Characteristics of General Soils, Natural Saltlicks and Articial Saltlicks for Wild Asian Elephants (Elephas Maximus) Conservation in the Western Forests of Thailand

Saltlicks are a keystone resource of wildlife. This study aimed to compare the characteristics among general soil, natural saltlicks and articial saltlicks in the natural forest of Salakphra Wildlife Sanctuary (SWS) and restoration habitat for elephant (Elephas maximus) in Kui Buri National Park (KNP), western Thailand. Twenty general soils, 33 natural and 35 articial saltlicks were analyzed. The chemical compositions in natural saltlicks were not consistent. The K, Mg, Fe and Cu in natural saltlicks were higher than articial saltlicks in both areas. The Ca and Zn in articial saltlicks in SES were higher than natural saltlicks of KNP. The salinity in articial saltlicks was highest and higher than in natural saltlicks in both areas and can provide supplemental Na, thereby increasing salinity in both areas. The articial saltlicks cannot provide Ca, K, Mg, Fe and Cu when compared to natural saltlicks that can provide a primary target for elephants at these sites. The ndings have consequences for conservation of elephants and other large herbivores by supplementing essential macro- and micro-nutrients in articial saltlicks. The key resources can provide information to help maintain the wildlife health and fecundity of the region. areas and consume agricultural products. These events chronically lead to conict between humans and wild elephants despite both SWS and KNP being divergent as conservation areas.


Introduction
The distribution range of the Asian elephants (Elephas maximus) is western India, Sri Lanka, northern India, and Himalayan countries such as Nepal, Bangladesh, Myanmar, Thailand, China, Lao PDR, Cambodia, Vietnam, Malaysia, Sumatra Island and Borneo Island 1 . The total population has been estimated to be 41,410-52,345 Asian elephants worldwide 2 . Perera 3 reported that the distribution of wild elephants has increased in Bhutan, India, Sumatra Island and Thailand. Elephants are the symbolic animal of Thailand. In Thailand, the overall population trend was considered constant between 2003 and 2009. The total wild elephant population in the protected areas of Thailand was estimated at 3,000-3,500 elephant in 56,270 km 2 4 . In SWS, an estimated 180 elephants with increased activity near saltlicks were reported 5 . While in KNP, approximately 168 elephants were concentrated in The King's project area, which highlights the importance of habitat improvement to elephants. In the King's project area, 12 small reservoirs were created, and many ponds and hundreds of check dams were built to keep water all year round 6 . About 30 arti cial saltlicks were also created. In addition, two ranger stations were established in 1998 and 2004 in both Kui canal valleys to provide safety for the elephants 7 .
Elephants need food every 12 hours with a dry weight of 1.5% of their body weight during dry season and 1.9% in the wet season 8 . Elephants directly receive minerals such as phosphorus (P), calcium (Ca), sodium (Na), and potassium (K) from food especially in mineral rich soils 9 , but some nutrients such as sodium (Na), sulfur (S) and magnesium (Mg) cannot be obtained from most plants 10 . Both Asian and African wild elephants (Loxodonta africana) must receive these minerals from others sources such as soil, drinking water [11][12][13][14] or termite mounds in the case of African elephants 15 . The macronutrients included Na, Ca, Mg, K, P, and S with an average value at 356.1, 1,540.2, 389.7, 158.4, 61.0, 62.9 and 16.0 ppm respectively. Various natural sources contain Na. Geophagia by herbivore mammals will assist in controlling Na and help in controlling hungriness 16 . Soil consumption is more obvious when the minerals in nature are decreased 17 . Mostly, wildlife was found to eat the saltlicks to supplement Ca and Na intake. In areas, which lack the required soil nutrients or natural saltlicks, arti cial saltlicks are suitable to manage the habitat and maintain the elephants' health 18-21 .
