Long-term monitoring of breeding successes of Great Hornbill ( Buceros bicornis ) and Rhinoceros Hornbill ( Buceros rhinoceros ) using artificial nest boxes in Budo-Su-Ngai Padi National Park, Thailand

. Nest boxes are a conservation tool for increasing the availability of nests. The efficacy of nest boxes can be assessed by examining the breeding parameters of birds in nest boxes to see whether data are comparable with tree cavities. The hornbill artificial nest research was conducted in Budo-Su-Ngai Padi National Park, Thailand since 2003. This project aims to increase the breeding propensity of two large species of hornbills; the Great Hornbill ( Buceros bicornis ) and the Rhinoceros Hornbill ( Buceros rhinoceros ) that coexist in the study area, by utilizing nest boxes. The nest boxes were constructed from fiber-reinforced plastic. There were two designs; small nest box: 50 x 49 cm at base and 95 cm height, and large: 65 x 70 cm at base and 95 cm height. Twenty nest boxes were installed between 2004 and 2006. These nest boxes were monitored from 2004 to 2021. The study results indicated that the proportion of nest occupancy of Great Hornbill breeding in natural nests was higher than nest boxes: 48.9% and 31.4%, respectively (the two binomial proportions test gives: Zc= 4.01, P < 0.0006 < α = 0.05). The results also showed that the Great Hornbill prefers to select a large nest box than a small one: 37.3% and 17.3%, respectively (the two binomial proportions test gives: Zc = 3.4, P < 0.00068 < α = 0.05). Eight nest boxes so far have been utilized by hornbills: six by Great Hornbill with 58 nesting successes and two by Rhinoceros Hornbill with 1 nesting success. The average life cycle of the fiberglass nest box is about 11.38±4.23 SD years. The study results suggest that reproduction parameters of hornbills in nest boxes are different from natural cavities in many aspects. Our two target species do not equally benefit from nest boxes.


