Advances in UAV Operations for Valley-Type Mapping with Different Duration Period PPP-AR Methods in GCP
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsIt is a very interesting study. The primary challenge addressed in this study is achieving high-accuracy positioning in rugged and narrow terrains such as valleys. In such environments, GNSS signals frequently experience interruptions or degradation due to topographical constraints, negatively impacting the performance of conventional methods. the unique challenges of valley-type terrains, such as abrupt slope transitions, narrow GNSS signal visibility windows, and multipath effects intensified by concave landforms, remain underexplored.
- What is the theoretical contribution?
- How to optimize flight altitude, image orientation, and overlap ratios?
- Please add the technical parameters of Javad Triumph-1 and Topcon Hiper Pro.
- Which stations are as the reference station? What is the baseline distance?
- Why do you choose both 3- and 10-minute observation intervals to evaluate?
- the differences between the terrain coordinates derived from CORS-TR at the checkpoints and those obtained from the various photogrammetric models were calculated. How about the coordinate accuracy?
- Is the PPP post solution?
- What kind of precision sp3 and clock file are used? GFZ or others? Final sp3 file or rapid sp3 file?
- CSRS-PPP 3 min is followed by PPP-Arisen 3 min with errors ranging from 6.7 to 30.9 cm and raPPPid 3 min with errors ranging from 16.1 to 75.9 cm. Figure 5 shows that the raPPPid-based model mostly gave errors above 30 cm, while other software solutions resulted in errors of 20 cm or below. The findings suggest that using CSRS-PPP or PPP-Arisen software would be more suitable for a 3-minute observation duration. Can the precision(~dm) meet the practical applications? If it is used for UAV mapping, which scale is available for practical use? 1/5000 or finer?
- Regarding vertical errors, CSRS-PPP is the best software for the PPP-AR method with a 10-minute observation duration. Could the authors discuss the reasons?
- As can be seen from figure 8, the precision with a 120-minute static duration is 10~20 cm. In which applications can this be used?
- As can be seen from figure 10 and figure 11, There are some differences in precision. Could the authors discuss the reasons for the differences among various PPP software?
- As can be seen from table 2, the vertical accuracy is ~dm. which applications can this be used?
Author Response
It is a very interesting study. The primary challenge addressed in this study is achieving high-accuracy positioning in rugged and narrow terrains such as valleys. In such environments, GNSS signals frequently experience interruptions or degradation due to topographical constraints, negatively impacting the performance of conventional methods. the unique challenges of valley-type terrains, such as abrupt slope transitions, narrow GNSS signal visibility windows, and multipath effects intensified by concave landforms, remain underexplored.
Response: Thank you for your useful comments and suggestions.
1) What is the theoretical contribution?
Response: In the introduction, the difference between the study and the existing literature and theoretical contribution are clearly stated. Please look at the 5th and last paragraphs in the introduction.
2) How to optimize flight altitude, image orientation, and overlap ratios?
Response: A new paragraph was added to the “2.2. Photogrammetric Process” section discussing UAV flight planning parameters. Specifically, we explained that flight altitude was chosen considering terrain relief, while image orientation and overlap ratios (80% forward, 70% side) were optimized to balance image redundancy and GNSS signal limitations in narrow valleys. You can look at lines 150 – 160.
3) Please add the technical parameters of Javad Triumph-1 and Topcon Hiper Pro.
Response: We added the technical parameters of Javad Triumph-1 and Topcon Hiper Pro. You can look at Page 5, Lines 182-183. Thank you for pointing out these shortcomings.
4) Which stations are as the reference station? What is the baseline distance?
Response: CORS-TR stations in the surrounding area are as reference points for determining the GCP coordinates. Reference CORS-TR stations were illustrated in Figure 1. The baseline distances to reference points vary between 1.2 km and 263 km. This information is given on page 6, between lines 193 and 198.
5) Why do you choose both 3- and 10-minute observation intervals to evaluate?
