Peat properties of a tropical forest reserve adjacent to a fire-break canal

Dayang Nur Sakinah Musa · Mohd Zahirasri Mohd Tohir · Xinyan Huang ·Luqman Chuah Abdullah · Mohamad Syazaruddin Md Said ·Muhammad Firdaus Sulaiman

Abstract Tropical peat comprises decomposed dead plant material and acts like a sponge to absorb water,making it fully saturated.However,drought periods dry it readily and increases its vulnerability to fire.Peat fires emit greenhouse gases and particles contributing to haze,and prevention by constructing fire-break canals to reduce fire spread into forest reserves is crucial.This paper aims to determine peat physical and chemical properties near a fire-break canal at different fire frequency areas.Peat sampling was conducted at two forest reserves in Malaysia which represent low fire frequency and high fire frequency areas.The results show that peat properties were not affected by the construction of a fire-break canal,however lignin and cellulose content increased significantly from the distance of the canal in both areas.The study concluded that fire frequency did not significantly influence peat properties except for porosity.The higher fibre content in the high frequency area did not influence moisture content nor the ability to regain moisture.Thus,fire frequency might contribute differently to changes in physical and chemical properties,hence management efforts to construct fire-break canals and restoration efforts should protect peatlands from further degradation.These findings will benefit future management and planning for forest reserves.

Keywords Peat fire · Peat properties · Fire-break canals ·Forest reserves

Introduction

Tropical peat swamps are prone to poor drainage,resulting in swampy,waterlogged conditions on the surface of peat soils.Water-logging develops plant residues which usually decay rapidly (Taufik et al.2019).Sabah’s peat swamp is unique and distinct from other peat swamps in Borneo and thus there is high potential for research and conservation at an international scale (Basintal et al.2007).Activities such as agriculture and land development threaten this unique ecosystem.Anthropogenic factors cause fire to spread from state and isolated lands to the forest reserve (FR).Based on the Annual Report of the Sabah Forestry Department(SFD 2020),open burning and arson contributed to cause of major fires,for example,land clearing (SFD 2019,2020).In the reports,the area burned annually was 706 ha (2015),20,000 ha (2016),60 ha (2017),279 ha (2018),1313 ha(2019),and 556 ha (2020).Fire in the peat swamp forest is associated with fire that smoulder underground and are persistent and lengthy(Rein 2013;Rein and Huang 2021),verifying the incidence of peat fires in Sabah that lasted about six months because of the El-Nino effect and about two months without El-Nino influence (SFD 2017,2020).Moreover,the peat fires damaged the peat swamp ecosystem,resulting in a slow recovery of the forest (Davies et al.2013).Peat moisture content is lowered as an intense water deficit is caused by the water table drawdown and massive burning(Afdeni et al.2016;Granath et al.2016;Taufik and Nukman 2019) which mainly occur during droughts.

In an earlier study,peat properties for distance from canal by Sinclair et al.(2020) focused on bulk density further from the canal.Previous research on canal blocks focused on smallholder farmers,and agricultural lands held by large companies to maintain water levels for yields(Ward et al.2021).Peat bulk density in post-fire areas was correlated to moisture content which controls ignition and burn depth (Nelson et al.2021).Fires will change peat properties and are governed by slow spread rate and low temperatures of smouldering fires (Rein 2009;Huang et al.2015;Santoso et al.2021).Post-fire effects can be determined by the physical and chemical properties of post-fire peat samples,as nutrient contents have changed(Könönen et al.2015;Handayani and Maswar 2019).For example,fire will consume organic matter and reduce its carbon content.

To protect peatlands in Malaysia,the government implemented prevention measures such as creating fire-break canals in Selangor (Munawi 2014) and Sabah peat swamp forests (Fig.1).Research on the presence of canals in peatlands are limited to the fire-break function.Fire-break canal implementation in Malaysia is like the canal blocking program in Indonesia,wherein their function is to raise the water table and prevent further drainage leading to peat fires(Sutikno et al.2019;Ward et al.2021).As degraded peatland easily triggers fire outbreaks,rewetting is important to reduce the number of fire events (Taufik et al.2023).In this study,the aims may be achieved through fire-break canal implementation in the study areas.In Sabah,a fire-break canal with a 6 m2cross-section area was constructed in Klias Forest Reserve,successfully reducing fire spread and was then implemented in Binsuluk Forest Reserve (Fig.1).Research on peat properties further from a fire-break canal and the effect of fire frequencies has yet to be understood.Therefore,this study aims to understand the influence of fire on peat conditions in different fire frequency areas near a fire-break canal.The post-fire effects near the fire-break canal caused by the peat fire will provide information on the impact of fire at different frequencies on peat properties.It is hypothesised that: (1) physical and chemical properties at different fire frequency areas are not the same;(2) different fire frequencies result in different moisture and fibre relationships;and,(3) fire-break canals do not influence the physical and chemical properties of peat.

