Research article

Multi-Spatiotemporal Analysis of Changes in Mangrove Forests in Palawan, Philippines: Predicting Future Trends Using Support Vector Machine Algorithm and Markov Chain Model

Authors
  • Cristobal B. Cayetano orcid logo (College of Fisheries and Aquatic Sciences, Western Philippines University, Sta. Monica, Puerto Princesa City, Palawan, Philippines)
  • Lota A. Creencia orcid logo (College of Fisheries and Aquatic Sciences, Western Philippines University, Sta. Monica, Puerto Princesa City, Palawan, Philippines)
  • Emma Sullivan orcid logo (Remote Sensing Group, Plymouth Marine Laboratory, Prospect Place, Plymouth PL4 7QP, UK)
  • Daniel Clewley orcid logo (Remote Sensing Group, Plymouth Marine Laboratory, Prospect Place, Plymouth PL4 7QP, UK)
  • Peter I. Miller orcid logo (Remote Sensing Group, Plymouth Marine Laboratory, Prospect Place, Plymouth PL4 7QP, UK)

This is version 2 of this article, the published version can be found at: https://doi.org/10.14324/111.444/ucloe.000057

Abstract

Multi-temporal remote sensing imagery can be used to explore how mangrove assemblages are changing over time and facilitate critical interventions for ecological sustainability and effective management. This study aims to explore the spatial dynamics of mangrove extents in Palawan, Philippines, specifically in Puerto Princesa City, Taytay and Aborlan, and facilitate future predictions for Palawan using the Markov Chain model. The multi-date Landsat imageries during the period 1988–2020 were used for this research. The support vector machine algorithm was sufficiently effective for mangrove feature extraction to generate satisfactory accuracy results (>70% kappa coefficient values; 91% average overall accuracies). In Palawan, a 5.2% (2693 ha) decrease was recorded during 1988–1998 and an 8.6% increase in 2013–2020 to 4371 ha. In Puerto Princesa City, a 95.9% (2758 ha) increase was observed during 1988–1998 and 2.0% (136 ha) decrease during 2013–2020. The mangroves in Taytay and Aborlan both gained an additional 2138 ha (55.3%) and 228 ha (16.8%) during 1988–1998 but also decreased from 2013 to 2020 by 3.4% (247 ha) and 0.2% (3 ha), respectively. However, projected results suggest that the mangrove areas in Palawan will likely increase in 2030 (to 64,946 ha) and 2050 (to 66,972 ha). This study demonstrated the capability of the Markov chain model in the context of ecological sustainability involving policy intervention. However, as this research did not capture the environmental factors that may have influenced the changes in mangrove patterns, it is suggested adding cellular automata in future Markovian mangrove modelling.

Keywords: change detection, image classification, Landsat, land use/land cover, Markov chain model, spatial dynamics, support vector machine

Rights: © 2023 The Authors.

1568 Views

1Citations

Published on
28 Apr 2023
Peer Reviewed

 Open peer review from Xiongjie Deng

Review

Review information

DOI:: 10.14293/S2199-1006.1.SOR-EARTH.A3RGAV.v1.RLSJNW
License:
This work has been published open access under Creative Commons Attribution License CC BY 4.0 , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Conditions, terms of use and publishing policy can be found at www.scienceopen.com .

ScienceOpen disciplines: Remote sensing , Earth & Environmental sciences , Quantitative & Systems biology , General earth science , Environmental change , Environmental studies
Keywords: Environmental science , Change detection , Image classification , Landsat , Land use/land cover , Markov Chain Model , Spatial dynamics , Support Vector Machine , Environmental modelling , Environmental protection

Review text

Review of

Multi-Spatiotemporal Analysis of Changes in Mangrove Forests in Palawan, Philippines: Predicting Future Trends Using Support Vector Machine Algorithm and Markov Chain Model

by Xiongjie Deng

This manuscript mapped the extent of mangrove forests in Palawan over multiple years using the Support Vector Machine (SVM) method, conducted change detection analysis across three periods, and predicted the trends of mangrove forests based on the Markov Chain model in the future. The generated accuracies demonstrated satisfactory classification results, and the Markov Chain model was evaluated to be a helpful tool for projecting mangrove forest areas in Palawan. Furthermore, I enjoy the analyses that involve policies and management interventions.