INTRODUCTION
The provision of nest boxes is a conservation measure widely used to increase nest site availability for secondary cavity nesting birds (Mänd et al. 2005).Different species of bird respond differently to nest boxes (Purcell et al. 1997).Thus monitoring of nest utilization was required in the artificial nest program to be able to assess the efficacy of the nest box.The data from nest monitoring could be compared between nesting of target species in artificial nests and tree cavities to examine whether breeding parameters of bird nesting in nest boxes are comparable with nesting in tree cavities (Purcell et al. 1997).
Nest boxes have been successfully used with various species of birds in wild bird conservation, e.g., Scarlet Macaw (Ara macao; Olah et al. 2014), and American Kestrel (Falco sparverius; Shave 2017).However, reports of nest boxes utilized by large secondary cavity nesters such as hornbills were rare, even though hornbills were selected as a conservation flagship species in many regions (Wilcove 2010) and 25 species of Hornbills were listed as either globally threatened or near threatened species by the International Union for Conservation of Nature (Poonsawad. et al. 2013).
Hornbills are large forest birds classified in the order of Bucerotiformes (Kemp 1995).Their prominent characteristics are an oversized, long, curved bill with a casque on top.There are two families, Bucorvidae and Bucerotidae, with 15 genera, and 62 species; 32 species are in Asia and 25 species are in Africa (Poonswad et al. 2013a).Thirteen of the 32 Asian hornbill species are found in Thailand, and six of them are found in Budo-Su-Ngai Padi National Park in Thailand (Poonswad et al. 2005).
Hornbills are omnivorous as they eat a combination of plant and animal food (Poonswad et al. 2013a).Asian hornbills predominantly feed on fruits and occasionally on small animals.From 252 genera and 79 families, 748 plant species are reported in the Hornbill diet (Kitamura 2011).Hornbills digest only the fruit pulp and then regurgitate or defecate the seeds (Kitamura 2011).Hornbills have relatively long seed retention times, so the propensity for depositing seeds far away from the parent trees is high (Kitamura 2011, Corlett 2017).These are the characteristics of an efficient seed disperser; hence Hornbills provide excellent seed dispersal services to tropical plants ( Kitamura 2011, Poonswad et al. 2013a).
Hornbills have unique nesting habits.The female seals herself into the cavity of large tree and leaves only a narrow slit through which the male passes the food to her and their chicks (James and Kannan 2009).Although hornbills nest in cavities, they are unable to excavate their own nest holes (Poonswad et al. 2013a).Because of their large size, they need large nest cavities to breed, so their breeding depends on large trees, which exist only in primary forests.The availability of appropriate nest cavities may be the most important population-limiting factor (Poonsawad 1995, James andKannan 2009).
Most Asian hornbill nests are in Dipterocarpus trees genus (James and Kannan 2009), which are heavily harvested for their timber (Poonswad and Kemp 1993).Therefore, logging seriously reduces the availability of both potential nest trees and suitable cavities.Also, human settlements in the forest areas result in fragmentation of hornbill habitats (Poonswad and Kemp 1993).It is known that hornbill populations are limited by the size of habitat patches (Sitompul et al. 2004, O'Brien et al. 1998, as cited in Poonsawad. et al 2005).In addition, to sustain hornbill populations and for comprehensive management of their habitats, enhancement of nest sites and provision of assisted nesting capability are necessary.To increase the chance of survival, this is especially important in secondary forest nest cavities because potential nest trees are scarce.
The aim of this study was to increase the breeding propensity of two large species of hornbills, the Great Hornbill (Buceros bicornis) and the Rhinoceros Hornbill (Buceros rhinoceros), by using artificial nests.These two species are classified as near threatened with CITES Appendix I for Great Hornbill and CITES Appendix II for Rhinoceros Hornbill (Poonswad et al. 2013a).In addition, this study also aims to explore the feasibility of usage of artificial nests by hornbills in their natural habitat.This study was divided into two parts (Pasuwan et al. 2011(Pasuwan et al. , 2015)).The first part was to design and implement artificial nests for hornbills by using their biological data as criteria for design.The second part was to determine the suitability of the artificial nests.Thereafter, we set up long-term monitoring and data collection in order to determine the results of hornbill utilization and the lifespan of nest boxes.This paper covers the data of 17-year monitoring from 2004 to 2021 and includes some records of observation from the second part of this project.
The data collected from this monitoring were analyzed to determine the efficacy of an artificial nest based on the hypothesis that could be described as follow: if nest boxes were in a suitable condition and located in an appropriate environment, the hornbills' nest occupancy proportion for the natural nests and the nest boxes does not differ significantly.
In addition, we also generated a sub-hypothesis to understand the hornbill selection preference.The sub-hypothesis could be described as follow: if nest boxes are in a suitable condition and located in an appropriate environment, the hornbills' nest occupancy proportion for large and small nest boxes does not differ significantly.

Study area
This study was conducted at Budo-Su-Ngai Padi National Park located in Thailand's three southern provinces: Narathiwat, Pattani, and Yala (Fig. 1).With an area of 341 km², the park is situated in two mountain ranges, Budo and Su-Ngai Padi.The area has forest patches that are separated and surrounded by human settlements and agricultural lands.The national park is in the Indo-Malayan tropical region that supports the Malaysian or Sundaic flora (Wells 1999and Woodruff 2003, as cited in Poonswad et al. 2005).Budo mountain range supports six species of hornbills: Great Hornbill (Buceros bicornis), Wreathed Hornbill (Rhyticeros undulatus), Rhinoceros Hornbill (Buceros rhinoceros), Helmeted Hornbill (Rhinoplax vigil), White-crowned Hornbill (Berenicornis comatus), and Bushy-crested Hornbill (Anorrhinus galeritus; Poonswad et al. 2005, Trisurat et al. 2013)

