Assessing feral swine damage in the western gulf region of Arkansas, Louisiana, and Texas

Feral swine (Sus scrofa) as an invasive species in the U.S. have caused severe damage to natural resources through competing for food resources with domestic livestock and destroying habitat by rooting and wallowing. Given their increased population and wide distribution in the southern U.S., feral swine control is becoming an urgent issue for both natural resource managers and landowners. We conducted a three-state mail survey to examine feral swine damage in Arkansas, Louisiana, and East Texas in the western gulf region in 2021. Our results reveal that feral swine activities like rooting and wallowing can cause severe and widespread damage to agriculture and forestry. Among the most reported and concerned damage by feral swine were crops, pastures, livestock, and loss of land value. Soybean damage was the most important agricultural crop damage in both Arkansas and Louisiana while it was hay, silage, and forage crops in East Texas. In terms of livestock damage, beef cattle and calves were the most common and severe damage in all three states. Average economic loss due to feral swine damage was estimated at $67.13/ha, $42.96/ha, $27.31/ha, and $57.54/ha for landowners in the region who owned cropland, forestland, pastureland, and multiple land types, respectively. Those findings could help both landowners and wildlife management personnel understand the negative societal impact of feral swine, and they could also facilitate the comparison between expected benefits and costs of control programs/options for feral swine in the southern states.


Introduction
Feral swine (Sus scrofa) are a particularly destructive invasive species in the United States (U.S. Department of Agriculture (USDA) 2005). Since they were first introduced to North America by the European colonists (Conover 2007), Feral swine have spread to more than 35 states with rapid geographic range expansion, especially in the last 3-4 decades (USDA Animal and Plant Health Inspection Service (APHIS) 2020). Despite their recreational benefits for hunters in certain areas (Centner and Shuman 2014), feral swine can cause severe damage to natural resources by competing for food resources with domestic livestock and destroying crops and habitats by rooting and wallowing. The nationwide damage caused by feral swine in the U.S. is estimated at around $1.5 billion per year, including damage to agricultural and forestry production and the natural environment (Wild Pig Info 2013).
Feral swine have a high reproductive capacity and early maturation, which makes their population eradication difficult once established. As reported in Higginbotham (2013) and Wild Pig Info (2013), feral swine generally have one or two litters per year with an average of five or six pigs per litter. In addition, they are very intelligent and adaptable and feed on various plants and animals across different geographic regions (Barrios-Garcia and Ballari 2012). Feral swine also carry many diseases, which could infect wildlife, livestock, and humans (Hartin et al. 2007;Conover and Vail 2007;Meng et al. 2009;Brown et al. 2018;Fredriksson-Ahomaa 2019). Given increased populations and wide distribution in the United States as well as the costly damage they inflict, controlling feral swine is becoming an urgent issue for both natural resource managers and landowners.
Feral swine travel in sounders, which is particularly problematic to landowners because this group activity can cause severe damage to agricultural crops, timberland, and pastures (Graves 1984;Seward et al. 2004;West et al. 2009). For example, the consumption of crops is a typical swine damage to agricultural producers (Bevins et al. 2014); common swine damage to forest landowners include girdling trees through rubbing, damaging the lateral roots by rooting and chewing, and removing the bark of trees by tusking (Mayer 2009). In the southern U.S., depredating newly planted tree seedlings has become a major concern of feral swine damage, including plantations of both pines (Fern et al. 2020) and hardwood species such as cherry bark oak (Quercus pagodaefolia), swamp chestnut oak (Q. michauxii), water hickory (Carya aquatica), and swamp tupelo (Nyssa sylvatica var. biflora) in a wetland restoration area in South Carolina (Mayer et al. 2000). In general, rooting is the most widespread and observed damage caused by feral swine considering that this is their primary method of searching for food (Mayer 2009).