Response: We explained clearly the reason for choosing 10-minute observation interval to evaluate in lines 319-320. The reason for choosing 3-minute observation interval to evaluate was explained in between lines 340-344. Thank you for pointing out these shortcomings.
6) The differences between the terrain coordinates derived from CORS-TR at the checkpoints and those obtained from the various photogrammetric models were calculated. How about the coordinate accuracy?
Response: The terrain coordinates derived from CORS-TR and their accuracies were given in Figure 2. The horizontal accuracies vary between 2.3 - 5.1 mm. The vertical accuracies vary between 4.3 - 7.2 mm. You can look at Figure 2.
7) Is the PPP post solution?
Response: Yes. We stated this issue in lines 256-257.
8) What kind of precision sp3 and clock file are used? GFZ or others? Final sp3 file or rapid sp3 file?.
Response: We have added this information. You can look at page 8 and 9, Lines 284-288. Thank you for pointing out this shortcoming.
9) CSRS-PPP 3 min is followed by PPP-Arisen 3 min with errors ranging from 6.7 to 30.9 cm and raPPPid 3 min with errors ranging from 16.1 to 75.9 cm. Figure 5 shows that the raPPPid-based model mostly gave errors above 30 cm, while other software solutions resulted in errors of 20 cm or below. The findings suggest that using CSRS-PPP or PPP-Arisen software would be more suitable for a 3-minute observation duration. Can the precision(~dm) meet the practical applications? If it is used for UAV mapping, which scale is available for practical use? 1/5000 or finer?
Response: Precision does not meet orthophoto and topographic maps in 1/5000 or finer scale. It is adequate for topographic mapping at medium scales (1/5000–1/10000). Also, it will be sufficient for applications such as environmental planning. The practical applications that the obtained precision will meet are stated in the conclusion section. Please look at lines 571-577.
10) Regarding vertical errors, CSRS-PPP is the best software for the PPP-AR method with a 10-minute observation duration. Could the authors discuss the reasons?
Response: CSRS-PPP benefits from precise satellite orbit and clock corrections provided by IGS, CODE and includes robust tropospheric delay modeling, which likely contributed to its superior vertical accuracy. The reason was given in conclusions section. You can look at lines 515-518.
11) As can be seen from figure 8, the precision with a 120-minute static duration is 10~20 cm. In which applications can this be used?
Response: We added a paragraph explaining that such accuracy is sufficient for applications such as topographic mapping, landslide monitoring, and other environmental monitoring studies where decimeter accuracy is acceptable. Please look at lines 392 - 401.
12) As can be seen from figure 10 and figure 11, There are some differences in precision. Could the authors discuss the reasons for the differences among various PPP software?
Response: We included an explanation highlighting that variations arise from different ambiguity resolution strategies, clock/troposphere modeling, and quality of precise orbit/clock products. You can look at lines 452-463.
13) As can be seen from table 2, the vertical accuracy is ~dm. which applications can this be used?
Response: We added that vertical accuracy at decimeter level is suitable for DEM generation at medium scales (1/5000–1/10000), hydrological modeling, and geomorphological studies, but not sufficient for engineering surveying requiring centimeter-level precision. Please look at lines 492-501.
Reviewer 2 Report
Comments and Suggestions for AuthorsPLEASE, use consistent geopositioning!
Fig1 shows the study area and location at three scales. The borders provide only a degree-minute second approximation for the area used in the study. Using OpenStreetMap and Google Earth. I locate the approximate centre of the ‘rectangle’ as
[38.0117,32.4979]. This decimal/Lat/Long, [dLL], value is a tuple of the csv values of lat and long to 4 decimal places. For this location 4 dp is quite sufficient but 3 dp isn’t. You can paste this value (even with the [ ]) into Google Earth. You need to make it easier for readers (and reviewers!) as well as people using the paper to make the data easier to get at (ie FAIR data principles).
Presumably the GCP locations are known to 6 or 7 dp location values. Why suddenly go to eastings northings in Fig 2?