The artificial nest box
The artificial nest boxes in this study were constructed from fiberreinforced plastic and insulated with polyurethane foam.The nest entrances were made of hardwood that was carved into ovalshaped entrances.The nest entrance was 20 cm tall, 13-15 cm wide, and 9 cm thick.
We designed the overall dimensions of a nest box based on the information from both an average natural nest (Poonsawad 1995) and sizes of nest boxes that have previously demonstrated success in captive breeding (Choy 1980, Golding andWilliams 1986).From the reviewed literature, the diameter of nesting chamber sizes at base ranges from 40 to 60 cm and nest height was about 1 meter.
After reviewing the resources mentioned above, we eventually agreed upon two designs of the nest boxes.Both designs were overall similar except for the size of a nesting chamber.The nesting chambers of these two designs were constructed based on the lower and upper bound found in the literature (Fig. 2).The small nest boxes were 95 cm tall, 50 cm long, and 40 cm wide.Twelve nest boxes of this size were made.The large nest boxes were 95 cm tall, 70 cm long, and 65 cm wide.Eight large nest boxes were made (Fig. 2).
The artificial nests were installed by hanging the nest boxes underneath a big branch of a large tree.The installation locations were selected using three main criteria (Pasuwan et al. 2011): 1.No active hornbill nesting within at least 300 meters from the installation area; 2. Existing reports of the targeted hornbill species perching on the tree or nearby trees of intended site of installation; 3. The locations of intended installations were at least 100 meters from an area with human activities.
The average distance between nest boxes was about 467.8 ± 169.98 meters (calculated from the main group of nine nest boxes installed on the eastern slope of the study site).

Data collection methods
This study was based on the field research design with data collected through direct observation.An observation blind was constructed at a distance of 30-50 meters away from each nest box.Ground observation was conducted from inside these blinds using binoculars to avoid disturbing the birds.Data collection was performed from February to August, coinciding with the breeding season of the hornbill.
We divided data collection frameworks into three phases according to hornbill nesting behavior (Chaisuriyanun et al. 2011): the pre-nesting, female imprisonment to female emergence, and chick fledging.The pre-nesting phase was between February and March for the Great Hornbill and from late February to May for the Rhinoceros Hornbill (Chaisuriyanun et al. 2011).In this phase, a pair of hornbills search for a breeding cavity.When they find a suitable one, the female will start sealing a nest entrance.
The nest sealing takes about 13-14 days to finish (Kaur et al. 2016).The completion of nest sealing is the end of the pre-nesting phase.This marks the start of breeding activities, which include egg-laying and chick-raising in the second phase (Poonswad et al. 2013a).As stated above, there are two main hornbill behaviors in the pre-nesting phase: nest visiting and nest sealing, so we adjusted the data collection practices to suit both the nature of each behavior and data requirements for our research analysis.
During early period of the pre-nesting phase, our purpose of observation was to determine the possibility of hornbills using nest boxes.At this period, the data collector conducted a weekly ground search for feces and seeds under the nest boxes to find signs of hornbill visiting.If these signs were present, the data collector set an observation schedule for 6 hours from 9 AM to 3 PM and continued for three consecutive days in order to verify hornbill visiting using visual evidence (Pasuwan et al. 2015).
If any nest sealing activity by a female hornbill was observed, a data collector would record first date of observed sealing.After that, they would conduct the nest checking by direct observation once a week and continue for two weeks to determine the completion of nest sealing.After two weeks, a nest entrance should be sealed and a small elongated vertical slit should be seen.We recorded nests with sealed entrances as a nest occupancy and otherwise as a non-occupancy The completion of nest sealing was considered to be the most pertinent data for analysis because it was a clear indication of the nest occupancy.The second phase or the breeding phase was the period when the female finished sealing the nest entrance and imprisoned herself until her emergence.In this phase, the female lays eggs, and raises a chick while only the male provides food.This period takes about 100 days (Chaisuriyanun et al. 2011).In this phase, we conducted a weekly nest check to make sure the process was running smoothly.
The last phase took about 21 to 28 days, starting from female emergence to chick fledging (Chaisuriyanun et al. 2011).Because this is the last phase of the breeding cycle, we could estimate the chick fledging date of occupied nest boxes by comparing our recorded data with the average nesting duration reported by other researchers.The average nesting duration of the Great Hornbill is about 102 to 144 days and 122 ± 10 days for the Rhinoceros Hornbill (Poonswad et al. 2013a, Pawar et al. 2018).
During this last phase, the data collector conducted weekly nest checks until the estimated chick fledging date.Nest checking was conducted more frequently (about 3 to 4 days a week and continued until the chick left the nest).In most cases, the observer might not actually see the chick emerge, as the chick could emerge at any time of the day.However, we could still determine the nest success by two methods.First, because the chick and its parents usually hang around the nesting area for a few days after the chick fledged, the data collector would likely find them or hear them calling when surveying the area.Second, when the data collector did not find the chick and its parents around the nesting area, we were still able to determine nesting success by comparing the total nesting duration of our recorded data with the average nesting duration reported by other researchers.
After the chick left the nest, we recorded any nest that hornbill used with full breeding cycle as a breeding success.It should be noted that although breeding success indicated the completion of a breeding cycle, this data was not appropriate for the assessment of nest box efficiency.Because of the fact that breeding success was caused by the combination of various factors, we did not use this data for hypothesis testing but mainly for discussion.
The details of a data collection form used in this study included name of data collector, the identification number of the nest box, date/time of observation, and observation results.The observation results section included a checklist of the hornbill visiting, time of visiting and leaving, date of the first observed nest sealing, and the estimated chick fledging date.