Materials and methods

Study area

Two forest reserves were selected to represent high fire frequency (HFF) and low fire frequency (LFF) areas.LFF is an area with minimum fire incident and the forest cover is still intact including asDryobalanopsspp.andAcaciaspp.(Fig.2a).The HFF is a severely frequently burnt area,waterlogged and with less forest cover (Fig.2b).The areas representing LFF and HFF are Klias FR and Binsuluk FR,respectively (Fig.3).Additional information regarding the fire history spanning from 2013 to 2020 indicates a higher fire frequency in Binsuluk FR,as depicted in Figures S1–S8,contrasting with a lower fire frequency in Klias FR (Appendix A).

Fig.2 (a) Intact forest in Klias FR (LFF);(b) Waterlogged condition on the surface in Binsuluk FR (HFF).(Photos by Dayang Nur Sakinah Musa)

Fig.3 Location of the Klias (LFF) and Binsuluk (HFF) forest reserves in East Malaysia

Sampling method

A 300-m transect line was established near a fire-break canal in each area for sampling at 75 m,150 m,225 m,and 300 m(Fig.4a),which are the study plots.Each plot has three randomly selected sub-plots within a 25-m radius as sampling replications.In addition,a 100-cm square pit 150-cm deep was dug at each sub-plot for sample collections at zero cm,50 cm,100 cm,and 150 cm (Fig.4b),in May 2021 for peat physical(n=48) and chemical properties (n=48).The peat auger and a core ring (98.17cm3) were used for chemical and physical analysis,respectively.Thus,the total samples collected is 96 (4 plots × 4 depths × 3 replications) for each HFF and LFF areas.

Fig.4 (a) Plot constructed near the fire-break canal at both HFF and LFF sites for peat sample collections;(b) Peat collected at different levels

Peat sample analysis

Physical properties

Peat samples were weight (Mw) before oven-drying for 24 h at 105 ºC.The oven-dried samples were weighted (Ms),and bulk density (g cm-3),moisture content (%),moisture regain (%),and porosity (%) calculated using the following Eqs:

Chemical properties

This is divided into (1) the acid detergent fibre method to analyse the decomposed peat fibre content (n=32);and(2) the loss on ignition (LOI) method (n=48) to determine ash content (%),organic matter content (%),and organic carbon content (Corg) (Agus et al.2011).Calculations for the LOI method are:

(*0.58 is the conversion factors from OM to Corg)

Statistical analysis

Statistical analysis for bulk density,moisture content,porosity,ash content,organic carbon content,organic matter content (n=48) lignin,cellulose,moisture regain,and moisture contentb(n=32) were done using SPSS software(IBM SPSS V.28).Comparisons between plots and sample depths were made using two-way analysis of variance(ANOVA) at a 95% confidence level are illustrated using SPSS (IBM SPSS V.28).AT-test was carried out between the two groups to compare the means of HFF and LFF.Pearson-correlation analysis was done to determine the relationship between the properties.Microsoft PowerPoint software was used to compile and organised the results.

Results and discussion

Peat properties of the high-frequency fire area (HFF)

In the HFF area,a fire-break canal was built [because the implementation of the fire break canal] in LFF demonstrated its effectiveness in mitigating the spread of fires into theto reduce fire spread into the LFF area.Compaction of the peat properties between the layers and from the distance of the fire-break canal was determined.For the HFF physical properties,bulk density (g cm-3) ranged from 0.14–0.22 g cm-3and was statistically lower than in LFF.Moisture content was statistically higher than in LFF,ranging from 354 to 1284% (Table 1).These two properties are related to each other,wherein the porosity of the peat shows that HFF peat have a statistically higher porosity than LFF.Peat swamp density,which is very low,will increase its water holding capacity,essential to regulate water,especially in the tropics (Leng et al.2019).Environmental conditions (Fig.2b) also contributed to the results of the peat samples.The low forest cover exposed the HFF area to direct sunlight,rain,and wind and contributed to surface saturation during the rainy seasons and its drying out during droughts.Therefore,the HFF area was fully saturated or waterlogged during peat sampling.Moisture content of the peat at 75 m and 150 m was statistically higher than at 225 m and 300 m,and ash content also decreased statistically further from the firebreak canal,indicating that fires usually started in the line of the fire-break canal.[The fire spread possibly due to the spotting fire that allows the fire spread to the forest reserve areas.]

