Application of precision agriculture technology to define and manage nematodes and diseases of soybean. John Rupe, Terry Kirkpatrick, Sreekala Bajwa, and Rick Cartwright, University of Arkansas and the Arkansas Cooperative Extension Service
Report Overview:
Preliminary results:
- Pine Tree field: harvest cyst densities of soybean cyst nematode (SCN) and yields were significantly affected by treatment. The highest cyst densities and the lowest yields were with the SCN-susceptible cultivar, Hutcheson, followed by Hutecheson plus the nematicide, aldicarb. The highest yields and lowest cyst densities were with the SCN resistant cultivar Anand, and were not detected with the resistant cultivar Anand.
- PrairieCounty field: average harvest cyst counts and yield did not differ between Hutcheson and Anand.
- Remote images of the fields revealed previous cropping history and differences in soil texture
Project Objectives:
- To detect the onset and map the development of SCN in soybean fields using aerial remote sensing
- To determine the practicality and effectiveness of applying site-
specific control measures for SCN
Importance of Research Area:
The soybean cyst nematode (SCN) is one of the most damaging pests of soybean in the US and is present in most soybean growing regions. When damage is severe, plants are stunted or killed and yield losses approach 100%. However, damage from the nematode is often symptomless. Growers may not know that they have an SCN problem, or if they know they have the nematode they may think that it is not reducing yields, yet yields can be reduced by as much as 10 to 15 bu/A. As a result, growers often do not take steps to control SCN.
Control of SCN is primarily by rotation, cultivar resistance, or a combination of both. In the south, the use of resistance is complicated because changes in races of the nematode can overcome the resistant cultivars used. As a result, growers may have to change the source of resistance to achieve effective control. Another control option is the use of nematicides, however it is thought to be too expensive to be of practical use. This may not be true. Like most nematodes, SCN is not evenly distributed across the field, but appears in clumps. This may be due to the slow spread within the field after the nematode is introduced or to differences in soil texture and other microenvironmental conditions that may limit the development of the nematode. Because of this uneven distribution, it may be cost effective to apply control measures like cultivar resistant or nematicides only to those parts of the field with the nematode. However, to do that, we need to be able to identify the infested areas.
Detecting the nematode is difficult. As mentioned, significant damage can occur without any visible symptoms and even where there are symptoms, the damage often extends beyond those areas. Determining the presence and the severity of SCN infestation requires taking soil samples, washing the nematodes out of the soil and counting the nematodes under a microscope. This very slow process makes grid sampling a field for SCN impractical for most growers. However, the whole field may not need to be sampled. Areas of greatest nematode damage may be detected by remote sensing. Since the nematode reduces the flow of water and nutrients from the root to the leaves, changes in leaf reflectance may result. By selecting the proper wavelengths, leaf reflectance can used to detect drought stress, nutrient deficiencies and some diseases well before visible symptoms appear. While remote sensing can detect stress, it may not be able to identify the cause of that stress, but by locating the areas of stress in the field, site specific sampling can be done to determine the cause of that stress. Determining what wavelengths are most sensitive to SCN plant damage and when the best time to monitor the fields will improve the effectiveness of this technique. Alternatively, locating areas of possible nematode damage might be identified from maps of yield or soil texture. Soil texture may be important in predicting both nematode development and plant damage from nematodes. With whatever method ultimately proves best in identifying areas that may be experiencing nematode damage, the goal is to reduce the number of soil samples necessary to accurately determine where SCN is in the field. This is essential for effectively applying site-specific controls for SCN.
Materials and Methods:
Tests were conducted in a field in PrairieCounty and another field at the Pine Tree Station, Colt, AR. Treatments were replicated strips of a soybean cyst nematode (SCN) susceptible cultivar (Hutcheson) and an SCN-resistant cultivar (Anand) at the Prairie County field and replicated strips of Hutcheson, Hutcheson + nematicide (aldicarb), and Anand at the Pine Tree field. At Prairie county, the strips were 32 rows wide and at Pine tree the strips were 16 rows wide. The treatments were replicated three and five times, respectively. Rows were planted on 30 inch centers and strips ran the length of each field, about 2,000 feet. The fields were planted on 2 July and 4 June for the PrairieCounty and the Pine Tree fields, respectively.
On 15 July, soil samples were taken from both fields. The widths of the quadrates corresponded to the treatments. At Pine Tree, the quadrates were 16 rows wide and 260 feet long and at Prairie county, they were 32 rows wide and 125 feet long. There were a total of 120 and 102 quadrates, respectively. Soil from each quadrate was assayed for plant parasitic nematodes. Soil from several quadrates were combined for nutrient analysis on a 2.5 A grid.
The fields were irrigated for optimum yields at both locations. The Prairie county field was furrow irrigated while the Pine Tree field was flood irrigated. The fields were monitored through out the season for the development of any disease or other problems.
Soil samples were taken again at harvest from both fields using the same quadrates. Yields were taken with yield monitoring combines. Soil electro conductivity was mapped using Veris at the Prairie county field on 30 January 2004. The Pine Tree field has not yet been mapped. Plant heights from each quadrate were taken in September.
Aerial images of the fields were taken on 24 September at Pine Tree and 7 October at Prairie county at four wavelengths, 550, 650, 750, and 850 nm.