In addition, feral swine can damage pastures, lawns, and fences (Adams et al. 2005) and even attack farmers, golfers, hikers, and picnickers (USDA APHIS 2020). Feral swine can also adversely affect wildlife, including predation of ground-nesting birds (e.g., quail, Rollins and Carroll 2001), threatening nesting success of rare birds (e.g., lesser prairie chicken (Tympanuchus pallidicinctus) in Texas), and reducing the population of game species such as deer (Mapston 2007). Other damage caused by feral swine on ecosystems is decreasing water quality and aquatic biota (Kaller et al. 2007). Overall, feral swine can cause a variety of damag to natural ecosystems including habitat degradation and predation on and competition with native species.
Several studies have attempted to estimate ferel swine damage at the state level. For example, damage in Texas was estimated at approximately $52 million annually excluding damage to urban and suburban areas (Higginbotham 2013); crop damage was valued at $75 million in 2009 in Alabama (Shi et al. 2009); the total economic impact to the Louisiana agricultural sector was estimated at $75 million (Tanger et al. 2015); in Georgia, damage to agriculture and properties was assessed at $81 million (Mengak 2012;South Carolina Wild Hog Task Force 2015); in Tennessee, damage was estimated at $26 million in 2015 (Poudyal et al. 2016). These studies focus on different scopes and contexts of economic damage using different estimation approaches, hindering comparisons across states.
With the increasing feral swine population and their broader impacts beyond agriculture, more data and studies are deemed necessary for uncovering the types and importance of damage to various private landowners. Such knowledge could serve as a useful baseline to compare with future assessments of damage caused by feral swine and prioritize future control strategies and resource allocations. Thus, the specific objectives of this study were to (1) Examine important feral swine damage to private landowners in three states (Arkansas, Louisianan, and East Texas) in the western gulf region of the United States, (2) Explore specific damage to agriculture (crop and livestock), forestry, and wildlife, and (3) Compare and aggregate the damage across the three states and different land uses. The findings of this study can help policymakers and the general public understand the negative societal and economic impacts of feral swine and facilitate wildlife management personnel and landowners to develop and implement more effective methods for controlling feral swine.

Study area
There are over six million feral swine in the United States (Mayer 2014), discovered in 47 states (Mayer 2014;USDA-APHIS 2020). Most of the swine population was established in the southern states (Gipson et al. 1998). They were first introduced to Texas by early Spanish explorers over 300 years ago (Taylor 2021) as the source of cured meat and lard for settlers. Currently, there are more than1.5 million feral swine in Texas (Taylor 2021). Feral swine have also been discovered in all 75 counties in Arkansas with an estimated population of roughly 200,000 (https:// www. agric ulture. arkan sas. gov/ arkan sas-depar tmentof-agric ulture-servi ces/ feral-hog/). Likewise, Feral swine are found in all 64 parishes in Louisiana, with an estimated population of 700,000 (https:// www. wlf. louis iana. gov/ page/ feral-hogs). The fast spread of feral swine has been observed in the U.S. from 1982 to 2020 (Fig. 1), becoming an invasive species of major concern for many states and counties, especially the in the U.S. South and West.
