Previous Article in event
Previous Article in session
Next Article in event
Next Article in session
Projecting the Potential Evapotranspiration of Egypt using a high-resolution regional climate model (RegCM4)
Published:
22 July 2022
by MDPI
in The 5th International Electronic Conference on Atmospheric Sciences
session Climatology
Abstract:
This study aims to use the regional climate model (RegCM4) to examine the influence of climate change on potential evapotranspiration (PET) of Egypt under two future scenarios. To address such a topic, the calculated PET is first corrected in the historical period with respect to the long-term gridded PET data (Climate Research Unit; CRU) using a linear regression model (LRM) between RegCM4 and CRU. After that, the LRM is used to correct the two future scenarios Representative Concentration Pathways (RCP4.5 and 8.5) of the period 2006-2100. The RegCM4 was downscaled by the medium resolution of the Earth System Model of the Max Planck Institute (MPI-ESM-MR) with 50 km horizontal grid spacing over Middle East and North Africa (MENA) and then nested over Egypt with 20 km horizontal grid spacing. The results showed that the RegCM4 is able to capture the monthly variability of PET with respect to the CRU; furthermore the RegCM4 overestimates/underestimates the PET depending on the location under consideration. Also, the simulated PET was notably improved when the LRM was used. Such improvement is indicated by a low mean bias and a high standard deviation ratio (close to unity) between the corrected PET and CRU. In addition, the future PET projects a strong increased trend under the RCP8.5 scenario; meanwhile the future PET projects a weak increased trend under the RCP4.5 scenario.
Keywords: Egypt; Regional climate model; Potential evapotranspiration; Climate Research Unit
Supplementary material: Download Supplementary material
View paper
View Poster
Download (presentation ppt)
Comments on this paper
Samy Anwar
23 July 2022
Dear All,
Good afternoon. Hope all are doing well. I am Samy and I am the corresponding author of this paper.
I noticed that the published file doesn't contain either the figures or the table. Therefore, I uploaded both in the attached zip file as a supplementary file.
Hope you can find the paper interesting.
Best Regards
Samy
Good afternoon. Hope all are doing well. I am Samy and I am the corresponding author of this paper.
I noticed that the published file doesn't contain either the figures or the table. Therefore, I uploaded both in the attached zip file as a supplementary file.
Hope you can find the paper interesting.
Best Regards
Samy
Ankur Srivastava
24 July 2022
Hi,
Nice work but have a few suggestions:
1. If the equations from the introduction can be shifted to the methodology
2. Also I am not aware of the structure of this journal, so is it is possible to keep figures in the results section of the paper. I found only the text because there are no figures in the main paper.
3. The authors have no described of using CRU data over other existing datasets.
4. It would be better to add a flowchart of the preprocessing of the datasets.
Nice work but have a few suggestions:
1. If the equations from the introduction can be shifted to the methodology
2. Also I am not aware of the structure of this journal, so is it is possible to keep figures in the results section of the paper. I found only the text because there are no figures in the main paper.
3. The authors have no described of using CRU data over other existing datasets.
4. It would be better to add a flowchart of the preprocessing of the datasets.
Samy Anwar
24 July 2022
Dear Ankur,
Many thanks for your comments and time to review the manuscript. Here, I reply to the suggested comments:
1 - Since I talk about potential evapotranspiration and its definitation, so it is better to be in the introduction section. Also, I refered to the Hargreaves–Samani method (equation I used to compute the PET) in different sections in the manuscript).
2 - I post a comment to illustrate that the published manuscript file doesn't contain figures and table. Therefore, I uploaded them in a zip file as a supplementary file.
3 - Because of inavailability of long-term records of station data, so I used CRU data as an alternative source of global PET (IPCC 2007) and there was no opportunity to compare between CRU and other gridded products with respect to station data. To account for the uncertainty associated with gridded-based products, I wrote the following sentence in the manuscript:
Evaluating the RegCM4 with high-resolution gridded PET product (e.g., Singer et al. 2020 and 2021) along with the CRU to account for the uncertainty of the observational-based dataset.
4 - We didn't use a specific flow-chart to process the data. The RegCM4 was converted from curvilinear to regular lat-lon grid using CDO package. To retrieve a time series either from the RegCM4 or CRU, NCO package was used for this purpose.
Many thanks for your comments and time.
