Granular Bait Applications for Management of Rangeland Grasshoppers Using a Remotely Piloted Aerial Application System

Abstract

Rangeland grasshoppers are an endemic species that play an essential role in the rangeland ecosystem but can cause severe economic damage when populations reach outbreak levels. Remotely piloted aerial application systems (RPAASs) offer an alternative method to carry out aerial insecticide applications in relatively small areas. The objective of this study was to investigate the efficacy of a granular bait, 2% Sevin (with the active ingredient carbaryl), applied by an RPAAS. The bait was applied on four replicated 4.05-hectare (10-acre) plots at a rate of 2.27 kg/ha (5 lbs/acre) with an RPAAS on a private ranch in New Mexico. Applications resulted in a normalized population reduction of 70.32% ± 16.54% standard error of the mean (SEM) of bait-susceptible species. Although some of the observed reduction in population may be attributed to aging, the net effect was most likely due to the ingestion of bait based on field observations of rapid mortality after ingestion and other factors, like past experience with carbaryl bait treatments on grasshoppers. Plots required at least two flights due to the Federal Aviation Administration’s (FAA) maximum takeoff weight requirement for small RPAASs. Combined, these results indicate that RPAASs can provide treatment capabilities in relatively small areas, i.e., population hotspots, preferably before outbreak levels are reached.

1. Introduction

The western United States contains ample rangeland habitats, which are home to more than 400 species of native grasshoppers (Orthoptera: Acrididae), about two dozen of which are considered pests by land managers [1,2]. These species are classified as such due to periodic population outbreaks during which they tend to feed voraciously on rangeland forage, as well as adjacent crops [2]. The 3rd nymphal instar stage is approximately when significant economic impact begins, with most species averaging five instars before becoming adults [3]. Outbreaks tend to start with “population hotspots”, which are localized areas of high population density [4] that are primarily composed of nymphs that have not yet dispersed away from hatching sites.

As a result of these pest grasshopper outbreaks, more than 20% of rangeland forage is consumed annually at an estimated loss of USD 1.25 billion [3,5], with further economic losses coming from adjacent crop destruction [6]. Therefore, land managers have developed management methods for suppressing populations that utilize insecticides (both in liquid and bait forms); in particular, the United States Department of Agriculture (USDA) Animal and Plant Health Inspection Service (APHIS) Rangeland Grasshopper and Mormon Cricket Suppression Program. Currently, fixed-wing aircraft are the default application method for insecticide treatments for three reasons: (1) historically being the only aerial option available, (2) the ability to cover large areas in a single flight, and (3) the ability to treat said areas in a relatively short period of time.

More recently, though, the use of remotely piloted aerial (or aircraft) application systems (RPAASs) for agricultural purposes is receiving increased attention, especially due to specialized abilities not available to fixed-wing aircraft that allow RPAASs to occupy unique agricultural niches [7,8]. Depending on the type of RPAAS, such abilities include being able to be deployed more rapidly, being able to be controlled remotely, flying autonomously using preprogrammed flight paths, flying at low altitudes with variable ground speed and at an array of heights, hovering in place, enabling greater precision and safety in terms of applications [9,10], being relatively safer for a pilot, and requiring less specialized training.

RPAASs have yet to be incorporated into U.S. grasshopper management programs, mainly due to a lack of studies demonstrating their utility for treating population hotspots. However, recently, Martin et al. [11] reported that a liquid insecticide (Sevin XLR PLUS) applied with an RPAAS platform significantly suppressed rangeland grasshopper populations on a New Mexico ranch and the results were comparable to those achieved via fixed-wing aircraft. Each 4.05-hectare (10-acre) hotspot test area was treated in a single flight lasting approximately 5 min. These applications, in comparison to fixed-wing aircraft, were relatively rapid, more targeted, and potentially more cost-effective [11]. This success suggested similar results could be obtained from assessing the utility of deploying bait-based insecticides for grasshopper management via RPAASs, a utility that had not been assessed until this present study. Such an ability would be welcomed by land managers because baits are efficacious against numerous pest grasshopper species [12] and certain management situations require the use of baits over liquids [13].

The most common type of granular spreaders for UAS are centrifugal systems [14,15,16,17]. Alternative systems include conveyor- [18], screw-, and pneumatic- [19,20]-based systems. Bait applications using these types of spreaders have been conducted for mammals [21,22,23], but, so far, there are relatively few for insects [24,25]. In this study, we seek to apply a centrifugal granular spreader to apply baits for grasshopper management.

The objective of this study was to investigate the utility of using RPAASs to treat grasshopper pest population hotspots, using a bait-based insecticide, 2% Sevin Bait. This bait contains the active ingredient carbaryl, an acetylcholinesterase inhibitor, which is effective against all grasshopper life stages and is used in the APHIS program.