This study focused on all counties in Arkansas (AR), Louisiana (LA), and a subset of 44 counties in eastern Texas (ETX). Although all counties in AR and LA were included, only a subset of counties in Texas were considered. This was largely because these counties in AR, LA, and ETX are

Data collection and analysis
To better understand the feral swine damage to private landowners, we conducted a mail survey in AR, LA, and ETX in 2021. The survey instrument was developed after a thorough review of the literature regarding private landowners' perspectives on feral swine. Questions regarding feral swine activities and damage were adopted from similar studies conducted elsewhere (e.g., Caplenor et al. 2017) and modified to fit the context in our three-state study area. The survey questionnaire was reviewed and approved by the University of Arkansas at Monticello's Institutional Review Board (IRB# FNRf-01). The targeted private landowners were those who own at least 30 acres of rural land and the associated mailing addresses were purchased from Dynata Inc. From the population of landowners, 4500 landowners were randomly selected, including 2000 in Arkansas, 1000 in East Texas, and 1500 in Louisiana, respectively. A survey questionnaire was mailed to each of these selected landowners, and the survey was administered according to the Dillman Tailored Design Method (Dillman 2000). A packet including a cover letter, 10-page questionnaire, electronic consent letter, and pre-paid business reply envelope was mailed to the selected landowners. A follow-up reminder postcard was then mailed to all of them two weeks later. The landowners were assured of the confidentiality and privacy of their information and their voluntary participation in this survey. In the questionnaire, participants were asked about their land use types, including cropland, forestland/timberland, pastures, grassland/lawns, and their mix (any combination of different land uses) (Table 1). Questions about what kind and amount (in monetary value) of damage by feral swine found on their land during the past five years and how important that damage was to them were also included in the survey (Table 2). Two major feral swine activities and fifteen damage types were listed in the questionnaire. The activities included rooting/grubbing and wallowing and the damage types included the damage to cropland (non-timber), pastures, food plots, streams or ponds, equipment (e.g., irrigation pipes), landscape (e.g., house yards, gardens), fences, and stored commodities; consumption of grains/hay; loss of land value; disease transfer to pets, livestock, or human; loss of timber value; and injuries to pets and livestock. For those landowners who had suffered crop and livestock losses, they were asked to list the specific types of crops (Table 3) and livestock (Table 4) damaged by feral swine. Respondents were asked to indicate all types of damage on their properties in the past five years and the importance (1 = extremely important, 5 = not important at all) of the damage from their perspective.
Descriptive statistics of key variables including land use, damage type, the importance of damage to landowners, and values of damage were calculated. Analysis of variance (ANOVA) was employed to compare the damage types and the importance and values of damage reported by the landowners in AR, LA, and ETX. In addition, 95% confidence intervals of damage values for different land use types in the three states were computed. All statistical test results were evaluated using a significant level of 95% (p ≤ 0.05).

Characteristics of survey respondents
Out of 4500 questionnaire packets sent to landowners, 285 landowners in AR, 175 in LA, and 86 in ETX were determined ineligible due to undeliverable addresses, death, and other reasons, and they were removed from the study. As a result, the sample size of eligible survey participants was 1,715 for AR, 1,325 for LA, and 914 for ETX, respectively. We received 361 valid survey responses from AR, 319 from LA, and 226 from ETX, yielding a response rate of 21.05%, 24.08%, and 24.73% for AR, LA, and ETX, respectively, and 22.91% for the three states combined. A non-response bias check was conducted by comparing the key demographic characteristics including age and gender between early (the first 10%) and late (the last 10%) respondents. Similarities were found between early and late respondents in all three states; for example, about a quarter of both early (25.7%) and late (24.8%) respondents in Arkansas indicated that their age were between 55 and 64.
Of the 361 landowners who responded in AR, 84.5% were male and 62.1% reported being 65 years or older. Respondents had a wide range of educational attainment, with 50.5% indicating high school education/GED and some college, 28.0% a bachelor's degree, and 14.0% reporting an advanced degree. About 17.0% of respondents reported an annual household income of $20,000-$50, 000, 67.6% between $50,000 and $100,000, and 12.5% more than $100,000. Regarding the ownership acreage, the median was 300 acres for all respondents and the average length of property ownership (tenure) was around 35 years.
Among the LA survey respondents, 85.9% were male and 64.5% were 65 years or older. Regarding education attainment, 27.2% of the respondents had a bachelor's degree, 15.6% had an advanced degree, and over half (50.2%) of them indicated high school education/GED and some college. Almost two-thirds (65.3%) of respondents indicated their annual household income was between $50,000 and $100,000. Those who reported an annual household income between $20000-$50000 and more than $100,000 were 16.7% and 15.4%, respectively. The median size of land owned by the survey respondents was 200 acres and the average tenure of ownership was 35 years.