Best Regards
Samy
Many thanks for your comments and time to review the manuscript. Here, I reply to the suggested comments:
1 - Since I talk about potential evapotranspiration and its definitation, so it is better to be in the introduction section. Also, I refered to the Hargreaves–Samani method (equation I used to compute the PET) in different sections in the manuscript).
2 - I post a comment to illustrate that the published manuscript file doesn't contain figures and table. Therefore, I uploaded them in a zip file as a supplementary file.
3 - Because of inavailability of long-term records of station data, so I used CRU data as an alternative source of global PET (IPCC 2007) and there was no opportunity to compare between CRU and other gridded products with respect to station data. To account for the uncertainty associated with gridded-based products, I wrote the following sentence in the manuscript:
Evaluating the RegCM4 with high-resolution gridded PET product (e.g., Singer et al. 2020 and 2021) along with the CRU to account for the uncertainty of the observational-based dataset.
4 - We didn't use a specific flow-chart to process the data. The RegCM4 was converted from curvilinear to regular lat-lon grid using CDO package. To retrieve a time series either from the RegCM4 or CRU, NCO package was used for this purpose.
Many thanks for your comments and time.
Best Regards
Samy
Samy Anwar
24 July 2022
Dear Ankur,
Many thanks for your comments and time to review the manuscript. Here, I reply to the suggested comments:
1 - Since I talk about potential evapotranspiration and its definitation, so it is better to be in the introduction section. Also, I refered to the Hargreaves–Samani method (equation I used to compute the PET) in different sections in the manuscript).
2 - I post a comment to illustrate that the published manuscript file doesn't contain figures and table. Therefore, I uploaded them in a zip file as a supplementary file.
3 - Because of inavailability of long-term records of station data, so I used CRU data as an alternative source of global PET (IPCC 2007) and there was no opportunity to compare between CRU and other gridded products with respect to station data. To account for the uncertainty associated with gridded-based products, I wrote the following sentence in the manuscript:
Evaluating the RegCM4 with high-resolution gridded PET product (e.g., Singer et al. 2020 and 2021) along with the CRU to account for the uncertainty of the observational-based dataset.
4 - We didn't use a specific flow-chart to process the data. The RegCM4 was converted from curvilinear to regular lat-lon grid using CDO package. To retrieve a time series either from the RegCM4 or CRU, NCO package was used for this purpose.
Many thanks for your comments and time.
Best Regards
Samy
Many thanks for your comments and time to review the manuscript. Here, I reply to the suggested comments:
1 - Since I talk about potential evapotranspiration and its definitation, so it is better to be in the introduction section. Also, I refered to the Hargreaves–Samani method (equation I used to compute the PET) in different sections in the manuscript).
2 - I post a comment to illustrate that the published manuscript file doesn't contain figures and table. Therefore, I uploaded them in a zip file as a supplementary file.
3 - Because of inavailability of long-term records of station data, so I used CRU data as an alternative source of global PET (IPCC 2007) and there was no opportunity to compare between CRU and other gridded products with respect to station data. To account for the uncertainty associated with gridded-based products, I wrote the following sentence in the manuscript:
Evaluating the RegCM4 with high-resolution gridded PET product (e.g., Singer et al. 2020 and 2021) along with the CRU to account for the uncertainty of the observational-based dataset.
4 - We didn't use a specific flow-chart to process the data. The RegCM4 was converted from curvilinear to regular lat-lon grid using CDO package. To retrieve a time series either from the RegCM4 or CRU, NCO package was used for this purpose.
Many thanks for your comments and time.
Best Regards
Samy
Samy Anwar
24 July 2022
Dear All,
Good evening. Hope you are doing well. Just wanted to report an important issue: "It is important to highlight that the calculating the PET (using the PM method) is feasible using version 4.7 of the RegCM and afterwards. However, such calculation hasn’t been tested yet. Therefore, the HG was used instead to compute the PET as recommended by Murat et al. (2018) and Srivastava et al. (2018). Another reason to use the HG method is that Egypt is characterized by hot dry summer and mild winter as reported in section (2.1); which make is the HG is a good choice for computing the PET in this study. "
Best Regards
Samy
Good evening. Hope you are doing well. Just wanted to report an important issue: "It is important to highlight that the calculating the PET (using the PM method) is feasible using version 4.7 of the RegCM and afterwards. However, such calculation hasn’t been tested yet. Therefore, the HG was used instead to compute the PET as recommended by Murat et al. (2018) and Srivastava et al. (2018). Another reason to use the HG method is that Egypt is characterized by hot dry summer and mild winter as reported in section (2.1); which make is the HG is a good choice for computing the PET in this study. "
Best Regards
Samy
ahmed shalaby
25 July 2022
Hi, Authors
This is study is significant to agricultural activity in Egypt that is vulnerable to global warming,
I have few questions
1- RegCM4 uses PM equation to calculate ET, why does PM fail to capture the ET climatology over such hyper-arid climate zone?