Maintenance checks
Maintenance checks of the nest boxes were conducted yearly either in the months of October or November.There were many causes of defects in the nest boxes that required maintenance.Defects such as wall cracks could be found by visual inspection from the ground using binoculars.However, defects such as a loss of bedding material cannot be examined with this method.So, we performed the preliminary maintenance checks prior to conducting detailed investigation of each nest.These preliminary maintenance checks were conducted in two ways: examining the documented nesting information and visually inspecting from ground by observer using binoculars.
Examine nesting data could reveal some clues that may point to some types of a nest box's defects.For example, if a hornbill suddenly stopped using a nest that was used previously with continuously breeding successes, this would likely indicate an unsuitable condition such as the loss of bedding material.With the information gathered, we could pinpoint the particular nest for detailed checking by using rope climbing to check and repair the nest box.

Data analysis
We found that the Great Hornbill was the only species that occupied the artificial nest boxes and succeeded in breeding in big numbers, whereas there was only one nesting success recorded from Rhinoceros Hornbill.Therefore, we could only perform the statistical analysis by using data recorded from the Great Hornbill.Microsoft Excel 2019 was utilized for analysis.
In this study, we used a two-sample binomial proportion test for analysis.This statistical method is widely used for comparing the proportion of data obtained from our binary variables (Oliveira 2020).Based on our hypothesis stated previously, we conducted a statistical comparison between the nest occupancy proportion of known natural nests collected by Thailand Hornbill Project during 2008-2020 and nest boxes collected during the same duration.Further, we conducted another statistical comparison between the nest occupancy proportion of the small and large nest boxes.We use the significant level of p < 0.05 in our hypothesis testing.Beyond hypothesis testing, we also performed a descriptive statistics analysis on the data obtained.
The analysis of nest occupancy proportion between natural nest and nest box was the comparison of the data between different groups, so in order to make them comparable, we selected the data collected during the same duration from both groups.However, when we considered the trend line of the nest box data collected throughout our study, we found that the nest occupancy number gradually increased during first three breeding seasons after the installations (2004)(2005)(2006)(2007) until relatively stable in 2008.In order to improve the data quality and consistency, we did not use the data recorded from the nest boxes during the first three breeding seasons from 2004 to 2007; rather, we used data from both groups recorded from 2008 to 2020 in our statistical analysis.
The analysis of the nest occupancy proportion between large and small nest boxes was the statistical comparison between data within the same group, so the selection conditions of the data were different.In this statistical analysis, we used the data collected from the nest boxes throughout our study from 2005 to 2020; however, we disregarded the data of each nest box recorded in the installation year.
In order to make the data appropriate for two proportion binomial analysis, we categorized and classified datum as follows: Each usable nest that was available in each specific breeding season was classified as an experimental trial.The nest occupancy data was classified as an experiment result and was categorized into binary choices: occupancy and non-occupancy.
From the classification stated above, there were a total of 417 trials with 204 occupancies for natural nests and 185 trials with 61 occupancies for the artificial nest boxes during 2008-2020 (Fig. 3).In terms of box sizes, there were a total of 115 trials with 43 occupancies for large nest boxes and 115 trials with 20 occupancies for small nest boxes during 2005-2020 (Fig. 4).