The majority (83.6%) of survey respondents in East Texas were male and 59.4% of them were 65 years or older. Nearly half (44.8%) of the respondents had a high school/GED or some college education, 27.6% had a bachelor's degree, and 17.6% had an advanced degree. Over half (51.6%) of the respondents reported income between $50,000 and $100,000, 25.5% between $20,000 and $50,000, and 20.8% more than $100,000. Their median ownership size was 467 acres, and the average length of land tenure was 44 years.
When asked about their familiarity with the invasive species of feral swine prior to receiving this survey, 86.9% of the AR respondents indicated that they had some familiarity; this percentage among LA and ETX respondents was higher, 91.4% and 99.5%, respectively. The dominant majority of landowners surveyed were aware of the presence of feral swine in their areas, suggesting that feral swine are a widespread issue in the region.
Land use and feral swine damage Cropland, the combination of different land uses, and pastureland were the top three land use types for the survey respondents in all three states (Table 1). The most populous land use type was cropland for AR (46.26%) and LA (34.17) respondents and pastureland  Rooting/grubbing and wallowing are two major activities of feral swine that caused damage to crops, pastures, and other land uses and farm establishments ( Table 2). The survey respondents seemed to be more concerned about rooting/grubbing than wallowing in all three states. There was a statistically significant difference in the landowners' concern about these feral swine activities across states. For instance, landowners in ETX were more concerned about rooting/ grubbing and wallowing than those in LA and AR while landowners in LA were more concerned about rooting/grubbing than those in AR. In ETX 96.9% of respondents indicated that the rooting/grubbing damage was important while only 76.1% in LA and 68.6% in AR said so.
In terms of damage inflicted by feral swine, the major concerns of the survey respondents were consistent with the shares of land use types. That is, the biggest concern for landowners in AR and LA was damage to crops or food plots while damage to pastures was the most important to the landowners in ETX. In addition to the direct damage to cropland and pastures, the landowners were also concerned about    Table 2, the average damage importance (mean) for all swine damage types in each state and ANOVA results among the three states were summarized. In addition, Figs. 2, 3 and 4 summarized the respondents' perceptions of each damage type in AR, LA, and ETX, respectively. Potential swine damage (a total of 16 options) included rooting/grubbing, crop damage, wallows, pastures, food plots, streams/ ponds, consumption of grain/hay, loss of land value, diseases transfer, equipment damage, timber values loss, landscape damage, fence damage, pet injury, livestock injury, and stored commodities damage.
Among all those potential damage types, the most important damage to landowners was rooting/grubbing: AR (2.27), LA (1.93), and ETX (1.28); in addition, a significant difference in this damage importance was found among three states, which indicated that respondents in ETX perceived this damage type more important than LA and AR. Of all the respondents, 96.9% indicated that the rooting/grubbing damage was important in ETX while the percentages were 76.1% and 68.6% in LA and AR, respectively. The following important damages in AR were crop damage (2.48), wallows (2.61), pastures damage (2.72), food plots (2.93), and streams/ponds damage (2.93). Specifically, 61.8% of respondents in AR reported that crop damage was important and the percentage for reporting wallows and pasture damage as important were 60% and 56.5%. Over half of respondents believed that damage to food plots (50.3%) and streams/ponds (52.4%) were important. In contrast, the following swine damage in LA were food plots (2.19), crop damage (2.21), wallows (2.23), pastures (2.38), consumption of grain/hay (2.51), land value loss (2.62), diseases transfer (2.80), and damage to streams/ponds (2.88) were important. To be specific, over two-thirds of respondents in LA indicated that swine damage to food plots (72.5%), crops (70.4%), through wallowing (67.4%), to pastures (67.1%) were important. Significant differences were found between AR and LA in the consumption of grain/hay and loss of land value, suggesting that these two damage types were reported more important in LA respondents than AR respondents. 