Table 1 Peat properties in high fire frequency (HFF) areas(n=48) based on distance from the fire-break canal

The fire-break canal-maintained moisture around 0.4 m or higher relative to the peat surface (Ward et al.2021),which influenced the findings of higher than 1000% moisture levels on the 75 m and 150 m plots.Thus,these results have added information that higher surface moisture might be influenced by the fire-break canal that maintains the water table.Further,the waterlogged condition of the peat surface might also be due to its water holding capacity.Higher porosity for peat layers in the HFF area have the ability to absorb water.Higher ash content has water-repellence characteristics due to slow burning (Dlapa et al.2013).A lack of oxygen occurs in waterlogged conditions,blocking near-surface macropores that affect peat hydrology,known as the hydrographic condition (Duchaufour 1982;Noble et al.2017).This condition might affect suppression efforts to extinguish underground peat fires.

Fire is also related to lower organic content (OMC) in HFF that produces char which increases the ash content and acts as a water repellent.This changes the burnt peat to a hydrophobic state up to 90%,wherein it will increase the water drop penetration time (Kettridge et al.2014).The OMC content in HFF was statistically lower than in LFF because of the severe fire incidents.It was <58%,proving that fire consumed the peat organic layers.In Indonesia,the rewetted site and the undrained site show 51.2% and 52.7% carbon content;in the rewetted site,a ditch network was present.Soil organic carbon is associated with peat depth,and the presence of a canal can reduce carbon loss as its input is preserved from the waterlogged conditions (Darusman et al.2022).As a fire consumes the organic matter intensively,it will impact the lower organic carbon content as it is the main deposit in tropical peatlands where almost 90% of carbon losses are related to the burning of organic soils (Che Azmi et al.2021).In this study,Corgwas lower in LFF than in the HFF;however,the distance from the fire-break canal does not significantly influence the peat properties based on ANOVA analysis.For the depths of zero cm (surface),50 cm,100 cm,and 150 cm,there was a significant difference (p<0.05) between the layers in each plot.

Peat properties in the low-frequency fire area (LFF)

In Sabah,the LFF area was the first with fire-break canal in the Beaufort peatland as part of a management effort to reduce fire spread into the Forest Reserve.Based on Fig.2a,the intact forest cover resulted in statistically higher bulk density and lower moisture content than in the HFF area,ranging from 172 to 924%.The porosity level was also statistically lower (56% -90%),which influences moisture availability of the peat.The results correspond to the actual situation of the LFF area,which is not waterlogged on the surface.Another parameter that indicates that the LFF area has lower incidents of fire is that the OMC is higher than in the HFF area,and ranges from 82 to 99%.Fewer fires lower the consumption of the organic layers.Therefore,carbon loss also is lesser than in the HFF area (48% -58%).Peat under saturated conditions has lower decomposition rates to reduce organic carbon loss (Szajdak et al.2020).During burning,peat emits numerous gases,including CO2,CO,CH4,and particle matter,which impact the environment and human health (Hamada et al.2016).Vegetation differences in both the LFF and HFF areas can result in significant small-scale variations in soil organic matter properties(Girkin et al.2019).Lower fire numbers resulted in low ash content (0.5% -18%);a burnt plank of wood (Fig.5) was found [at 225 m away from the canal at below the surface at the depth of–50 cm] (ash content: 11.3%;OMC: 89%;MC: 297%).But fewer fires did not contribute to surface waterlogging such as in the HFF area.A lower ash content also indicates lower water repellence in the LFF area due to the lower fire frequency.Although the area had lower moisture content,fire frequency was low,reflecting the fire-break canal’s function to reduce fire spread.However,it does not fully stop fires from spreading into this area as firebrands from adjacent burning sites can jump to the Forest Reserve with the help of wind (Table 2).