2-What is the land-surface processes that could be attributed to the different pattern or response of these different location?
in other word, how do land and vegetation cover of these location determine the PET in increasing temperature environment?
3-What are the advantages of CRU over MODIS in PET data?
Minor comments
1-In table one, it might be better, instead to write the full regression equation for each location, just write a formal equation at the table's head (y=ax+b) and then refer to a and b pair, this would be easier to read
This is study is significant to agricultural activity in Egypt that is vulnerable to global warming,
I have few questions
1- RegCM4 uses PM equation to calculate ET, why does PM fail to capture the ET climatology over such hyper-arid climate zone?
2-What is the land-surface processes that could be attributed to the different pattern or response of these different location?
in other word, how do land and vegetation cover of these location determine the PET in increasing temperature environment?
3-What are the advantages of CRU over MODIS in PET data?
Minor comments
1-In table one, it might be better, instead to write the full regression equation for each location, just write a formal equation at the table's head (y=ax+b) and then refer to a and b pair, this would be easier to read
Samy Anwar
25 July 2022
Dear Dr Ahmed,
Many thanks for your questions. I will be happy to answer them:
1 - the RegCM4 doesn't use the PM method to compute the ET, instead it calculates the components of ET as soil evaporation, vegetation evaporation and vegetation transpiration for each vegetation or soil land unit. In this study, such calculation is restricted to delta and Nile river regions only (where vegetation exists). Also, it is written in the manuscript that calculating the PET using PM is not recommended in arid/hyper arid regions because it requires a surplus of soil moisture (which doesn't exist in the study region). Moreover, from version 4.7 of the RegCM4 and upcoming versions, calculation of PET using PM is not fully tested. Therefore, the HG equation is used instead as recommended by FAO.
2 - In the static data (read by the RegCM4), desert covers majority of Egypt, therefore calculating ET is only feasible over delta and Nile river regions otherwise it is zero. Therefore, calculating PET is mainly based on meteorological variables: mean air temperature and global incident solar radiation without need to consider the soil moisture status particularly in arid and hyper arid regions (as in this study).
3 - MODIS-PET is mainly based on considering NDVI to calculate PET (which is only restricted to delta and Nile river regions as discussed in points 1 and 2). Also, MODIS data is integrated from 2000 to present. On the other hand CRU data is available over all land units considered in the present study and it is based on upscaling of station observation around the globe and it is integrated from 1901 to present. Also IPCC (2007) and other literature reviews recommended CRU-PET as the best source for assessment of PET across the globe. Therefore, CRU product was used in this study.
4 - Thanks for your notice, it will be considered when the revision is available.
Many thanks for your time.
Best Regards
Samy
Many thanks for your questions. I will be happy to answer them:
1 - the RegCM4 doesn't use the PM method to compute the ET, instead it calculates the components of ET as soil evaporation, vegetation evaporation and vegetation transpiration for each vegetation or soil land unit. In this study, such calculation is restricted to delta and Nile river regions only (where vegetation exists). Also, it is written in the manuscript that calculating the PET using PM is not recommended in arid/hyper arid regions because it requires a surplus of soil moisture (which doesn't exist in the study region). Moreover, from version 4.7 of the RegCM4 and upcoming versions, calculation of PET using PM is not fully tested. Therefore, the HG equation is used instead as recommended by FAO.
2 - In the static data (read by the RegCM4), desert covers majority of Egypt, therefore calculating ET is only feasible over delta and Nile river regions otherwise it is zero. Therefore, calculating PET is mainly based on meteorological variables: mean air temperature and global incident solar radiation without need to consider the soil moisture status particularly in arid and hyper arid regions (as in this study).
3 - MODIS-PET is mainly based on considering NDVI to calculate PET (which is only restricted to delta and Nile river regions as discussed in points 1 and 2). Also, MODIS data is integrated from 2000 to present. On the other hand CRU data is available over all land units considered in the present study and it is based on upscaling of station observation around the globe and it is integrated from 1901 to present. Also IPCC (2007) and other literature reviews recommended CRU-PET as the best source for assessment of PET across the globe. Therefore, CRU product was used in this study.