When the agricultural areas were expansion invading in forest areas; especially slash and burn agriculture at the beginning of 19th century reduced the wild elephant habitat in both African and Asian elephants 22 . These activities cause elephant-human con ict (HEC) for more than 100 years [23][24][25] . Causes of con ict can be summarized as land use changes such as forest land use, logging and secondary forest. This made the elephants moved closer to human habitats 22 . In addition, arti cial water sources built adjacent to forests attracted wild elephants to come out of the forest, particularly in dry season 26 . Moreover, building barriers on the travel routes of the elephants affected the elephant's behavior to be more aggressive27. Wild elephant's behavior and ecological characteristics been changed such as some populations that have become familiar with people 28 . Socioeconomics and politics have changed; for example, expansion of pineapple crops by Dole Company from Hawaii, USA in Kui Buri District in Prachuap Khiri Khan province in 1967 29 . Later, there was agricultural expansion as the governmental policy aimed for the local people to settle in forest areas during that period. As a consequence, it has caused con ict between people and wild elephants.
In Thailand, critical issues from wild elephants include damaging agricultural and crop products, accidents and killing people. The trends have continually increased. It found in at least 14 preserved areas such as Phu Luang Wildlife Sanctuary, Phu Kradueng National Park, or KNP, due to expansion of the pineapple crop industry, wild elephants damaged crops by eating the pineapples both in dry and rainy season. This has been a longterm issue until projects were initiated by His Majesty the King and the guidelines set to solve the problems between people and wild elephants 6 . Nelson et al. 23 and Perera 3 suggested methods to solve con icts between people and wild elephants which included traditional methods, killing elephants, translocation, repellant methods, physical barriers, compensation schemes, wildlife utilization schemes and land use planning. Arti cial saltlicks are one of several methods to solve the problems between elephant and human con ict 30 . Consequently, distribution of natural saltlicks for elephants is not suitable. Even if elephants are able to take up nutrients from plants, some minerals are insu cient for wild elephants' needs, resulting in lack of minerals in their bodies. This is evident in male wild elephants that require high Ca to form teeth, tusks and bones18-21. The female wild elephants also need Ca during pregnancy or nursing a baby, because primary minerals of elephants are highly vital for breast milk. The World Wide Fund for Nature 29 reported that the S intake of wild elephants was not adequate, and it is necessary for wild elephants and other mammals to supplement minerals by consuming saltlicks 31,32 . However, there is no previous study exhibiting that minerals in saltlicks are su cient for wild elephants. SWS and KNP located in the western forest of Thailand are well known for habitat of wild elephants. Wild elephants in these two areas are known to utilize in agricultural areas and consume agricultural products. These events chronically lead to con ict between humans and wild elephants despite both SWS and KNP being divergent as conservation areas.
The SWS is a natural forest, while KNP is a restored forest. Although both areas have saltlicks, there has been no clear approach and standard analysis of mineral content in either arti cial or naturally occurring saltlicks. We aim to determine whether or not minerals in both the arti cial and natural saltlicks of SWS and KNP are su cient for wild elephants. These nding are important in terms of providing standardized approach to ensure that the mineral requirements of elephants are met with the provision of arti cial saltlicks, thereby potentially decreasing human-elephant con ict. Study sites. Salakphra Wildlife Sanctuary is located in Kanchanaburi province and is 858.55 km2 in size. It is located from latitude 14° 9' to 14° 40' N and longitude 99° 9' to 99° 30' E. Most areas form a mountainous complex with 50 − 1,178 m height differences from mean sea level. The highest mountain peak is Kao Hua Loan Mountain, 1,178 m from mean sea level. Average highest temperature is 34°C and lowest temperature is 23.1°C. Important water sources are Kao Huai Sadong and Huai Salakphra which are located in the central part of the area. Its length is approximately 26 km. Plant communities include dry dipterocarp forest, dry evergreen forest, mixed deciduous forest, bamboo forest and tropical grassland. The geology is Karst topography, metamorphic rock and sedimentary rock. There is also a small amount of igneous (granite) rock in this area 33 (Fig. 1). Kui Buri National Park is located in Prachuap Khiri Khan province and is 969 km2 in size (14° 9' to 14° 41' N and 99° 10' to 99° 25' E). The geography is mostly part of the mountain range complex of Tenasserim ridge with an average elevation of 750 m above sea level. The forest community consists of mixed deciduous forest, dry evergreen forest and tropical rainforest 34 (Fig. 1). Soil chemical analysis. Cation exchange capacity (CEC) and saturation of exchangeable bases (Ca, Mg, K and Na) were determined using ammonium acetate (one natural and neutral). Electrical conductivity (EC) was determined in a saturation extract, and particle size (texture) identi cation was performed using the hydrometer method 37 . Phosphorus (P) was determined using the Bray II method. Trace elements (Fe, Cu, Mn, and Zn) were extracted with DTPA 38 . Organic carbon concentration was determined by the Walkley-Black method 39 . The pH was also determined for each sample 40 . The results of these analyses were compared to the control samples (n = 7) that were obtained from 30 cm below the soil surface in places where there was no evidence of geophagy.