From GE I estimate by eye that GCP3[dLL] is [38.01197,32.49619] to 5 dp.
If I am using a UAV to survey an area of look at a location I want to know a digital location, [dLL] provides this uniquely. The results of your work should be to make life easier for they end user.
The need for good geolocation is shown in Fig.2. I can see the big tree in GEarth but I want to know where it is so I (and anybody else) needs to know where it is. You will need to know where to send your UAV for repeats under different vegetation/me conditions (for example) or perhaps to compare the effectiveness of new sensors etc. All these point to the need for data reproducibility, specification of images and data results. So please says something about this as Future work.
Please
- redraw the maps with decimal degrees on the margin (i.e. do away with d,m,s).
- Locate image/view centres with 4dp [dLL]
- Give your GCPs [dLL] values
- Show how you can use a circle/ellipse error/precision to, say 5dp as appropriate, for better ground control and mission planning.
Author Response
PLEASE, use consistent geopositioning!
Fig1 shows the study area and location at three scales. The borders provide only a degree-minute second approximation for the area used in the study. Using OpenStreetMap and Google Earth. I locate the approximate centre of the ‘rectangle’ as [38.0117,32.4979]. This decimal/Lat/Long, [dLL], value is a tuple of the csv values of lat and long to 4 decimal places. For this location 4 dp is quite sufficient but 3 dp isn’t. You can paste this value (even with the [ ]) into Google Earth. You need to make it easier for readers (and reviewers!) as well as people using the paper to make the data easier to get at (ie FAIR data principles).
Presumably the GCP locations are known to 6 or 7 dp location values. Why suddenly go to eastings northings in Fig 2?
From GE I estimate by eye that GCP3[dLL] is [38.01197,32.49619] to 5 dp.
If I am using a UAV to survey an area of look at a location I want to know a digital location, [dLL] provides this uniquely. The results of your work should be to make life easier for they end user.
The need for good geolocation is shown in Fig.2. I can see the big tree in GEarth but I want to know where it is so I (and anybody else) needs to know where it is. You will need to know where to send your UAV for repeats under different vegetation/me conditions (for example) or perhaps to compare the effectiveness of new sensors etc. All these point to the need for data reproducibility, specification of images and data results. So please says something about this as Future work.
Response: We sincerely thank the reviewer for the detailed and insightful comments regarding geolocation consistency and fair data principles. We can share all data upon request, and we stated that in the data availability statement section. We added new sentences about future work to the lines 585 – 590. We agree that the use of decimal degree coordinates is more effective for data accessibility and reproducibility. Following the reviewer’s suggestions, we have implemented the following changes in the revised manuscript:
Please
- redraw the maps with decimal degrees on the margin (i.e. do away with d,m,s).
Response: We have redrawn the maps (Figures 1 and 2) with decimal degrees on the margins instead of degrees-minutes-seconds.
- Locate image/view centres with 4dp [dLL]
Response: The center coordinates of the study area is now presented in decimal latitude/longitude values (dLL) with 4 decimal places, which ensures adequate spatial resolution for the study area. Please look at lines 129 – 131.
- Give your GCPs [dLL] values
Response: We provided the GCP coordinates in decimal degree format (to 5 decimal places for maximum precision). You can look at new Figure 2.
- Show how you can use a circle/ellipse error/precision to, say 5dp as appropriate, for better ground control and mission planning.
Response: We illustrated the errors of GCPs using circles in Figure 2. Please refer to new Figure 2
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have revised the manuscript. Some interesting research results are obtained. The comment is as follows,
The title in Figure 5 is incorrect. 10-minute should be revised to 3-minute.Horizontal errors for static method and 3-minute PPP-AR.
Author Response
Comment 1: The title in Figure 5 is incorrect. 10-minute should be revised to 3-minute.Horizontal errors for static method and 3-minute PPP-AR.
Response 1: Thank you for your attention and helpful contribution.