Table 2 Peat properties in Low Fire Frequency (LFF) area based on distance from the firebreak canal

Fig.5 A piece of burnt wood from the sampling site (Photo by Dayang Nur Sakinah Musa)

A rehabilitation approach such as tree planting (SFD 2019) may help the area to recover and improve peat properties.However,the survival of seedlings depends on adaptation to waterlogged conditions,drought,and to fire (Tata and Pradjadinata 2016).A prevention measure of fire-break canal construction can help reduce fire threat and help rehabilitate the area.A study in Riau,Indonesia showed that frequency of extreme fire declined by 40% after rewetting (Taufik et al.2023),which is important for site rehabilitation.Based on the simulation analysis by Warren et al.(2017),the optimal peat swamp forest condition may be achieved over 25 years with complete restoration,including drainage like the state of this study area.Based on the study plot,a fire-break canal did not influence the physical and chemical properties of peat.

Relationship between the HFF and LFF areas properties

Soil properties are not completely independent,and numerous interrelations between properties can be identified(Borůvka et al.2002).Therefore,a detailed analysis of the relationships between variables in this study demonstrates their interactions,especially with regards to moisture availability.In this aspect,a correlation between two variables indicated by r=1.000,and variables with r=<0.200 will not be considered correlated.The hypothesis is:

Ho: There is no relationship between two variables (r=0).

Ha: There is a linear relationship between the two variables (r=1).

An appreciation of the physical and chemical properties of soil as affected by fire frequency is crucial to understanding the relationships between the variables.The findings in this study suggest that porosity plays a vital role in determining bulk density and moisture content in both areas (Fig.6).In the HFF area,repeated fires demonstrate that porosity does not have a correlation (r=<0.200) with chemical contents such as ash,OMC,and Corg.Notably,the firebreak canal in the HFF area does not dry the area as porosity has a reasonably strong relationship with bulk density(r=0.988) and a moderate relationship with moisture content (r=0.584).As moisture is the main control of smouldering material (Nelson et al.2021),it is crucial to understand the properties that influence moisture availability.This will affect the significantly high moisture levels at the surface,indicating the weaker peat’s ability to retain water (Perdana et al.2018),known as a hydrophobic condition because of a previous fire event.In HFF,the porosity level and moisture content have a moderate relationship (0.584) but the three basic peat properties depend on each other.

Fig.6 Physical properties correlation-regression analysis: (a) Influence of distance on porosity (%),bulk density (g cm-3) and moisture content(%) of Binsuluk FR (HFF);(b) Influence of distance on porosity (%),bulk density (g cm-3) and moisture content (%) of Klias FR (LFF) -

The physical properties correlation-regression analysis in the LFF area found that moisture content and porosity had a strong correlation coefficient of 0.838,or 70% of the moisture content depended on peat porosity (p=0.002).Further,the high correlation in the LFF area proved that this area had not been affected by smouldering fires to any extent.There is a relationship between bulk density and Corg.(<0.001,2-tailed) in the HFF area but not in the LFF area.This indicates that,by increasing Corg,bulk density will be lower,and means that denser soil is more capable of storing Corgbased on its structure.

Despite the fire history,the chemical properties were correlated for both study areas (Fig.7).In the HFF area,the organic carbon and organic matter contents are scattered,with a strong positive correlation (0.998) and a strong negative correlation between ash,organic carbon and organic matter contents.The LFF area soil chemical properties in this study have a strong relationship (r=1.000).The LFF area experienced peat fire but not as frequently as the firebreak canal was constructed to prevent the fire from spreading.The chemical properties of OMC,Corg,and ash were strongly correlated (r=1.000) in LFF.OMC is a major component of peat which repels water (Szajdak et al.2020) and influences the compaction of the peat in the HFF area.But its moisture content was not influenced by OMC,as other variables contributed to the presence of moisture in this area.

Fig.7 Correlation regression of peat chemical properties based on the distance from the fire-break canal in both Binsuluk FR: (a)organic matter content (%),organic carbon content (%);(b) ash content (%),organic carbon content (%);(c) ash content (%),organic matter content (%),and Klias FR: (d) organic matter content (%),organic carbon content (%);(e) ash content (%),organic carbon content (%);(f) ash content (%),organic matter content (%)