4 - Thanks for your notice, it will be considered when the revision is available.
Many thanks for your time.
Best Regards
Samy
Samy Anwar
26 July 2022
Dear Prof Lupo,
Good evening. Hope you are doing well. I noticed a difference between the manuscript I submitted with the one published online. I sent the conference secretary to inform them concerning this issue. Can you check with them? As I sent two e-mails but didn't receive any response yet.
Thanks in advance
Best Regards
Samy
Good evening. Hope you are doing well. I noticed a difference between the manuscript I submitted with the one published online. I sent the conference secretary to inform them concerning this issue. Can you check with them? As I sent two e-mails but didn't receive any response yet.
Thanks in advance
Best Regards
Samy
Shabnam Pourshirazi
26 July 2022
Dear authors,
It is an excellent manuscript with very good results, which will greatly help PET measurements in the future, considering the onset of global warming in hot and dry countries like Egypt.
I had some criticisms and questions as follows:
1. It is better not to mention the equations in the introduction (just noting who used them is enough), mention them in the M&M, and compare them with the results of other equations in the discussion section.
2. I think it is better to show Figure 1B in Figure 1A (merge both).
3. How reliable is the CRU dataset, and was there another data source that could have been used?
4- It is better to Equation 4 to M&M and only shows the figures in the results.
Thanks
It is an excellent manuscript with very good results, which will greatly help PET measurements in the future, considering the onset of global warming in hot and dry countries like Egypt.
I had some criticisms and questions as follows:
1. It is better not to mention the equations in the introduction (just noting who used them is enough), mention them in the M&M, and compare them with the results of other equations in the discussion section.
2. I think it is better to show Figure 1B in Figure 1A (merge both).
3. How reliable is the CRU dataset, and was there another data source that could have been used?
4- It is better to Equation 4 to M&M and only shows the figures in the results.
Thanks
Samy Anwar
26 July 2022
Dear Shabnam,
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Samy Anwar
26 July 2022
Dear Shabnam,
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Samy Anwar
26 July 2022
Dear Shabnam,
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Many thanks for your comments and time. Here I respond to them:
1 - I do agree with you. But I preferred to put the equations in the introduction section because I initially talked about climate change and its impact on PET. Also, I talked in details about the equations to show that PM is not linear equation with many variables, so calculating the PET using the PM equation from the RegCM4 output can induce a large uncertainty associated with the simulated PET. Regarding HG equation, I wanted to show what is the original equation and there is another one based on knowledge of global incident solar radiation and mean air temperature. So I had to show what the second HG formula is to show that PET can be projected using mean air temperature as a proxy.
2 - I do agree with your point of view. Initially I made both of them in one figure, but recently I had to separate them so the reviewer can see the figure clearly.
3 - I do agree with you that there are many sources of PET such as CRU, GLEAM, MODIS and other products. But I preferred to use CRU over GLEAM and MODIS for the following reasons:
a - GLEAM and MODIS compute the PET for every vegetation unit, otherwise it is zero
b- CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
4 - For the reasons reported in point 1, I had to mention equation 4 in the introduction section. Also, I referred to it again section (3.1) as:
"Furthermore, it can be noticed that the PET changes are notably affected by the 2-m mean air temperature changes either in the historical period or in the two future scenarios as reported by Li et al. (2018).
Best Regards
Samy
Samy Anwar
29 July 2022
Dear Participants,
Good afternoon. Hope all are doing well. I would like to talk about simulating or calculating of potential evapotranspiration (PET) in semi-arid, arid (e.g., Egypt) or hyper-arid regions).
It is well known that Penman-Monteith method (PM; Allen et al. 1998) is the standard method to compute the PET either from station data or using the output of the regional climate models (RCMs). However, calculating the PET using PM method in arid regions (e.g., Egypt) faces important challenges such as:
1 - The PM method requires many meteorological variables (each one with its own source of uncertainty), so the calculated PET (either from station data or RCMs) will possesses a high-degree of uncertainty because the PM method is not-linear.
2 - The PM requires specific thresholds and unlimited supply of soil moisture. Such condition doesn’t exist under extreme dry conditions (Brutsaert and Parlange 1998).