Preliminary Results:
Even though both fields had high levels of SCN in 2002, planting populations in 2003 were relatively low ranging from 0 to 455 and 0 to 1,591 juveniles/200cc of soil at PineTree and Prairie county, respectively (Fig. 1 and 2).
Figure 1. Planting distribution of the soybean cyst nematode (J2 larvae/200cc soil) in the Pine Tree field, 2003.
Figure 1. Planting distribution of the soybean cyst nematode (J2 larvae/200cc soil) in the PrairieCounty field, 2003.
Distribution of the nematode was uneven at both locations. At Pine Tree, most of the nematodes were in the lower half of the field. At Prairie county, most of the nematodes occurred on the east side of the field and at the bottom third. We were able to obtain a soil map of the Pine Tree station and most of the nematodes occurred on the Calloway silt loam (Fig. 3).
Figure 3. Soil map of the Pine Tree field.
The harvest SCN densities (cysts/200cc soil) at Pine tree were scattered throughout the field and reflected treatment (Fig. 4).
Figure 4. Harvest distribution of soybean cyst nematodes (cysts/200cc) at Pine Tree, 2003.
The highest cyst counts were in Hutcheson followed by Hutcheson + aldecarb (Table 1). There were no cysts found in the plots of the resistant cultivar Anand. Yields reflected nematode levels and were highest where the cyst densities were lowest (Anand) and lowest where the cyst densities were highest (Hutcheson).
Table 1. Plant heights and yields at the Pine Tree and Prairie county fields, 2003.
Treatment / Plant Height (cm) / Yield (Bu/A) / SCN, planting (J2/200cc) / SCN, harvest (cysts/200cc)Pine Tree
Hutcheson / 68 / 74 b1 / 1409 / 914aHutcheson + Temik / 70 / 77ab / 909 / 216 b
Anand / 62 / 82a / 545 / 0 c
PrairieCounty
Hutcheson / 48 / 77 / 361 / 217Anand / 49 / 77 / 832 / 272
1 Number followed by the same letter not statistically different (P=0.05) by LSD within
a location.
At the Prairie county field, harvest densities of cysts were lower than at the Pine Tree field and there was no difference between cultivars (Fig. 5, Table 1). Most of the cysts occurred in the lower portion of the field.
Figure 5. Harvest distribution of soybean cyst nematodes (cysts/200cc) at PrairieCounty, 2003.
In both fields, planting populations did not reflect harvest populations of SCN. This was probably due to nematode levels being present, but below the detection threshold at planting. The relationship of yield to SCN densities will be analyzed in more detail once the harvest levels of SCN eggs and juveniles have been determined.
The remote sensing images from Pine Tree shows a lighter area at the north end of the test (Fig. 6).
AB
Fig. 6. A. Four band image aerial image of the Pine Tree field taken on 24 September 2003 at 550, 650, 750, and 850 nm of wavelength. B. Classified image using 20 classes (classes are made based on spectral reflectance – pixel based).
This corresponded to a five year wheat/soybean double crop study that was concluded in 2001. Differences between cultivars were not evident. At the Prairie county field, there were differences in the classified image that might correspond to soil texture (Fig. 7 A). The electro conductivity map (Fig. 7 B) indicates that the soil had more silt and less sand in the northern end of the field and became sandier at the lower end of the field. In the four band image, Hutcheson plots were darker than the Anand plots (Fig. 7 C).
ABC
Fig. 7. A. Classified image of Prairie county field with 20 classes. B. Electro conductivity (Veris) map of field (lower portion was flooded when assayed). C. Four band aerial image taken on 7 October 2003 at wavelengths of 550, 650, 750, and 850 nm.
This corresponded to a visible color difference between the two cultivars during reproductive growth. Also, Anand had more Cercospora leaf blight late in the season compared to Hutcheson, but not enough to affect yield.
Plant heights were greater at Pine Tree than Prairie county field reflecting the earlier planting date at Pine Tree. There were not statistically significant differences between treatments in plant height at either location.
Continuing Work:
Analysis of 2003 data:
More detailed analysis will be conducted once all of the 2003 analyses are completed. These include: harvest nematode counts, soil nutrient analysis, electro conductivity, and soil texture analysis. Soil samples for textural analysis will be selected based on the electro conductivity data collected with Veris.
2004 Research:
Field: Due to a reduction in funding, only the Pine Tree field will be used in 2004. In 2004, this field will be planted with Hutcheson only to determine the carry over effect of the treatments. The field will be sampled as in 2003, except there will be more aerial images of the field and electro conductivity will not be repeated.
Microplot: Microplots with and without the nematode will be established with Hutcheson at the experiment station in Fayetteville, AR. These will be monitored with a hyperspectral radiometer to determine the optimum wavelengths, or combinations of wavelengths to detect damage by SCN.
Project Status:
Table 2 outlines key milstones and dates work was initiated and completed to meet the milestones.
Table 2. Key milestones for the SCN project
Milestone / Initiated / CompletedSelect and plant fields / Spring 2003 / Spring 2003
Planting samples / Spring 2003 / Spring 2003
SCN analysis / Spring 2003 / Summer 2003
Soil nutrient analysis / Fall 2003
Harvest samples / Fall 2003 / Fall 2003
SCN analysis / Fall 2003
Electro conductivity / Winter 2004
Aerial Imagery / Summer 2003 / Fall 2003
Overall analysis of data / Fall 2003
1