2. Materials and Methods

2.1. RPAAS

The RPAAS used for the study was a six-rotor Precision Vision 35 (Leading Edge Aerial Technologies, New Smyrna Beach, FL, USA). It was equipped with a centrifugal granular spreader (UGS-4G, CFR Innovations, Chicoutimi, QC, Canada) for applying the bait. The spreader system was set up with a No. 28 orifice at 5.3 V to achieve a total system flow rate of 2.29 kg/min (5.04 lbs/min). To achieve an application rate of 2.27 kg/ha (5.0 lbs/acre) over 4.05 ha (10 acres), the RPAAS was flown at 4.3 m/s (9.62 mph) at an altitude of 6.1 m (20 ft). This altitude was selected to provide adequate clearance over obstructions in the landscape consistent with label requirements. Treatments with 2% Sevin Bait (22.68 kg (50 lbs)/plot) required two to three flights to cover each 4.05 ha (10-acre) plot depending on the distance ferried before reaching the plot, with the hopper refilled halfway through each plot’s treatment. Multiple flights were needed because current Federal Aviation Administration (FAA) regulations restrict the total weight of an RPAAS, including load, to 24.95 kg (55 lbs). The plots were approximately square at 201 m by 201 m (660 ft × 660 ft), but this fluctuated with other boundaries, such as fence lines and high voltage electrical wires. The label for 2% Sevin Bait allows the APHIS program and allied state programs to make aerial treatments up to 4.54 kg/ha (10 lbs/acre). However, 2.27 kg/ha (5 lbs/acre) was chosen here because this rate is recommended by APHIS-Plant Protection and Quarantine (PPQ)-Science & Technology (S&T) as it has been demonstrated to have comparable efficacy at a lower cost when used to suppress populations of species that are susceptible to bait [26]. Due to this, using 2% Sevin Bait at a rate of 2.27 kg/ha (5 lbs/acre) is increasingly common in the APHIS program.

2.2. Bait Swath Measurement

Twenty-five two-gallon buckets were placed in a line perpendicular to the flight path, approximately 0.91 m (3 ft) apart, to capture the bait distribution (Figure 1). Ten sequential passes over the line were made at a 6.1 m (20 ft) altitude, each following the same direction of travel at 5.6 m/s (12.5 mph). Buckets of collected bait were emptied into weighing trays and the mass of the bait was measured with a balance (Mettler PB153-S, Columbus, OH, USA). Bait application rate was plotted over distance to determine effective swath based on APHIS program standards, which are when the cumulative average drops below 75% of its initial value.

2.3. Field Trials

A randomized plot design with two treatments and eight plots was established on rangeland habitat near Estancia, New Mexico, with each rectangular plot measuring approximately 4.05 ha (10 acres) (Figure 2). This region was selected based on reported grasshopper densities meeting economic thresholds. Plot dimensions were created based on linear measurements from ground instruments resulting in a slight deviation from rectangular plots due to changes in terrain. The two treatments consisted of the following: (1) untreated control and (2) 2% Sevin Bait (by Wilbur-Ellis, the active ingredient is 2% carbaryl). Meteorological data were collected from a fixed weather station (Vantage Pro2, Davis Instruments, Hayward, CA, USA) located within 1.93 km (1.2 miles) of the plots.

Grasshopper population density estimation methods were performed on day 0 (the day of the treatments), 3, 7, 10, and 14 using both visual estimation and sweep net sampling following the protocols in Martin et al. [11]. Collected specimens were identified to species and life stages. Untreated control plots had their population densities estimated each time the treated plots were assessed except for plot D (Figure 2). Due to time constraints caused by the onset of dusk during the bait study, plot D was treated the day after the other three plots, so population estimations for both sets of untreated control plots were averaged together. The sweep net samples were the basis for calculating the efficacy statistics because of the variation in the degree of bait acceptance by species (

Table 1. Overview of species collected from all experimental plots, organized by their expected susceptibility to bait [12] and then in alpha-order by genus, then species name [27].

Expected Susceptibility to Bait	Species
Highly Sensitive (70–85%)	Ageneotettix deorum (Scudder, 1876)
Highly Sensitive (70–85%)	Aulocara elliotti (Thomas, 1870)
Highly Sensitive (70–85%)	Hadrotettix trifasciatus (Say, 1825)
Highly Sensitive (70–85%)	Melanoplus foedus Scudder, 1878
Highly Sensitive (70–85%)	Melanoplus sanguinipes (Fabricius, 1798)
Sensitive (55–70%)	Melanoplus occidentalis (Thomas, 1872)
Highly Vulnerable (42–72%)	Eritettix simplex (Scudder, 1869)
Highly Vulnerable (42–72%)	Psoloessa delicatula (Scudder, 1876)
Vulnerable (12–42%)	Aulocara femoratum Scudder, 1899
Nonsusceptible (15–30%)	Amphitornus coloradus (Thomas, 1873)
Nonsusceptible (15–30%)	Cordillacris crenulata (Bruner, 1889)
Nonsusceptible (15–30%)	Cordillacris occipitalis (Thomas, 1873)
Nonsusceptible (15–30%)	Metator pardalinus (Saussure, 1884)
Highly Nonsusceptible (0–15%)	Phlibostroma quadrimaculatum (Thomas, 1871)
Unknown	Heliaula rufa (Scudder, 1899)
Unknown	Melanoplus regalis (Dodge, 1876)
Unknown	Xanthippus corallipes (Haldeman, 1852)

).