3 - Using station observations, the PM cannot be feasible to calculate for a long time for many locations. Also, calculating PET using PM from the output of the RCMs has a serious problem because of it's dependence on the relative humidity or dew-point (to compute the actual vapor pressure); which is quite low over arid/hyper-arid regions (e.g., Egypt) leading to low values of the simulated PET (from the RCMs).
Therefore, there was an urgent need to compute the PET using a simple empirical method with minimum number of meteorological inputs (to reduce the uncertainty of the calculated PET). Another important feature of this simple method is that the proposed methods needs to work under different climate regimes (e.g., dry areas such as Egypt) unlike the other empirical methods.
In this case, the proposed method can be Hargreaves–Samani method (HS; Hargreaves et al. 1985 and 2003). The HS is recommended by the FAO as an alternative method to compute the PET because of its dependence on the temperature and precipitation only (Shahidian et al. 2012; Fisher and Pringle 2013).
This is my justification to use the HG method in my study concerning "Projecting the Potential Evapotranspiration of Egypt using a high-resolution regional climate model (RegCM4)"; which is discussed in many posts in my page or the group recently made "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations".
Please feel free to post comments, so the discussion can be useful for every researcher interested in such field particularly drought assessment or calculating the crop water needs.
Have a nice day.
Best Regards
Samy
Good afternoon. Hope all are doing well. I would like to talk about simulating or calculating of potential evapotranspiration (PET) in semi-arid, arid (e.g., Egypt) or hyper-arid regions).
It is well known that Penman-Monteith method (PM; Allen et al. 1998) is the standard method to compute the PET either from station data or using the output of the regional climate models (RCMs). However, calculating the PET using PM method in arid regions (e.g., Egypt) faces important challenges such as:
1 - The PM method requires many meteorological variables (each one with its own source of uncertainty), so the calculated PET (either from station data or RCMs) will possesses a high-degree of uncertainty because the PM method is not-linear.
2 - The PM requires specific thresholds and unlimited supply of soil moisture. Such condition doesn’t exist under extreme dry conditions (Brutsaert and Parlange 1998).
3 - Using station observations, the PM cannot be feasible to calculate for a long time for many locations. Also, calculating PET using PM from the output of the RCMs has a serious problem because of it's dependence on the relative humidity or dew-point (to compute the actual vapor pressure); which is quite low over arid/hyper-arid regions (e.g., Egypt) leading to low values of the simulated PET (from the RCMs).
Therefore, there was an urgent need to compute the PET using a simple empirical method with minimum number of meteorological inputs (to reduce the uncertainty of the calculated PET). Another important feature of this simple method is that the proposed methods needs to work under different climate regimes (e.g., dry areas such as Egypt) unlike the other empirical methods.
In this case, the proposed method can be Hargreaves–Samani method (HS; Hargreaves et al. 1985 and 2003). The HS is recommended by the FAO as an alternative method to compute the PET because of its dependence on the temperature and precipitation only (Shahidian et al. 2012; Fisher and Pringle 2013).
This is my justification to use the HG method in my study concerning "Projecting the Potential Evapotranspiration of Egypt using a high-resolution regional climate model (RegCM4)"; which is discussed in many posts in my page or the group recently made "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations".
Please feel free to post comments, so the discussion can be useful for every researcher interested in such field particularly drought assessment or calculating the crop water needs.
Have a nice day.
Best Regards
Samy
Samy Anwar
29 July 2022
The page I mentioned is on LinkedIn: https://www.linkedin.com/in/samy-ashraf-anwar-rateb-41382572/
The group "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations" is also available on this page.
Thank you.
Samy
The group "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations" is also available on this page.
Thank you.
Samy
Samy Anwar
29 July 2022
The page I mentioned is on LinkedIn: https://www.linkedin.com/in/samy-ashraf-anwar-rateb-41382572/
The group "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations" is also available on this page.
Thank you.
Samy
The group "Potential Evapotranspiration: In-situ Measurements, Reanalysis and Numerical Simulations" is also available on this page.
Thank you.
Samy
Somayeh Hejabi
29 July 2022
Dear authors,
The subject is so interesting. I have few questions:
1 - Why Hargreaves–Samani method is used instead of Penman-Monteith method in the present study?
2 – Why CRU gridded data is used as a source of observation?
The subject is so interesting. I have few questions:
1 - Why Hargreaves–Samani method is used instead of Penman-Monteith method in the present study?
2 – Why CRU gridded data is used as a source of observation?