Materials And Methods
Statistical analysis. Box plots were used to illustrate the distribution of data among different treatments. ANOVA were performed to identify whether the soil reaction, salinity and chemical composition (macro-and micro-nutrient) of the soil types between the two study areas were different from one another with signi cant differences set at the p-value = 0.05 level. The correlative coherence analysis (CCA) was used to test the relationships between 10 minerals and environmental parameters (pH and salinity) among general soil, natural saltlicks and arti cial saltlicks samples in Salakphra Wildlife
Macronutreints. Concentrations of P was signi cantly different (F = 7.116, df1, 2 = 5, 82, p < 0.0001) among the experiments, but inside the SWS and KNP were not signi cantly different (p > 0.05). Phosphorus was highest in natural saltlicks in SWS. In general soils and natural saltlicks in SWS were higher than general soils and natural saltlicks in KNP (p < 0.05). After supplement of minerals P in arti cial saltlicks of SWS was higher than KNP (p < 0.05) (  Fig. 2) The Na was signi cantly different among general soils, natural saltlicks and arti cial saltlicks (F = 3.197, df 1, 2 = 5, 82, p = 0.011). Concentrations of Na were not signi cantly different between natural and arti cial saltlicks in both areas (p > 0.05). These saltlicks were signi cantly higher than in general soils in both areas (p < 0.05) ( Table 1 and Fig. 2).
Concentrations of Se among general soils, arti cial saltlicks and natural saltlicks were not signi cantly different (F = 1.348, df 1, 2 = 5, 82, p = 0.252) ( Table 1 and Fig. 2) Relationships among arti cial, natural saltlicks and general soils. The relationships among arti cial, natural saltlicks and general soils in SWS and KNP, environmental factors and minerals was analyzed by using CCA (Table 2   Elgon National Park 44 (Table 3). SWS = Salakphra Wildlife Sanctuary, KNP = Kui Buri National Park, N/A = > 70% of the sample were either non detectable or below the limit of quanti cation and therefore at very low concentration, -= not analyze.
The most common chemical component in soils in SWS and KNP were Fe and Mg (natural saltlicks > arti cial saltlicks > general soils). At our study area, the amount of Se was consistently the lowest of all nutrient types in both areas with natural saltlicks, followed by arti cial saltlicks and then general soils having the lowest concentrations respectively. Therefore, it can be seen that in both areas Fe and Mg were highest and Se was lowest. This nding was similar with Huai Kha Khaeng Wildlife Sanctuary 45 . Minerals were found in natural saltlicks at higher levels than general soil as was found in natural saltlicks of Huai Kha Khaeng Wildlife Sanctuary. These minerals were higher than baseline levels of trace metals in the food (Mg, Na, Zn and Fe) in zoo 46 and blood (Fe, Zn, Na, Mg and Ca) of captive Asian elephants 47 (Table 3).
Higher concentrations of Mg, Ca, Mn, Fe, Zn and Cu in natural saltlicks than in general soil can thus represent an important nutrient supplement to wild elephant enabling their forage to sustain their body and health 45 , especially Ca required to support their bones and ivory 48 . Seidensticker and McNeely 45 also describe Na and an important driver of geophagy.