The results indicate that all chemical variables are interdependent.Ash content was studied to identify the area burnt before showing that it was present from 0 to 150 cm and up to 300 m further from the canal.Based on the analysis of variance,the significant difference between the depth for the HFF area was at 150 m,225 m,and 300 m from the fire-break canal.The results indicate that peat properties were not influenced by the distance from the fire-break canal but by the depth of the peat.The correlation regression analyses were used to relate each chemical property.The total peat organic matter in the HFF area varied from 71.4 to 98.0% at 75 m from the canal,indicating no significant difference (p>0.05) between the depths.The HFF area showed a strong correlation (r=0.998,0.998,1.000) for OMC,Corg,and ash content in the Pearson correlation analysis.The findings were similar to the LFF area,where the variables were strongly related (r=1.000) and had no difference with depth regardless of the distance from the canal (p>0.05).LFF area findings do not affect the chemical properties of the organic carbon and organic matter contents.From the results,lower fire frequency showed that the results gathered and point in similar places in the graphs,contributing to the high correlation (r=1.000).Even though a higher frequency indicates a high correlation of r=0.998,the pattern is straight and the lower fire frequency had little influence.

Fibre analysis and moisture regain of HFF and LFF areas

Peat formation is associated with a high-water table,and this study explored the role of fibre content in influencing moisture.A fibre analysis was carried out to understand the role of the lignin and cellulose fibre contents related to the high moisture content of the decomposed peat near the firebreak canal.In poorly humified peat,the fibre is expected to have higher lignin and cellulose content (McMorrow et al.2003).As moisture plays an important role in protecting the peat from drying,the ability to regain moisture after samples were dried was also studied (Table 3),where the fibre content was theorised to influence moisture availability.Lignin and cellulose contents were determined to understand the potential of lignin to repel water and the hydrophilic cellulose content.Water holding capacity is related to the fibre content and will decrease with increased water retention (Taufik et al.2019).The proportion of cellulose in this study was lower in the LFF area and affected water retention.

Table 3 Moisture regains after drying for HFF and LFF areas based on distance from the fire-break canal

In this study,moisture regain was determined to understand the role of decomposed peat in regaining moisture after drying (Eq.3).Moisture regains compared with the initial moisture content were expressed as a percentage of the total weight of the material.As the samples for fibre analysis were chosen randomly,the ability of the dry peat to regain moisture related to rewetting and the fibre content.The ability for dry peat to regain moisture [through the observation,it was elucidated that the higher moisture content in the HFF area was due to increased fibre content.] This study found that the HFF area had a significantly higher ability to regain moisture (755 ± 40%) than the LFF area after drying (507 ± 39%),(t(94)=4.385,p=<0.001.) A comparison of the dry base moisture content for the HFF area showed that it was lower(p=0.09) with a mean of 655 ± 41% compared to the ability of peat to regain moisture (755 ± 41%).This corresponds to the waterlogging in the HFF area as the fibre content contributes to this.Moreover,in the LFF area,there were no significant differences (p=0.07) between moisture content(407 ± 39%) and the ability of the peat to regain moisture(508 ± 39%).Based on the ANOVA,there was no significant difference between the HFF and LFF areas (p>0.05),indicating that the ability of peat to regain moisture after drying is similar.In contrast,the distance from the canal played a significant role in influencing the ability of the peat to regain moisture.It was significantly greater (p=0.03) at the furthest plot from the fire-break canal (300 m) compared to 75 m,150 m,and 225 m in the LFF area.

For the HFF area,the ability to regain moisture is similar by distance and was not influenced by fibre and cellulose content.[The findings resulted a significant difference(p<0.05) in the HFF area irrespective of the fibre content.] However,the results are considered to be the effect of repeated fire occurrences on the peat to regain moisture.Pearson correlation analysis for the HFF area showed no relationship between cellulose and moisture content and a weak relationship between lignin and moisture contents(Fig.8).However,the LFF area showed that lignin and moisture content have a moderate relationship (r=0.451),which is a fair relationship with r=0.692 between cellulose and moisture content contributing to the increasing of moisture availability.The significantly higher (p<0.05) lignin and cellulose content in the HFF soil contributed to the higher moisture at the surface and lowest at the 100 cm depth where the water table is usually found.The presence of lignin rather than cellulose in peat is higher as it will decrease with high groundwater levels but not cellulose (Iglovikov and Motorin 2019).Warren et al.(2017) reported that lignin depends on reduction as the groundwater increases with depth,where high moisture at the HFF surface increased with depth for the LFF area,which influences moisture availability as lignin is higher.However,underground peat fires might change water absorption properties.Burning dead material contributes to these changes as the HFF soil shows no relationship between moisture content and lignin and cellulose.