Samy Anwar
29 July 2022
Dear Dr Somayah,
Many thanks for your comments. Here I reply to the proposed comments:
1 - The PM shows some limitation to be applied over an arid region (e.g., Egypt) such as:
a - PM method requires many meteorological variables (each one with its own source of uncertainty), so the calculated PET (either from station data or RCMs) will possesses a high-degree of uncertainty because the PM method is not-linear.
b - It requires specific thresholds and unlimited supply of soil moisture. Such condition doesn’t exist under extreme dry conditions (Brutsaert and Parlange 1998).
c - PM cannot be feasible to calculate for a long time for many locations. Also, calculating PET using PM from the output of the RCMs has a serious problem because of it's dependence on the relative humidity or dew-point (to compute the actual vapor pressure); which is quite low over arid/hyper-arid regions (e.g., Egypt) leading to low values of the simulated PET (from the Regional Climate Models; RCMs).
d - Therefore, we used the Hargreaves–Samani method (HS; Hargreaves et al. 1985 and 2003). The HS is recommended by the FAO as an alternative method to compute the PET because of its dependence on the temperature and precipitation only (Shahidian et al. 2012; Fisher and Pringle 2013).
e - The HS method has been widely used in many studies (e.g., Almorox et al. 2015). Another advantage of using the HS approach is that, temperature and solar radiation can explain about 80% of the PET variability (Martí et al. 2015). Furthermore, it was found that using the HS method indicates a good performance in semiarid and arid regions (López-Urrea et al. 2015). In data-scarce regions, the HS can be used to compute the PET with a reasonable accuracy (Traore et al. 2010 and 2013). In addition, the HS method is more suitable for calculating PET under climate change scenario (Li et al. 2018).
2 - CRU products is used over other available products because CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
Many thanks for your comments.
Best Regards
Samy
Many thanks for your comments. Here I reply to the proposed comments:
1 - The PM shows some limitation to be applied over an arid region (e.g., Egypt) such as:
a - PM method requires many meteorological variables (each one with its own source of uncertainty), so the calculated PET (either from station data or RCMs) will possesses a high-degree of uncertainty because the PM method is not-linear.
b - It requires specific thresholds and unlimited supply of soil moisture. Such condition doesn’t exist under extreme dry conditions (Brutsaert and Parlange 1998).
c - PM cannot be feasible to calculate for a long time for many locations. Also, calculating PET using PM from the output of the RCMs has a serious problem because of it's dependence on the relative humidity or dew-point (to compute the actual vapor pressure); which is quite low over arid/hyper-arid regions (e.g., Egypt) leading to low values of the simulated PET (from the Regional Climate Models; RCMs).
d - Therefore, we used the Hargreaves–Samani method (HS; Hargreaves et al. 1985 and 2003). The HS is recommended by the FAO as an alternative method to compute the PET because of its dependence on the temperature and precipitation only (Shahidian et al. 2012; Fisher and Pringle 2013).
e - The HS method has been widely used in many studies (e.g., Almorox et al. 2015). Another advantage of using the HS approach is that, temperature and solar radiation can explain about 80% of the PET variability (Martí et al. 2015). Furthermore, it was found that using the HS method indicates a good performance in semiarid and arid regions (López-Urrea et al. 2015). In data-scarce regions, the HS can be used to compute the PET with a reasonable accuracy (Traore et al. 2010 and 2013). In addition, the HS method is more suitable for calculating PET under climate change scenario (Li et al. 2018).
2 - CRU products is used over other available products because CRU is mainly based on upscaling of station observations around the globe and it is integrated over the period 1901-2020 (version 4.05) and now to 2021 (version 4.06). Also, it uses the PM to compute the PET. In addition, CRU is considered as the best available reference PET data and it is used as the ground truth of observation for global assessment of PET (Droogers and Allen 2002; Mitchell and Jones 2005; IPCC 2007; Sperna et al. 2012; Potop and Boroneant 2014).
Many thanks for your comments.
Best Regards
Samy
Anthony Lupo
29 July 2022
Dear Authors:
Great presentation!! One question - was the RCP 2.6 examined or is there no reason to expect this one to happen? - would it provide for the complete range?
Great presentation!! One question - was the RCP 2.6 examined or is there no reason to expect this one to happen? - would it provide for the complete range?