60.4% of the responding landowners   in LA believed that swine damage in the consumption of grain/hay was important, and the responded percentage for the damage of land value loss was 58.1%. Both AR and LA respondents believed that other damage types were not important as the mean score in each case exceeded neutral (neutral = 3) and above three indicates "not important" on the five-point Likert scale. Of the ETX respondents, damage to pastures (1.37) and damage type of wallows were significantly more important than in the other two states of LA and AR. Of all ETX respondents, 95.6% believed that swine damage to pastures were an important damage type while the percentage for wallows was 88.7%. In addition, other important reported damage types in ETX were crop damage (1.72), loss of land value (1.83), damage to streams/ponds (1.88), and consumption of grain/hay (1.88). The percent of respondents who reported those damage types as important were: crop damage (85.1%), loss of land value (79.1%), damage to streams/ponds (80.1%), and consumption of grain/hay (78.9%). In addition, other important damage types that were reported by the ETX respondents were landscape damage (2.07), disease transfer (2.11), fence damage (2.14), food plots (2.26), and injury to livestock (2.70). Respondents' perceptions of the damage importance of landscape, fences, and livestock injury were distinctly different from LA respondents. Specifically, 70.5% of ETX respondents indicated that swine damage to landscape (e.g., house yard, garden) was important to them; the percent for reporting disease transfer to pets, livestock, or human, damage to fences, and food plots were 67.9%, 70.4%, and 70.9%. By contrast, just less than half (46.6%) of respondents indicated that livestock injury was an important damage type. Other damage types such as equipment damage, loss of timber value, and injury to pets were not important as the mean score in each case exceeded neutral (neutral = 3), and above three indicates "not important" on the five-point Likert scale.
Damage to specific types of crops and livestock A total of 13 options (corn; grain sorghum; peanuts; wheat; cotton; oats; rice; sunflower; soybeans; hay,   silage, and forage crops; edamame; sugarcane; and timber) were included under the question of specific feral swine damage to agricultural crops (Table 3). Among all the listed agricultural crops, the most reported damage was soybeans in both AR (57.3%) and LA (44.7%), whereas hay, silage, and forage crops were reported in ETX (87.3%). The next most widely damaged crops were rice (40.0%) and corn (37.8%) in AR, corn (36.5%) and hay, silage, and forage crops (34.0%) in LA, and wheat (21.5%) and corn (12.7%) in ETX. Damage to timber was reported by 25.9% of LA respondents. There were respondents in all three states who reported damage to other crops with 22.2% in AR and relatively low percentages in LA and ETX (Table 3). Similarly, the respondents were also asked about damage to livestock including beef cattle and calves, sheep/goats, dairy cows, horses, rabbits, swine, poultry, llamas and alpacas, aquaculture, buffalo, camels, donkeys, and mules. Of all respondents in the three states, damage to beef cattle and calves was the most reported damage to livestock: AR (32.5%), LA (86.8%), and ETX (93.2%). Second was horses with 23.5% of respondents in LA, 12.8% in ETA, and 5.6% in AR. In addition, 10.3% of LA respondents also indicated damage to aquaculture. Relatively low or no respondents reported damage to other types of livestock.

Estimated values of damage
To estimate landowners' economic losses due to feral swine, a question was asked about the respondents' losses based on their recent five years' experience. Linking the reported losses with the answers received from the respondents regarding their land use type and acreage, we calculated the average damage per acre and the confidence intervals at the 95% confidence level for different land types in all three states (Table 5). In addition, Tukey's test was used to test whether any significant difference among the three states in the average loss for each land use type. As reported by the respondents, the average loss in the past five years in agricultural cropland was estimated at around $28/acre ($69.19/ha) in both AR and LA and around $25/acre ($61.78/ha) in ETX. Similarly, landowners' loss in forestland/timberland caused by feral swine was approximately $17/acre ($42.00/ha) in both AR and LA and $12/acre ($29.65/ha) in ETX, which is significantly different from that in AR and LA. By contrast, there was no remarked difference in the loss for pastureland in all three states, estimated at around $11/acre ($27.18/ha). Regarding the combination of different land types, the average loss was estimated at $22/acre ($54.36/ha) in AR, $40/acre ($98.84/ha) in LA, and $24/acre ($59.31/ha) in ETX; a significant difference was found for LA when compared to AR and ETX. Average economic loss due to feral swine damage was estimated at $67.13/ha, $42.96/ha, $27.31/ha, and 57.54/ha for landowners in the region who owned cropland, forestland, pastureland, and multiple land types, respectively.