Still, I notice that several same descriptions are in different format. For example, “modeling” in line 420 but “modelling” in lines 87, 380, Figure 2, etc.; “ e.g. ” are in italic type in line 128 and 146, etc., but are in roman type in line 151 and 153, etc. I would be really delighted if the authors could keep the style consistent across the whole article. Besides, for the entire article, I think it would be better to add a comma behind “et al.” and the author’s name, for example, (Brown et al., 2006; Mukherjee et al., 2014) (line 94), (Ball and Pidsley, 1995) (line 96).

What’s more, I find some errors in this manuscript, and please find my suggestions and doubts below:

Line 85: I think “may had” should be “may have”.

Line 100: I think the comma behind “intrusion” is redundant.

Line 124: The phrase “to do” may be wordy, consider removing it.

Line 149: Perhaps there should be a comma between “technique” and “training”; consider modifying “time consuming” to “time-consuming”.

Line 148 - 150: It confused me that you specified “Landsat imagery”. I agree that extracting training samples is time-consuming when using supervised classification methods, but I do not think it is only for Landsat imagery, so I would advise you to modify “Landsat imagery” to a more general term, like remotely sensed imagery.

Line 154: I think you should add “of” between “one” and “many”.

Line 167: Perhaps you should modify “(c)” to “(3)” since you used “(1)” and “(2)” before.

Line 178: I think “borders” should be “border”.

Line 193: It seems the “ecosystem” should be in plural format.

Line 226, 243, 249, 259, 272, 391: It looks like you typed the letter “x” instead of the multiplication sign “×”.

Line 254: The “ L min – (L max L min ) ” confused me, I guess you want to explain Q min and Q max here?

Line 319: Perhaps “was” should be “were”.

Line 344: The comma behind “where” may be unnecessary.

Line 358, 360: Please keep both “Kappa” consistent.

Line 361: It seems that “qualify” does not agree with the subject.

Line 420 – 426: I would advise you to move this part to the beginning of Section 2.

Figure 2: In the left box, I would advise you to modify “7 ETM+” to “Landsat 7 ETM+”; you mentioned LULC classification using SVM in this figure, but in the title, you missed this part, so I would advise you to describe it more specifically.

Line 438: Perhaps “were” should be “was”.

Line 445: Perhaps “are” should be “is”.

Figure 4: As I understand, the left y-axis is for Palawan, and the right y-axis is for the other three cities. If so, I would advise you to specify it in the title of this figure, or consider mapping them separately. The original figure is confusing to me.

Figure 5: I would advise you to add ticks to the y-axis.

Figure 6: Please add a legend to indicate bars in different colours.

Line 493: I think it would be more precise to say “SPOT satellite sensor’s data” or “SPOT satellite sensor’s images”.

Line 494: Perhaps “Based on”.

Line 521: “in” may be unnecessary.

Table 2: Is it normal that the estimates in 2018 and 2020 from this study are exactly the same in Puerto Princesa City and Taytay?

Line 537, 538: The first sentence is ambitious to me, I guess you tried to mean that the total mangrove forests extent in Puerto Princesa City at 3,201.8 ha in 2003 was estimated by Pagkalinawan and Ramos (2013)? If so, I would advise you to modify the place of the term “in 2003” because if it presents at the beginning of the sentence, it may mislead readers that Pagkalinawan and Ramos conducted their research in 2003.

Line 562 - 564: I only found “WRSP/R” information in Table 2; I would advise you to add cloud cover percentages of the two images you mentioned to make this sentence more compelling.