Fig.8 Relationship between lignin (r=0.451) and cellulose contents (r=0.692) with moisture content in both Klias FR and lignin (r=0.280)and cellulose contents (r=0.062) with moisture content in both Klias FR

This research on tropical peat properties was limited to near the fire-break canal in low and high fire frequency areas.Malaysia’s fire management measures on fire prevention using fire-break canals and their impact on peat properties improve our knowledge of post-fire soil properties for future conservation programs (Basintal et al.2007).Since the fire-break canal implementation has reduced fire incidences in the forest reserve (SFD 2019,2020),this research is important for global views of peat swamp forests in Borneo (Basintal et al.2007).The findings also add to the importance of the canal blocking study of tropical peatlands in Indonesia,which are similar to this study’s fire-break canal function to rewet and restore the area (Sutikno et al.2019;Ward et al.2021).Results on physical and chemical properties are crucial for future peatlands and forest rehabilitation.

Conclusion

Less attention has been paid to peat-based land-use changes to physical and chemical properties (Könönen et al.2015).This research focused on peat properties near fire-break canals of two other forest reserves representing high fire frequency (HFF) and low fire frequency (LFF) areas.Peat fires release a substantial amount of carbon into the environment(Davies et al.2013).Ash content is higher in the HFF area by 8%,which indicates carbon loss,yet moisture content is high due to the fire-break canal maintaining the water table level.However,changes in peat properties and the environment indicates that peat will become more vulnerable to fire from outside of the forest reserve.Comparing the HFF and LFF areas,there were no significant difference between the physical and chemical properties except for peat porosity(t72.908=5.066,p<0.05).The mean porosity in the HFF area was 9% higher than in the LFF area.This contributes to the higher moisture levels by 434% in the HFF area but lower bulk density and organic matter content by 0.05 g cm-3and 8%,respectively.Due to this lower OMC,the HFF area was lower in Corgby 5% compared to the LFF area.

For fibre content analysis of lignin and cellulose content,the HFF area had a higher average with 397 g kg-1and 54 g kg-1,respectively.However,moisture content did not influence the moisture and regain moisture availability in the HFF area,even though the lignin and cellulose contents were significantly increased further from the fire-break canal.In contrast to the LFF area,the increase of cellulose and lignin content moderately influenced the moisture content and regainability of the peat.Other peat properties might also influence peat moisture in each area.Pearson correlation analysis indicated that ash,organic matter,and organic carbon content do not correlate with the LFF area moisture content.These findings can fill the peat gap at different depths and at maturity,will produce different water contents (Junedi et al.2021).Through this study,distance from the fire-break canal did not influence moisture content,but peat depth up to 150 cm from the surface under different fire frequencies will.

In summary,forest cover and fire frequency in areas near fire-break canals influence peat properties differently and it is challenging to detect the effects of fire-break canals in different fire frequency areas under different environments.This research highlights the protected peat swamp forests in Malaysia and provides information on soil properties near fire-break canals in different hydromorphic settings.It is important to identify type-specific threats and implications for conservation,cultivation and restoration (Gabriel et al.2018).Canal construction such as a fire-break may not critically influence peat physical and chemical properties.But fire frequency might contribute,suggesting that different forest reserves have different properties when fire happens frequently.When peat is repeatedly burnt,its properties will change along with its structure.This study should help to understand the influence of fire-break canals under different fire frequencies.Restoration efforts could reduce drought on soil moisture by at least 19% (Taufik et al.2020),and minimum fire incidents might improve peat properties.Knowing the physical and chemical properties near a fire-break canal will provide soil scientists and land managers with important information on peat soils.

AcknowledgementsThe authors gratefully acknowledge the support of the soil laboratory of the forest research centre and the assistance of the soil laboratory of the faculty of tropical forestry,Universiti Malaysia Sabah.

Author contributionsConceptualisation: DNSM,MZMT,LCA,MSMS,MFS and XH;data curation: DNSM,MZMT and XH;formal analysis: DNSM;methodology: DNSM,MZMT LCA,MSMS,MFS and XH: experimental test: DNSM;supervision: MZMT;writing:DNSM-original draft;writing– review and editing: DNSM,MZMT,XH,LCA,MSMS,and MFS.

Data availabilityData are available under request to the corresponding author: Mohd Zahirasri Mohd Tohir;zahirasri@upm.edu.my.

Declarations

Conflicts of interestThe authors declare no conflicts of interest.