Evaluating Commercial Cranberry Beds for Variability and Yield Using Remote Sensing Techniques

Peter V. Oudemans2, Marilyn G. Hughes1, and Larisa Pozdnykova

1Rutgers Cooperative Extension and the Grant F. Walton Center for Remote Sensing and Spatial Analysis

New Brunswick, NJ USA

2Rutgers University, Blueberry and Cranberry Research and Extension Center

Chatsworth, NJ, USA


Cranberry growers are using new technologies to decrease the impact of farming on the wetland environment, analyze crop health, and maximize and model cranberry (Vaccinium macrocarpon Ait.) yield. Extensive field sampling has been used in the past as a means of estimating potential bed yields; however, problems have been encountered due to high intra-bed spatial variability. Color-IR aerial photography available since 1987 (Ocean Spray Cooperative) and historical records of bed yields are used to identify and analyze two common diseases found in cranberry, Phytophthora root rot (PRR) and fairy ring (FR), and their effect on bed yield. Imagery was geo-referenced and imported into a GIS database for analysis. Results indicate that PRR occurs over larger areas than FR and causes more significant yield losses. Treatment appears to provide long term control and significant positive economic impact for the grower. FR appears not to be an economically important disease, and the cost of treatment exceeds the impact on crop productivity.


The large American cranberry (Vaccinium macrocarpon Ait.) is a low growing perennial vine indigenous to the sandy wetland soils found in the Pine Barrens region of New Jersey. Due to stringent wetland laws, expanding the cultivated acreage to meet an increasing demand for the crop is difficult. A viable alternative to expansion is improvement of the existing acreage. Current yields in New Jersey vary from 5400 - 54000 kg/ha on average. Within an individual bed however, yields may vary from 0 - 86000 kg/ha. Many factors influence yield, including disease, variety, nutritional status and soil type. Two diseases, Phytophthora root rot (PRR) and fairy ring (FR) are of serious concern to cranberry growers. Previous work has shown that, in New Jersey, Phytophthora cinnamomi is a nearly ubiquitous water borne fungus that is distributed through the irrigation systems . This pathogen infects and rots the plant roots resulting in reductions in plant biomass and productivity. Disease development occurs in areas with poor drainage and often results in plant death. Although acute symptoms (vine death) are easily recognized, the pathogen is also capable of causing chronic injury leading to significant yield loss. Chronic infections are often undetectable in the field. Control strategies for PRR include improvement of soil drainage, followed by fungicide use. Early detection of root rot prone areas aids in disease management and helps decrease the time required for bed establishment. Little information is available on the pathogen effects on plant growth and yield loss and there has been no research describing loss due to factors such as reduction in bud set, or reduction in fruit size. Thus, the economic impact of this disease is poorly understood. Fairy ring is caused by the fungus Psilocybe agrariella. This disease manifests itself as expanding rings, reaching up to 20m in diameter. The advancing edge of the ring typically appears as a region of enhanced growth followed by an inner region of dying vines. The internal portion of the ring is usually slow to recover and may remain stunted for several years. Treatment of this disease requires the application of fungicides to an area larger than the affected regions. Treatment usually costs $25,000 per hectare of treated area for material alone. The cost benefit of fungicide applications is uncertain and yield losses have not been determined.The objectives in this study were to 1) examine the ability to use readily available color-IR aerial photography to identify the spatial and temporal distribution of these two cranberry diseases 2) assess the impact of these two diseases on yield.


For this study, we used the available 1:12000 color-IR aerial photography obtained and archived each May from Ocean Spray, Inc. (Lakeville-Middleboro, MA). Data are collected in three bands, the green (.5-.6), red (.6-.7), and near-IR (.7-2.0. The photography is available for the time period covering 1987 - 1999 with the exception of 1988 and 1998. Ten years of photography for several cranberry farms known to be infected with PRR and FR were scanned into a digital format using a high-resolution color scanner. All of the imagery was imported into ERDASâ Imagine (Atlanta, GA) an image processing software package. Using a series of ground control points obtained in the field with a Trimble Pathfinder Pro XR GPS the imagery were rectified and re-projected into a UTM NAD83 coordinate system with a ground resolution of 0.3m. Multi-spectral techniques were used to related differences in spectral reflectance to observed differences on the field. These techniques exploit the fact that in general, healthy vegetation absorbs energy in the red part of the spectrum and reflects energy in the near-IR part of the spectrum. The ratio of the near-IR to red reflectance known as a Normalized Difference Vegetation Index (NDVI) is often used as an indicator of plant stress. Larger NDVI values indicate more green vegetation and depending on the time of the over flights, a healthier crop. The second is an unsupervised classification based on an ISODATA clustering algorithm that statistically groups pixels together into similar classes. Extensive field checking is required for assignment of physical features to the results; however, results show that this method is useful for identifying variations in vegetative cover, irrigation system features, and areas of beds impacted by insect damage and root diseases. To evaluate the relative effects of fairy ring and Phytophthora root rot beds were selected that displayed symptoms of both diseases. Once identified, the diseased areas within each bed were delineated using a technique called "heads-up" digitizing in ArcViewâ GIS (ESRI, Redlands, CA) using the imagery as a backdrop. Fairy rings are easily identified from CIR aerial photographs (Fig. 1). The area of each polygon (beds and rings) was computed using an avenue script in ArcView. The periphery of each diseased area and the bed boundaries were recorded in the field using GPS. Yield estimates were made for diseased and healthy areas and the total loss for the bed was estimated. Areas affected by Phytophthora root rot tend to be irregular and often appear as areas of high reflectance in the CIR aerial photographs. This is due to exposed sand in areas where the vines are dead. Other crop anomalies such as some root feeding insects, cause damage similar to PRR, therefore, diseased areas were verified visually in the field and confirmed by isolation of the fungus from the root and stem tissues. To estimate yield effects by both diseases, in-situ sampling points were places both within diseased areas as well as in adjacent healthy areas and yields were estimated by counting the number of vegetative and flowering shoots (called uprights), as well as berry number inside of a 232 cm2 area.