Figure 8 (a) and (b), (c) and (d): It will be better if the x-axis of (b) and (d) can align with the x-axis of (a) and (c), respectively.

Line 757: I think “had” is not in the correct form.

Line 763: It seems that “presumed” does not appear to be in the proper form.

Line 764: I think the “is” after “there” is not in the correct form.

Line 774: It seems the “and” before “infrastructure” is unnecessary.

Line 788: The word “shifting” may be in the wrong form.

Line 805: I think it would be more precise to add “an” before “increase”.

Line 812: It seems that “continues” should be “continue”.

Line 819 - 823: You state that integrating Cellular Automata in Markov Chain modelling can evaluate the impacts of different policies. Still, I suggest you to explain why you recommend future research should integrate the Cellular Automata in Markov Chain modelling more specifically, for example, you can provide more details about the advantages, capabilities, and effectiveness of Cellular Automata in mangroves-related research to assess policies.

Line 830: Two “also” in a sentence seem unnecessary.



Note:
This review refers to round 1 of peer review.

 Open peer review from Banashree Thapa

Review

Review information

DOI:: 10.14293/S2199-1006.1.SOR-EARTH.AATWRS.v1.RJURJZ
License:
This work has been published open access under Creative Commons Attribution License CC BY 4.0 , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Conditions, terms of use and publishing policy can be found at www.scienceopen.com .

ScienceOpen disciplines: Remote sensing , Earth & Environmental sciences , Quantitative & Systems biology , General earth science , Environmental change , Environmental studies
Keywords: Environmental science , Change detection , Image classification , Landsat , Land use/land cover , Markov Chain Model , Spatial dynamics , Support Vector Machine , Environmental modelling , Environmental protection

Review text

The paper mapped the spatio-temporal trends of mangrove forests in the city of Palawan, Phillipnes using Markov chain model highlighting the response of mangrove forests to policy interventions, thus speaking to the role of protective mechanisms in impacting health and abundance of mangrove forests.

Overall, the manuscript is a great attempt by Cayetano et al., Following are a few suggestions for the authors' consideration.

Line 10: I think the authors meant "titled" not "entitled"

Line 16: Mangrove conservation isnt the right term I feel. One can not conserve mangroves using remote sensing, rather study mangroves and thus promote conservation

Line 124: This sentence can be improved, reflecting the exact reasons why repeated, accurate on-ground sampling exercises cant be done. For instance, talking about the inaccessibility of mangrove forests, distance from monitoring locations etc, cost associated with the field visits and high technical skills required to navigate these ecosystems make in-situ sampling difficult for mangrove forests.

Line 125: If the above suggestions are accepted, then consider changing "however" to "thus"

Line 136-141: Consider rewording "The mangroves of Palawan have been protected under the direct human inventions through the International Union for Conservation of Nature (IUCN) protected area Category I-IV (Long and Giri 2011) and 1992 Republic Act No.7611, commonly known as the Strategic Environmental Plan for Palawan Act (SEP Law, FAO 2021); yet this unique ecosystem remains under threat due to climate change and associated rising sea levels (Gilman et al. 141 2008; Giri et al. 2011)."

Line 143: replace "for" with "in"

Line 161-163: sentence hard to understand; may consider rewording

Line 196-197: Repeated info as Introduction Para 2 Line 4

Line 285: If the abbreviation has not been elaborated before, then recommend you to write Electromagnetic spectrum here

Line 371: a brilliant explaination of the Markov tool, nice!

Line 378: replace "have been" with "is"

Line 425: consider adding "diagram of multi-temporal mangrove change detection"

Line 481: Figure description: please do mention the time frame as well

Line 674: Please elaborate the ECAN abbreviation

Line 728: replace "caused" to "cause"

Line 757: replace "may had" with "may be"

Line 788: replace "becomes slightly shifting" to "starts to shift"

Line 814: Maybe you meant "vitally important" instead of "viably important"



Note:
This review refers to round 1 of peer review.