Samy Anwar
29 July 2022
Dear Prof Anthony,
Many thanks for your kind words. According to the International Center for theoretical Physics (ICTP), the RCP2.6 scenario was provided as demonstrated in the link: http://clima-dods.ictp.it/Data/RegCM_Data/MPI-ESM-MR/
However, only the two future scenarios RCP4.5 and RCP8.5 were considered for the following reasons:
1 - The RCP4.5 represents the moderate scenario (stabilize at end of year 2100) and the RCP8.5 (represents the highest increase at end of year 2100).
2 - Based on available storage and computational power (MPI-ESM-MR) was downscaled first over the MENA region and then nested to Egypt domain), only the scenarios were considered
Other scenarios and GCMs can be considered in a future study as more storage and computational power become available. We also highlighted that we used only one GCM to downscale the RegCM4. Therefore, to account for the uncertainty of the atmospheric forcing, multi-GCMs (CMIP5/CMIP6; (Taylor et al. 2012; Li et al. 2015; Eyring et al. 2016) and their ensemble will be used to further examine the impact of climate change on the Tmean, Rs and PET.
Many thanks for your notes.
I just have some questions to ask:
1 - Based on the proposed comments, will the manuscript be accepted or not?
2 - It was requested from me to submit a manuscript matching the requirements of the Environmental Proceedings Journal (6-8 pages), however I saw that the published manuscript was the original one I sent without figures or tables and different from what was requested from me to submit.
3 - In case the paper is accepted, will there be a restriction on number of pages, i.e. 6-8 pages free and the rest will require fees. I ask because I don't have a source of fund in the mean time.
Kindly let me know your response.
Best Regards
Samy
Many thanks for your kind words. According to the International Center for theoretical Physics (ICTP), the RCP2.6 scenario was provided as demonstrated in the link: http://clima-dods.ictp.it/Data/RegCM_Data/MPI-ESM-MR/
However, only the two future scenarios RCP4.5 and RCP8.5 were considered for the following reasons:
1 - The RCP4.5 represents the moderate scenario (stabilize at end of year 2100) and the RCP8.5 (represents the highest increase at end of year 2100).
2 - Based on available storage and computational power (MPI-ESM-MR) was downscaled first over the MENA region and then nested to Egypt domain), only the scenarios were considered
Other scenarios and GCMs can be considered in a future study as more storage and computational power become available. We also highlighted that we used only one GCM to downscale the RegCM4. Therefore, to account for the uncertainty of the atmospheric forcing, multi-GCMs (CMIP5/CMIP6; (Taylor et al. 2012; Li et al. 2015; Eyring et al. 2016) and their ensemble will be used to further examine the impact of climate change on the Tmean, Rs and PET.
Many thanks for your notes.
I just have some questions to ask:
1 - Based on the proposed comments, will the manuscript be accepted or not?
2 - It was requested from me to submit a manuscript matching the requirements of the Environmental Proceedings Journal (6-8 pages), however I saw that the published manuscript was the original one I sent without figures or tables and different from what was requested from me to submit.
3 - In case the paper is accepted, will there be a restriction on number of pages, i.e. 6-8 pages free and the rest will require fees. I ask because I don't have a source of fund in the mean time.
Kindly let me know your response.
Best Regards
Samy
Ahmed Kheir
31 July 2022
This is an exciting paper, but there are some points that should be considered for further improvements:
1- Abstract should include some quantifying results.
2- The introduction needs major improvements by shifting the equations to Methods, highlighting the hypothesis and novelty, and adding more references related to CC in Egypt. Also, why you used CMIP5 (RCPs) and did not use the recent scenarios , CMIP6 (SSPs)?
3- Add a future direction in the conclusion. I suggest to refer to using compare current scenarios with high resolution scenarios from CMIP6.
1- Abstract should include some quantifying results.
2- The introduction needs major improvements by shifting the equations to Methods, highlighting the hypothesis and novelty, and adding more references related to CC in Egypt. Also, why you used CMIP5 (RCPs) and did not use the recent scenarios , CMIP6 (SSPs)?
3- Add a future direction in the conclusion. I suggest to refer to using compare current scenarios with high resolution scenarios from CMIP6.
Samy Anwar
31 July 2022
Dear Dr Ahmed,
Many thanks for your time and constructive comments. Here, I reply to the suggested comments point-by-point:
1 - I agree with you. I will add a part of quantifying the RegCM4 performance at Aswan (as an example) when the manuscript is returned to me for a revision.