Figure 1. A color-infrared aerial photograph represented in gray scale, showing a
cranberry bed exhibiting symptoms of fairy ring disease.


3.1 Fairy Ring

Results from the four adjacent cranberry beds of the cultivar Ben Lear show that over ten years, fairy rings within a bed generally increased in both size and number (Fig. 2). Some rings appear to merge creating a larger ring, while other rings only appear for one or two growing seasons and then disappear altogether. In general, the majority of rings were perennial and increased in area by approximately 20% each year. The rings ranged in size from 2 - 20m in diameter. The reductions in cranberry yield caused by fairy ring disease are summarized in Table 1. Results indicate that on average fairy ring causes approximately a 52% reduction in cranberry yield.

Figure 2. Digitized fairy rings from a cranberry single bed shown for
four years from 1987-1997.

Table 1. Effect of fairy ring on yield and yield components of cranberry

Diseased1 Healthy2
Upright density *** 43.13±36.20 101.00 ±71.15
Flowering uprights (%) ** 37.6±13.8 47.8±15.3
Yield (estimated) *** 122.9-492.6 420.0-784.0

*** indicates values for diseased and healthy vines are significantly different at the 0.001 level.
1Samples were taken from areas visually impacted by the disease
2Samples were taken from unaffected areas adjacent to the diseased areas

3.2 Phytophthora Root Rot

Phytophthora root rot occurred in areas ranging from 5 - 5000 m2. The areas with acute symptoms provide no yield and recently killed vines generally lose their fruit to fungal rots. Results from the unsupervised classification of the color-IR imagery show that areas of acute infection tend to be surrounded by areas of chronic infection. Whereas, areas of acute infection are clearly visible both in the photography and on the ground, chronic infections are often invisible to the eye (Fig. 3). The ability to identify areas of chronic infection is significant due to their negative impact of yield . This avenue of investigation, examining the chronic effects of Phytophthora root rot, is presently underway and is presented in another paper in this volume (Pozdnykova et. al.). Based on the yield estimates from this study Phytophthora root rot had significant effects on all measured components of yield (Table 2).

Figure 3. Illustration of acute and chronic injury in a cranberry bed based
on a CIR photograph with an unsupervised classification (10 classes) applied.

Table 2. Effect of Phytophthora root rot on yield and yield components of cranberry

  Diseased1 Healthy2
Upright density *** 29.2±28.5 60.8±33.5
Flowering uprights (%) *** 17.8±17.2 52.8±9.4
Yields (estimated) *** 0-200 100-550

*** indicates values for diseased and healthy vines are significantly different at the 0.001 level.
1Samples were taken from areas visually impacted by the disease
2Samples were taken from unaffected areas adjacent to the diseased areas


Cranberry yield is controlled by several components, the most significant of which is the density of fertile shoots called uprights . Uprights may be either fertile or infertile and the percentage of fertile or flowering uprights as well as the density of uprights appear to vary considerably over the sampling areas in beds infected by both types of fungal diseases. The overall effect on yield from these two diseases is shown in Table 3. In general, Phytophthora root rot occurred over larger areas and had larger impacts on yield than fairy ring. No adjustments are made at this time for chronic injury.

Table 3. Comparison of disease effects on a cranberry bed displaying both diseases

Area (ha)
Actual yield 


Potential yield


Root rot 0.04
Fairy ring 0.018
Healthy 1.67
Total 1.728


Current agricultural methodology is aimed at maximizing productivity while minimizing the area of cultivated land. With little opportunity to expand the area of production, cranberry growers must utilize new technologies to decrease the impact of farming on the environment and maximize yields. These results indicate that color infrared photography provides useful information regarding crop health and yield One significant practical conclusion from this work is that fairy ring is not an economically important disease and the cost of treatment exceeds the impact on crop productivity. Phytophthora root rot however, can cause significant crop losses, and treatment by improving soil drainage should provide long term control and significant positive economic impact for the grower.


Eaton, G. W., and Macpherson, E. A. (1978): Morphological components of yield in cranberry. Horticultural Research 17, 73-82.

Oudemans, P. V. (1999): Phytophthora species associated with cranberry root rot and surface irrigation water in New Jersey. Plant Disease 83, 251-258.

Shawa, A., Eaton, G. W., and Bowen, P. A. (1981): Cranberry yield components in Washington and British Colombia. Journal of teh American Society of Horticulural Science 106, 474-477.