L429: Remove “For most of the observatories, “ Then, the sentence will make more sense.
P20L434–437: Maybe move this paragraph before the previous one. Definitely, it is better to start with the statement that all stations host permafrost at present, and then describe that it will warm and likely degrade or disappear at some of them.
P20L439: Again, it is unclear what the first trend of 2.59±0.20 °C refers to.
P21L469–470: Change “At Cierva despite a warming rate of 0.51 ± 0.01 °C.dec-1 positive MAGT is projected…” to “At Cierva, despite a warming rate of 0.51 ± 0.01 °C.dec-1, positive MAGT is projected…”
P21L476: Table 5 has a misleading caption because it presents other temperature variables in addition to the mean annual ground temperature at 20 m.
P22L480: Fig. 5 has a misleading caption because air temperature cannot be referred to as ground temperature.
P24L505–509: The interpretation that ground temperature warming in the periods 1950–1975 and 2015–2020 was favoured by decreased snowfall and shorter duration of the snow cover contradicts former explanations that thicker and longer snow cover tends to warm the ground. Likewise, cooling simulated for 2000–2015 is difficult to explain by reduced snowfall.
P24L513–514: Remove the first sentence in the section 5.2; it is redundant.
P24L516: Why are the periods 2000–2015 and 2016–2022 discussed in the section entitled “Future evolution of permafrost temperature (2022 - 2100)”?
P25L532: Change “Under the more optimistic SSP1-2.6 scenario, …” to “Under the SSP1-2.6 scenario, …”
P25L539: Add “with values ranging from 1 to 2 °C in 2100.”
P26L557: Specify to what depth the bias refers to.
P26L574: Change “In the worst-case scenario (SSP5-8.5), …” to “Under the SSP5-8.5 scenario, …”
P27L582: Change “thermokarst and landslide activity should increase” to “thermokarst and landslide activity will likely increase”
P31L666: Change “An historical period …” to “A historical period …”
P31L674: Change “improved coupled between” to “improved coupling between”
P31L679: Change “a RMSE of 1.33 °C. and a standard deviation” to “a RMSE of 1.33 °C and a standard deviation”
References
Baptista, J., Brito Guapo Teles Vieira, G., Manuel De Carvalho Soares Correia, A., Lee, H., Westermann, S., 2025. Modelling the evolution of permafrost temperatures and active layer thickness in King George Island, Antarctica, since 1950, The Cryosphere, 19, 3459–3476, https://doi.org/10.5194/tc-19-3459-2025
Biskaborn, B.K., Smith, S.L., Noetzli, J., Matthes, H., Vieira, G., Streletskiy, D.A., Schoeneich, P., Romanovsky, V.E., Lewkowicz, A.G., Abramov, A., Allard, M., Boike, J., Cable, W.L., Christiansen, H.H., Delaloye, R., Diekmann, B., Drozdov, D., Etzelmüller, B., Grosse, G., Guglielmin, M., Ingeman-Nielsen, T., Isaksen, K., Ishikawa, M., Johansson, M., Johansson, H., Joo, A., Kaverin, D., Kholodov, A., Konstantinov, P., Kröger, T., Lambiel, Ch., Lanckman, J.-P., Luo, D., Malkova, G., Meiklejohn, I., Moskalenko, N., Oliva, M., Phillips, M., Ramos, M., Sannel, A.B.K., Sergeev, D., Seybold, C., Skryabin, P., Vasiliev, A., Wu, Q., Yoshikawa, K., Zheleznyak, M., Lantuit, H., 2019. Permafrost is warming at a global scale. Nature Communications, 10, 264. https://doi.org/10.1038/s41467-018-08240-4
Guglielmin, M., Cannone, N., 2012. A permafrost warming in a cooling Antarctica? Climate Changem, 111, 177–195. https://doi.org/10.1007/s10584-011-0137-2
Hrbáček, F., Uxa, T., 2020. The evolution of a near-surface ground thermal regime and modeled active-layer thickness on James Ross Island, eastern Antarctic Peninsula, in 2006–2016. Permafrost and Periglacial Processes, 31, 141–155. https://doi.org/10.1002/ppp.2018
Hrbáček, F., Vieira, G., Oliva, M., Balks, M., Guglielmin, M., de Pablo, M.Á., Molina, A., Ramos, M., Goyanes, G., Meiklejohn, I., Abramov, A., Demidov, N., Fedorov-Davydov, D., Lupachev, A., Rivkina, E., Láska, K., Kňažková, M., Nývlt, D., Raffi, R., Strelin, J., Sone, T., Fukui, K., Dolgikh, A., Zazovskaya, E., Mergelov, N., Osokin, N., Miamin, V., 2021. Active layer monitoring in Antarctica: an overview of results from 2006 to 2015. Polar Geography, 44, 217–231. https://doi.org/10.1080/1088937X.2017.1420105
Hrbáček, F., Oliva, M., Hansen, C., Balks, M., O'Neill, T. A., de Pablo, M. A., Ponti, S., Ramos, M., Vieira, G., Abramov, A., Kaplan Pastíriková, L., Guglielmin, M., Goyanes, G., Rocha Francelino, M., Schaefer, C., Lacelle, D., 2023. Active layer and permafrost thermal regimes in the ice-free areas of Antarctica, Earth-Science Reviews, 242, 104458, https://doi.org/10.1016/j.earscirev.2023.104458
Hrbáček, F., Kňažková, M., Láska, K., Kaplan Pastíriková, L., 2025. Active Layer Warming and Thickening on CALM‐S JGM, James Ross Island, in the Period 2013/14–2022/23, Permafrost and Periglacial Processes, 36, 378–389, https://doi.org/10.1002/ppp.2274
Kaplan Pastíriková, L., Hrbáček, F., Matějka, M., 2025. Validation of ERA5-Land-based reconstructed air temperature and near-surface ground temperature on James Ross Island. Polar Geography, 48, 95–115, https://doi.org/10.1080/1088937X.2024.2434744
Oliva, M., Navarro, F., Hrbáček, F., Hernández, A., Nývlt, D., Pereira, P., Ruiz-Fernández, J., Trigo, R., 2017. Recent regional climate cooling on the Antarctic Peninsula and associated impacts on the cryosphere. Science of the Total Environment, 580, 210–223. https://doi.org/10.1016/j.scitotenv.2016.12.030
Ramos, M., Vieira, G., de Pablo, M.A., Molina, A., Abramov, A., Goyanes, G., 2017. Recent shallowing of the thaw depth at Crater Lake, Deception Island, Antarctica (2006–2014). Catena, 149, 519–528. https://doi.org/10.1016/j.catena.2016.07.019
All the best,
Tomáš Uxa