RESULTS
In 2005, one year after installation of the artificial nest boxes, the Rhinoceros Hornbill was the first species observed when inspecting nest boxes.However, no breeding attempts by any hornbill species were reported using the nest boxes that season.
In 2006, the Great Hornbill was the first species that we observed to breed successfully in an artificial nest box.The utilization of nest boxes by Great Hornbill increased over the years until it reached six nests out of total 15 in 2014; this is the highest number of nest boxes occupied by Great Hornbill ever recorded.After that, Great Hornbills have been observed using the same nest boxes every season (Fig. 5).One pair has been nesting in the same box for 14 years.Although the maximum number occupied by hornbills never exceeded six nests, this number has been relatively stable (Fig. 5).Great Hornbills have been observed to be visiting other vacant nest boxes in recent years so it is possible that the number of nest boxes used by Great Hornbill could increase in the future.
Rhinoceros Hornbills were also spotted visiting and inspecting the nest boxes but only two nest occupancies were made according to our long-term monitoring; one failed attempt in 2011 and one successful nesting in 2021 (Fig. 5).
From our record of eight nests that have so far been utilized by hornbills, none were occupied during the first breeding season after installation; however 66.6% of the utilized nest boxes were used between the second to the fourth breeding seasons after installation.The total occupancy of nest boxes by Great hornbills during 2008-2020 was 31.4%.
The comparison of the nest occupancy proportion between known natural nests and nest boxes using two binomial proportion analysis can conclude (at a significance level of 0.05) that the nest occupancy proportion of natural nests by Great Hornbill was higher than artificial nests (Z c = 4.01, P < 0.0006 < α = 0.05).
When considering the nest occupancy of Great Hornbill recorded from two designs of nest boxes during 2005-2020, the cumulative occupancy of the large nest box was 37.3% and 17.3% recorded from the small nest box.The comparison of the nest occupancy between small and large nest boxes using two binomial proportion analysis can conclude that the Great Hornbill prefers to select the larger nest boxes than the smaller ones (Z c = 3.4,P < 0.00068 < α = 0.05).
In order to collect some visual evidence related to hornbills' choices for the nest boxes size, we installed one large and one small nest box in the same tree.Observation records from these indicated that both Great and Rhinoceros Hornbills visited both nest boxes for six breeding seasons but both were not occupied.
In 2012, we took the smaller nest box down and left the larger one on the tree.After that, in 2014, the second breeding season after we took the smaller nest boxes down, Great Hornbill occupied the larger nest boxes until the end of the study.
In terms of nest box management, the average maintenance period from the first installation stage to the first maintenance stage of the artificial nest boxes was about 11.38 ± 4.23 SD year.The causes of maintenance were as follows: 13 cases of loss of bedding material, four cases of loose steel wires, and two cases of wall cracks.
From a total of 20 nest boxes installed at the study site from 2004 to 2021, 13 remained in use until the end of the study, three were broken from trees falling down, two were taken down because of the cracks found on the box walls, and two were study samples that were removed as the sampling process was discontinued.

DISCUSSION
The potential explanations of these study results were discussed with the findings of other similar study to generate the implications for future conservation strategy.Three discussion topics were conducted as follows: the Great Hornbills' nest selection, breeding success rate of the Great Hornbill in artificial nests, and the effects of nest boxes on target species.