Two types of population density reduction results were reported using the sweep net sample data: standard average percent population density reduction and the standard error of the mean (SEM), and normalized population reduction (P_R), which factors in population density reductions in both treated and untreated plots caused by the treatment and natural causes using the equation:

$$P_R=1-{\displaystyle \frac{T_a{C}_b}{T_b{C}_a}}$$

(1)

where T_a is the final total population density of the treated plots, T_b is the initial total population of the treated plots, C_a is the final total of the untreated control plots, and C_b is the initial total of the untreated control plots [28]. SEM was also calculated for this value based on independent and random error propagation [29] using the equation:

$$\Delta P_R=\sqrt{{\left[{\displaystyle \frac{C_b}{T_b{C}_a}}\Delta T_a\right]}^2+{\left[{\displaystyle \frac{T_a{C}_b}{T_b^2{C}_a}}\Delta T_b\right]}^2+{\left[{\displaystyle \frac{T_a{C}_b}{T_b{C}_a^2}}\Delta C_a\right]}^2+{\left[{\displaystyle \frac{T_a}{T_b{C}_a}}\Delta C_b\right]}^2}$$

(2)

A repeated measures analysis of variance (ANOVA) was used to measure the following: (1) the significance of the treatment, (2) time after treatment, (3) species of grasshopper, and the interactions of these three variables. Statistical analyses were performed using R statistical packages (version 3.6.0) [30].

3. Results

3.1. Bait Swath Measurement

The measured effective swath width of the 2% Sevin Bait treatment was 14.4 m (47.2 ft) while the average application rate during the calibration was 5.27 kg/ha (4.70 lbs/acre) (Figure 3). The speed of the RPAAS was correspondingly decreased to ensure that the application rate of the label was delivered.

3.2. Bioassay

Treatments with 2% Sevin Bait were conducted on 4–5 June 2020. Meteorological data for all treatments are contained in

Table 2. Average meteorological data collected during treatments on 4–5 June 2020.

Plot	Wind Direction	Wind Velocity (ms−1/mph)	Temperature (°C/°F)	Relative Humidity (%)
A	SSW	0.16/0.36	27.0/80.6	33.1
B	S	1.41/3.15	30.5/86.9	26.0
C	NE	2.32/5.19	26.9/80.4	30.6
D	NW	1.23/2.75	28.6/83.5	29.1

.

In terms of percent population density reduction of the bait-susceptible species (from vulnerable to highly sensitive, see Table 1), the treated plots averaged 63.45% ± 14.38% SEM by day 14 while the untreated plots averaged 39.25% ± 9.71% SEM. Normalized population reduction for these same species was 70.32% ± 16.54% SEM. Percent population density reduction specific to each bait-susceptible species by day 14 was 95.6–100% for Highly Sensitive, 93.33% for Sensitive, 24–50% for Highly Vulnerable, and 86.84% for Vulnerable. An initial analysis of the visual estimations of population density following treatment with 2% Sevin Bait showed no significant difference due to the treatment.

Further analysis based on the species composition in the sweep net samples showed a significant effect from species, time, and treatment, as well as significant interaction effects from species and treatment, plus species and time (

Table 3. Results of repeated measures ANOVA of 2% Sevin Bait treatments of grasshoppers across 14 days. Grey shading indicates statistically significant factor(s) with p < 0.05.

Effect	df	Sum of Squares	Mean Square	F	p
Treatment	1	53	53.11	10.717	0.001
Time	4	177	44.33	8.944	<0.001
Species	16	2014	125.86	25.397	<0.001
Treatment:Time	4	7	1.63	0.330	0.858
Treatment:Species	16	496	30.98	6.251	<0.001
Time:Species	64	509	7.95	1.605	0.002
Treatment:Time:Species	64	71	1.11	0.233	1.000
Residuals	3910	19,378	4.96

, Figure 4, Figure 5 and Figure 6). Based on the significance of these factors, the most likely causes of variation in the observed population density reduction of the plots were due to bait susceptibility (species, treatment, treatment:species) and aging (time, time:species). The aging of the grasshopper population can be seen clearly in Figure 7, which shows an example of a gradual reduction in earlier life stages and an increase in adult presence.