In general soils, SWS and KNP differed in their amounts of minerals with soil in SWS having higher mineral contents than in KNP due to SWS having been designated as a protected area by the Royal Forest Department pursuant to Wildlife Reservation and Protection Act 2019 and not having been subjected to resource degradation over time such as in KNP 49 . The degradation of KNP was mostly caused by other land use demands as humans were unrestricted in converting forest to cropland and hunted wild animals for food. The Conservation and Restoration of Kui Buri National Forest Project under His Majesty the King's Royal Initiation began in 1997 34 . However, levels of K, Na and Se were similar in both study sites. While KNP had salinity in general soils at higher levels than in SWS. This result agreed with the study on the development scheme of lands with agricultural di culties in the western area of Thailand, which reported that the soil series under soil salinity problems were the Tha Chin Series, which were founded in Prachuap Khiri Khun and the Nong Kae Series 50 .
In natural saltlicks, P, K, Ca, Mg, Ze and Cu are rich at both study areas, while Ca and Mg were lower in SWS than KNP. Na, Fe and Se were not signi cantly different in both areas. Hence, SWS is not faced with minerals de ciency in natural saltlicks, but KNP may need to supplement some minerals to enrich the minerals in the arti cial saltlicks depending on the spatial condition and composition of soil in each area. This difference in mineral concentration may lead to differing mineral concentration in the forage of elephants 51 .
Arti cial saltlicks were found to be important to enrich the minerals in general soils and elephants, but, P, K, Mg, Mn, Ze, Fe and Cu in SWS and KNP were different, because of differences in historical land-use practices over time, and differences in the provision of arti cial saltlicks in each area 51 . Currently, providing arti cial saltlicks is not speci cally formulated and hence, each area should add minerals associated with the speci c soil condition of the local area rather than merely providing arti cial saltlicks with salt that attracted few wild animals for utilization 52 .
Mineral compositions in general soils and natural saltlicks in both areas were different. For preparation of future arti cial saltlicks, the mineral composition of the local area needs to be understood as a guideline for making cost-effective and suitable arti cial saltlicks for elephants and other geophagies.
This study suggests that the concentration of macronutrients in descending order for natural saltlicks in SWS was Ca > Mg > Na > K > P, while macronutrients in arti cial saltlicks were arranged in the following descending order as Mg > Na > K > Ca > P. For KNP, the natural saltlicks had a descending concentration order of Mg > Ca > Na > K > P, while for arti cial saltlicks it was found to be Mg > Na > Ca > K > P. The salinity was found signi cantly higher in the arti cial when compared to the natural saltlicks, while P, Ca, Na, Mn and Zn were not different. Overall SWS contained higher levels of nutrients when compared to KNP. According to the high number of arti cial saltlicks in KNP, salinity in KNP was much higher than in SWS as these arti cial saltlicks add salt to the system. Other minerals such as P, K, Mg, Mn, Ze, Fe and Cu in both areas are not signi cantly different. Ca, Na and Se were higher in SWS than in KNP. Finally, managers should provide arti cial saltlicks that are speci cally formulated by adding minerals based on the local spatial distribution of nutrients found in the general soil types of a speci c area to make them suitable for wild elephant and other geophagies. We recommend further studies on the relationships between wild elephant distribution and spatial distribution and the occurrences of natural saltlicks in the area. Forage samples should be investigated for their composition of minerals to ensure the quantities are suitably available for elephants. Arti cial saltlicks are important where low concentrations of mineral elements are found in forage or natural saltlicks. Figure 1 Salakphra Wildlife Sanctuary in Kanchanaburi province and Kui Buri National Park in Prachuap Khiri Khan Province Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.

Figure 2
Soil reaction, salinity and chemical composition of general soils (Soil), natural saltlicks (Natural) and arti cial saltlicks (Arti cial) from Salakphra Wildlife Sanctuary (gray) and Kui Buri National Park (black)