Great Hornbills' nest selection
Although six nest boxes were occupied by hornbills, the analysis indicated that Great Hornbills have a higher tendency to select natural nests compared to artificial nests (31.4% for nest boxes and 48.9% for natural nests).Despite not knowing the exact reason for this, this lower tendency to select nest boxes was unlikely to be associated with the unsuitability of artificial nest boxes.Because nest boxes (used in this study) were constructed based on the characteristics of the natural nest cavity of the targeted species (Pasuwan et al. 2011) and there had been more than 50 cases of breeding success recorded from nest boxes throughout our study, there was enough convincing evidence to confirm the suitability of nest boxes in many aspects.
Though, this lower tendency to select nest boxes could be caused by a combination of various factors, based on evidence obtained in our study, we propose a possible explanation as follows.The searching process and the method for locating suitable natural cavities in this study relied on visual evidence of hornbill nest occupancy and nest visiting, e.g., following male hornbills' flying direction during breeding season usually leads to their nest location.From this method, natural nests were able to be located only after they were occupied by hornbills.The unoccupied and non-visited cavities could not be found by this method, so some more natural cavities may have existed in the study area without being detected by data collectors.If the number of these undetected nests was high, this could have affected the number of available natural nests recorded in our database.This could have caused the nest selection tendency of natural cavities to be higher than it should be.
Moreover if the number of suitable but undetected nests was high, it would have caused an increase in nest density in some areas without being detected.When nest boxes were installed in an area with a number of undetected nests, the addition of nest boxes could cause the nest density to increase in that area.The high density of suitable nests could cause undesirable consequences in nest selection of birds.Robertson and Rendell (1990) described that the high density of suitable nests will attract more breeding pairs to visit the area, which could cause fiercer competition for nests, and finally could result in nest desertion as described by Poonsawad et al. (2005).Although the above scenario has never been reported during our study period, the possibility of this occurrence cannot be ruled out.The relationship between nest density and nest selection was also described by Deeming et al. (2017) and Serrano-Davies et al. (2017) in their studies of nest boxes with Blue Tits (Cyanistes caeruleus) in which some nest boxes were unused because of more spatial separation distance required between nests of each breeding pair of the same species.Poonsawad et al. (1987) also describe that the distance between Great Hornbill nests was approximately 200 meters.Therefore this study of Great Hornbills' home range and nest spacing in our study area could be useful to determine the optimal density of nests to make more appropriate decisions on the management of nest boxes.