2 - Since I talk about climate change and its influence on water resources including potential evapotranspiration (as an important component in the terrestrial hydrology cycle), it was better to talk about it in the introductory section. Also in the methodology, I focus on the experiment design by giving a description of the RegCM4, downscaling by MPI-ESM-MR and observational dataset I used in my study (Climate Research Unit; CRU). In addition, in the results section
I referred to equation 4, as I discussed the impact of climate change on Tmean first, then on PET (as a proxy of Tmean).
In the introduction section, I referred to previous climate change study and its impact on the PET of Egypt as written as:
Khalil et al. (2015) used remote sensing products estimate the water loss from agricultural lands using the HS method.
Abdel Wahab et al. (2018) used the regional climate model (RegCM4; Giorgi et al. 2012) model to estimate the water loss in the period 2015-2025 under RCP4.5 scenario.
Also, in the discussion and conclusion section, I linked the results of my study to previous studies conducted over Egypt as: "The results showed that the projected PET shows a gradual increase from the North zone to Upper Egypt zone; such finding is consistent with the results reported in (Attaher et al. 2006; Farag et al. 2016).
At the time of conducting experiments, the RegCM4 was supported by an interface with the GCMs participated in the CMIP5 simulations. As for the CMIP6, such an interface with the RegCM4 still undergoes frequent developments. Therefore, we used GCMs belongs to the CMIP5. As long as the interface of CMIP6 with the RegCM4 is ready, we will include downscaling CMIP6 (by the RegCM4) in a future study.
3 - At the end of section 4 (discussion and conclusion), we added the following paragraph to conduct a future study as:
"A future work will consider using multi-GCMs (CMIP5/CMIP6; (Taylor et al. 2012; Li et al. 2015; Eyring et al. 2016) and their ensemble to further examine the impact of climate change on the Tmean, Rs and PET. "
Therefore, instead of comparing the results of this study with the output of the CMIP6, we will downscaled the GCMs participated in both CMIP5/6, comparing between them concerning Tmean, Rs and PET with the CRU (as the observational dataset) and for different future projections (RCPs and SSPs).
Again for your time and constructive comments.
Best Regards
Samy
Many thanks for your time and constructive comments. Here, I reply to the suggested comments point-by-point:
1 - I agree with you. I will add a part of quantifying the RegCM4 performance at Aswan (as an example) when the manuscript is returned to me for a revision.
2 - Since I talk about climate change and its influence on water resources including potential evapotranspiration (as an important component in the terrestrial hydrology cycle), it was better to talk about it in the introductory section. Also in the methodology, I focus on the experiment design by giving a description of the RegCM4, downscaling by MPI-ESM-MR and observational dataset I used in my study (Climate Research Unit; CRU). In addition, in the results section
I referred to equation 4, as I discussed the impact of climate change on Tmean first, then on PET (as a proxy of Tmean).
In the introduction section, I referred to previous climate change study and its impact on the PET of Egypt as written as:
Khalil et al. (2015) used remote sensing products estimate the water loss from agricultural lands using the HS method.
Abdel Wahab et al. (2018) used the regional climate model (RegCM4; Giorgi et al. 2012) model to estimate the water loss in the period 2015-2025 under RCP4.5 scenario.
Also, in the discussion and conclusion section, I linked the results of my study to previous studies conducted over Egypt as: "The results showed that the projected PET shows a gradual increase from the North zone to Upper Egypt zone; such finding is consistent with the results reported in (Attaher et al. 2006; Farag et al. 2016).
At the time of conducting experiments, the RegCM4 was supported by an interface with the GCMs participated in the CMIP5 simulations. As for the CMIP6, such an interface with the RegCM4 still undergoes frequent developments. Therefore, we used GCMs belongs to the CMIP5. As long as the interface of CMIP6 with the RegCM4 is ready, we will include downscaling CMIP6 (by the RegCM4) in a future study.
3 - At the end of section 4 (discussion and conclusion), we added the following paragraph to conduct a future study as:
"A future work will consider using multi-GCMs (CMIP5/CMIP6; (Taylor et al. 2012; Li et al. 2015; Eyring et al. 2016) and their ensemble to further examine the impact of climate change on the Tmean, Rs and PET. "
Therefore, instead of comparing the results of this study with the output of the CMIP6, we will downscaled the GCMs participated in both CMIP5/6, comparing between them concerning Tmean, Rs and PET with the CRU (as the observational dataset) and for different future projections (RCPs and SSPs).
Again for your time and constructive comments.
Best Regards
Samy