4. Discussion

The primary objective of our study was to evaluate the efficacy of treating a grasshopper population with an RPAAS-delivered bait-based insecticide and then compare the results informally with those from fixed-wing aircraft treatments based on our collective experience and knowledge. The significance of the treatment and interaction between treatment and species indicates that the 2% Sevin Bait was effective and specific, ranging from 24–100% population density reduction depending on a species’ relative susceptibility to bait (Table 1 and Figure 4, Figure 5 and Figure 6). These findings and the percent population density reduction of 63.45% + 14.38 SEM for all bait-susceptible species combined are relatively similar to the results of a past APHIS program carbaryl bait study in South Dakota that used fixed-wing aircraft to apply 2% Sevin Bait using the same rate as in this study, 2.27 kg/ha (5 lbs/acre) [12]. However, direct comparisons of these two studies are confounded by multiple variables, such as population density estimation method (visual-only) and days assessed (only at 0, 2, 4, and 7 days), differing species complexes, relative susceptibility to bait, and environmental conditions. Furthermore, the combined percent population density reduction is less than that seen previously in studies using carbaryl sprays [11]. This difference is expected due to the specific mode of entry (feeding) for the granular formulation, which limits the effectiveness of granular baits due to the susceptibility concept (Table 1) [31].

However, it should be noted for our present study that the additional significant factors of time, species, and the interaction between time and species indicate that aging was a confounding cause of mortality during the experiment (Table 3). This was further supported when data were analyzed on a species and life cycle basis (Figure 7). Regardless, field observations of bait ingestion by grasshoppers that resulted in rapid mortality, past author experience with carbaryl bait grasshopper experiments, and the species complex strongly suggest that the observed population density reductions were primarily a result of the bait treatments. Also, it should be noted that sweep-sampling specimens in the untreated control plots across two consecutive days (caused by the need to treat plot D on a different day) may have contributed to a lower normalized population reduction value as a possible result of oversampling.

The application rate of 2.27 kg/ha (5 lbs/acre) resulted in the RPAAS being flown close to maximum takeoff weight even after splitting the plot into at least two loads to satisfy current Federal Aviation Administration (FAA) restrictions that limit the maximum takeoff weight of small unmanned aircraft systems (which the RPAAS used in this study falls under) to less than 25 kg (55 lbs). Additional ferry time to the second half of the plot resulted in one plot requiring a third flight. These additional logistical hurdles, in addition to various issues loading the bait spreader (e.g., the need to wear an OSHA-approved respirator and difficulties in physically moving the bait into the hopper), make granular applications via RPAASs currently less attractive compared to liquid applications reported earlier [11]. Additionally, using bait tends to not reduce population densities as well as liquid insecticides for two main reasons, the first being that baits only induce mortality via ingestion whereas, depending on the formulation, sprays can act both topically and orally [13,31]. Second, as evident from Table 1, baits act more selectively on grasshopper populations (especially depending on the species complex) than sprays [12,13,31].

Still, despite these various issues, baits do have several advantages over liquid insecticides. For example, because baits only work when eaten orally, they have an advantage in potentially affecting fewer non-target arthropods and mostly ground-dwelling ones since the bait tends to fall through vegetation [13,31]. Furthermore, bait is less susceptible to drifting off-target, which is caused by wind, temperature, and the relative weight of the particles being applied [13].

In the near future, based on the encouraging results of this study and the previous one that used liquid insecticides [11], USDA-APHIS plans to conduct several more RPAAS-focused studies on pest grasshopper species. The first of these plans to evaluate the efficacy of a different liquid insecticide on larger plots (40 acres vs. 10) compared to our previous study [11] while the second will be similar, but focus on treating at night, which has not yet been assessed for insecticide treatments made via RPAASs. There are two potential advantages to the latter concept, the first of which is that wind speeds are often lower, so spray drifting would be more limited and the other is that grasshoppers (and other insects) often go into a dormant state of inactivity at night, so they could possibly be targeted with a treatment more precisely.

5. Conclusions

Overall, the efficacy results suggest that using an RPAAS to apply bait for treating grasshopper outbreaks was effective compared to a past carbaryl bait experiment using fixed-wing aircraft [12], but more studies and innovative development are needed to improve upon existing performance, mainly related to loading the bait spreader and enabling larger payloads to be carried. Despite these issues, using RPAASs to deploy bait treatments for suppressing grasshopper outbreaks is still enticing for the reasons noted in the text and, as with liquid insecticides [11], there are myriad potential advantages compared to fixed-wing aircraft applications including relative portability, more targeted applications, the potential to treat hotspots more rapidly, and possible cost-savings depending on the hotspot’s size. Additional work may consider time of day and alternative treatment products, such as bioinsecticides [32].