Discussion
Feral swine are a destructive exotic species in the United States and have caused severe damage to landowners particularly in the southern, southwestern, and western coastal states in the U.S. Considering that feral swine are invading landscapes across state boundaries with an increasing population in the western gulf region, understanding the types of damage and the importance of those damages to private landowners is increasingly necessary and important. Those findings could serve as a useful baseline to compare with future assessments of damage caused by feral swine across states based on different land uses (i.e., agricultural land, forestland/timberland, etc.) and guide the development and deployment of policy and management practices for mitigating feral swine damage. This study examined the importance feral swine damage types in the western gulf region of Arkansas, Louisiana, and East Texas to private landowners by conducting a regionwide survey. Of all the landowners surveyed, the most important feral swine damage activities were rooting/grubbing and wallowing. This observation is not surprising considering that rooting/ grubbing is the major and primary food-searching method, and in doing so causes serious damage to natural resources (Mayer 2009;Bevins et al. 2014). Feral swine have caused widespread damage to landowners in Arkansas, Louisiana, and East Texas. The types of damage are correlated with land use types in each state. Specifically, agricultural cropland was the most populous land use type reported by landowners in both AR and LA and the most important damage was to crops and food plots (Table 2). Bevins et al. (2014) reported that consumption of crops was a very common damage to crop producers/landowners. This problem is exacerbated because large sounders can cause grievous damage to agriculture crops (Seward et al. 2004;West et al. 2009). In addition, they can damage crops through their behaviors of rooting, trampling, and wallowing (Mayer 2009). By contrast, the most populous land use type was pastureland in ETX where the damage to pastures was reported to be the most important. Feral swine can damage pastures by killing desired plant species and turning over sod and pastures through rooting to expose the tender roots of plants, grubs, and invertebrates, which ultimately destroys the pasture (Adams et al. 2005).
In terms of feral swine damage to specific agricultural crops, our survey results indicated that the most reported damage by landowners was to corn, soybean, rice, hay, silage, and forage crops, as well as wheat in AR, LA, and ETX. This finding echoes the report of USDA APHIS (2020) which indicated that commonly targeted field crops by feral swine mainly include corn, grain sorghum, soybeans, wheat, oats, peanuts, and rice as swine will eat almost any crop they can find. Pimentel et al. (2002) found that crops can comprise over 50% of the vegetative portion of feral swine diets when they forage in agricultural lands, resulting in yield losses. Anderson et al. (2016) reported that the highest yield loss estimates were for peanut and corn production in Southeast and Texas. Mckee et al. (2020) also reported that yield loss to hay was 6.59% in Texas and 6.03% in Louisiana due to feral swine damages. In terms of damages to livestock, beef cattle and calves were the most reported damage in all three states, which is in line with the finding of Mapston (2007), Anderson et al. (2019), and USDA APHIS (2020), as feral swine are capable of killing young calves and vulnerable adult animals during their birthing process. Moreover, feral swine can eat or contaminate livestock feed, mineral supplements, and/or water sources (Kaller et al. 2007).