Breeding success rate of Great Hornbill in artificial nests
Contrary to the nest selection rate that was higher in natural cavities, the percentage of breeding success of the Great Hornbills in the nest boxes was higher than the natural nests in the same timeframe (83.8% for nest boxes and 69.65% for natural nests).This result conforms with the results of some other nest box studies conducted on various secondary cavity nesting species that described the higher breeding success rate of nesting in nest boxes than in natural cavities (East and Perrins 1988, Alaalo et al. 1990, Kuitunen and Alekninis 1992, as cited in Purcell et al. 1997).However, all the above studies that reported higher breeding success in nest boxes than natural cavities were conducted with Passeriformes, whereas our study first reported the same results with Bucerotiformes.
However the higher breeding success rate of birds nesting in nest boxes, as stated above, was not reported by all studies; some studies reported otherwise, such as lower breeding success rate in nest boxes (Evans. et al. 2002) or a similar breeding success rate between nest boxes and natural nests (Slevin et al. 2018).These differences are attributed to differences in various factors of studies such as study location and habitat type, or even the different responses of species to nesting in nest boxes (Purcell et al. 1997).
Artificial nest boxes differ from natural nests in several ways.These differences create conditions unlike those occurring in natural nests.These different conditions both directly and indirectly affect the reproduction parameters of nest box residents.The implications of this discussion suggest that Hornbills' reproduction parameters collected from artificial nest boxes should be used very carefully when used as reference to the reproduction parameters of population nesting in natural cavities.As also suggested by many studies, the breeding ecology of birds nesting in artificial nest boxes may not be a good representative of birds nesting in natural cavities (Robertson and Rendell 1990, Purcell et al. 1997, Evan. et al. 2002, Lambrechts et al. 2010) The effects of nest boxes on target species Different species of birds may respond differently to nest boxes (Purcell et al. 1997).Therefore any species that adapt to breed in nest boxes better than others will get more benefits from these additional nests.In this study, our target species (Great and Rhinoceros Hornbill) were similar in many aspects.Poonswad et al. (2013b).This ratio of chick fledging from nest boxes of both of our target species was part of the evidence of the effect of additional nest boxes on the community structure in our study area.
Although the low breeding rates of Rhinoceros Hornbill in nest boxes could be caused by the combination of various factors, evidence from both our study and others suggested that the interspecific competition for nests between these two species could be one of the factors that affected the Rhinoceros Hornbill's utilization of the nest boxes in our study site.Poonswad et al. (2005) described that the evidence of competition between species for cavities in our study area accounted for about 26% of total available cavities.In addition, there was an observation reported in this study that Great Hornbills chased Rhinoceros Hornbills prior to visiting artificial nests in 2005.
In addition, our study area at Budo-Su-Ngai Padi National Park is located in a transition area called the Kangar-Pattani line (Hughes et al. 2003).The Kangar-Pattani line is the biogeographical transition area between mixed moist deciduous forest of the northern Thai-Malay peninsula and wet seasonal evergreen dipterocarp rain forest (Hughes et al. 2003).Although the core range of the Great Hornbill lies in mixed moist deciduous forest to the north (Van Steenis 1950and Whitmore 1984, as cited in Gale and Thongaree 2006), the Rhinoceros Hornbill is the resident of the Malayan evergreen rain forest of the south.Our study area is considered to be in an overlapping distribution area between these two species and also the northernmost range of the Rhinoceros Hornbill in mainland Southeast Asia.Because these two species coexist in the same area, and because similar nest characteristics are required by both species, they have to compete for limited availability of the same resources.This causes the interspecific competition for nests between these two in our study area.
In terms of conservation management, when comparing the average maintenance period between nest boxes (counting from the first installation period to first maintenance period) and natural nest cavities (counting from first report of hornbill utilization to first maintenance time) recorded by Thailand Hornbill Project, artificial nests are more durable than natural nest cavities (11.38 years for artificial nest boxes and 8.46 years for natural nest cavities).

CONCLUSION
Although both the Great and Rhinoceros Hornbills were reported breeding in captivity (Wilkinson 2005), there are very few published reports of these species breeding in nest boxes in their natural habitat.The study results indicated that Great Hornbills prefer to select natural nests over nest boxes.Further, they prefer large nest boxes more than small ones.Because there was some breeding success of Great and Rhinoceros Hornbill in the artificial nest boxes, the utilization of an artificial nest was considered to be one useful conservation strategy to increase the breeding propensity of these two large species in their natural habitats, especially when large trees are scarce.
This study results also raised the question of whether the introduction of an artificial nest box had an impact on the hornbill breeding cycles as well as the relationship between the species.These topics are worthy of further studies.
We also recommend future studies that focus on the interactions and competition between the Great and Rhinoceros Hornbills with attention on the niche of less adaptable species, especially the Rhinoceros Hornbill in this area.Future conservation efforts should also be focused on these less adaptable species and the species' ecological niche so that they can co-exist in the long term.

Fig. 1 .
Fig. 1.Location and terrain map of study area: Budo mountain range, Thailand.(Created by Thailand Hornbill Project).

Fig. 2 .
Fig. 2. Top left, top right, and bottom right: Diagrams showing the fabrication and the arrangement of parts for the artificial nest-box.Bottom left: Small and large nest boxes.
They are both large hornbills and are sister taxa(Chamutpong et al. 2013) and also have similar breeding habits(Chamutpong et al. 2013).Great Hornbills seem to be able to adapt to nests in nest boxes better than Rhinoceros Hornbills.Although the Rhinoceros Hornbill was the first species reported to visit nest boxes, there was no report of breeding success until 2021; the number of nest occupancy was also very low.Throughout this study, the number of chicks produced by Great Hornbill using the nest boxes