Landowner losses due to feral swine damage for different land use types were assessed in this study. Our results indicated that there was no statistically significant difference in the average landowners' loss in agricultural cropland damage across the three states. The estimated loss ranged between $24/ acre to $29/acre during the past five years. However, there was a significant difference in the average loss for forestland/timberland damage among the three states: AR and LA were approximately $6/acre higher than ETX. Similarly, a significantly higher loss to combined land use type was found in LA compared to AR and ETX. A possible reason behind this could be more landowners reported this land use type in LA than in the other two states.
Given the widespread populations of feral swine, control/management efforts made by individual landowners will not be effective, this calling for regionwide coordinated efforts of both the private and public sectors. Federal and state agencies can play a leadership role in developing feral swine control/management strategies and coordinating the efforts of different stakeholders. In 2018, the Farm Bill established the Feral Swine Eradication and Control Pilot Program (FSCP) under the USDA. This program costs $75 million over the life of the bill (2019-2023) and since its enactment, many states have initiated their feral swine pilot projects. For example, there was the Feral Hog Eradication Task Force (FHETF) in AR, the Louisiana Feral Hog Management Advisory Task Force (FHMATF) in LA, and the Central Texas Feral Hog Task Force (CTFHTF) in TX. Those task forces are playing important roles in connecting different stakeholders such as private landowners, companies, and professional natural resource managers to guide feral swine management and control. Controlling invasive species like feral swine requires not only substantial funding but also cooperation from many private landowners and natural resource managers across the landscape (Caplenor et al. 2017;Watkins et al. 2019). Therefore, the importance of landowners' perceptions, attitudes, and behaviors when making their feral swine management decisions cannot be overstated. Effective control of feral swine populations also entails educating the public and offering technical support to landowners, so effective outreach and education programs are also needed in this regard.

Conclusions
This study identified important damage types of feral swine and corresponding economic loss to private landowners for different land use types across the three states in the western gulf region of the U.S.: Arkansas, Louisiana, and East Texas. The results from this multi-state study provide a broader and deeper understanding of landowners' perception and assessment of feral swine damage on their rural lands, which is of value for developing and implementing regionwide control measures. The important damage caused by feral swine activity reported by landowners in all three states included rooting/grubbing and wallowing, followed by damage to crops, pastures, food plots, consumption of grains/hay, and streams/ponds. Specifically, the most reported feral swine damage to agricultural crops was soybeans in both AR and LA whereas it was hay, silage, and forage crops in ETX. In terms of damage to livestock, the most reported damage was to beef cattle and calves in all three states. Landowners' losses to feral swine damage were estimated for different land use types in all three states. For agricultural cropland, landowners' average loss in the past five years was assessed at $28/acre ($69.19/ha) in both AR and LA and approximately $25/acre ($61.78/ha) in ETX. In terms of forestland/ timberland, the losses were estimated at $17/acre ($42.00/ha) in both AR and LA while it was $12/ acre ($29.65/ha) in ETX. By contrast, the reported damage to pastureland was very similar in all three states with an estimation of $11/acre ($27.18/ha). For the damage loss to the combination of different land types, LA had a higher estimation ($40/acre ($98.84/ ha)) than both AR ($22/acre ($54.36/ha)) and ETX ($24/acre ($59.31/ha)). Those results provide important insights for understanding feral swine damage to landowners in the region and could be used to inform future incentive-based policies for encouraging landowners to actively manage/control this invasive pest.
To be noted, this study can be extended on several fronts. First, this multi-state study should motivate additional efforts to assess feral swine damage in other regions. Thus, the results can be compared across other states and regions, contributing to the development and deployment of coordinated control efforts. Second, more work needs to be done to examine landowners' attitudes toward controlling/managing feral swine on their properties and the associated costs, so an economic tradeoff analysis could be further examined. Third, incentive-based policy instruments such as tax reductions and cost-sharing need to be explored so that they can be appropriately designed and implemented to encourage more landowners to take coordinated and collective actions to manage/control feral swine on their properties.
[containing information that could compromise research participant privacy/consent] but are available from the corresponding author on reasonable request.

Conflicts of interest
The authors declare no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.