© 2011 Plant Management Network.
Cucurbit Downy Mildew ipmPIPE: A Next Generation Web-based Interactive Tool for Disease Management and Extension Outreach
P. S. Ojiambo, Department of Plant Pathology, North Carolina State University, Raleigh, NC 27695; G. J. Holmes, Valent USA Corporation, Cary, NC 27519; W. Britton, T. Keever, and M. L. Adams, Department of Plant Pathology, North Carolina State University, Raleigh, NC 27695; M. Babadoost, Department of Crop Sciences, University of Illinois, Urbana, IL 61801; S. C. Bost, Department of Entomology and Plant Pathology, University of Tennessee, Nashville, TN 37211; R. Boyles and M. Brooks, Department of Marine, Earth and Atmospheric Sciences, North Carolina State Climate Office, North Carolina State University, Raleigh, NC 27695; J. Damicone, Department of Entomology and Plant Pathology, Oklahoma State University, Stillwater, OK 74078; M. A. Draper, USDA-NIFA, Washington, DC 20250; D. S. Egel, Department of Botany and Plant Pathology, Purdue University, West Lafayette, IN 47907; K. L. Everts, Department of Plant Sciences and Landscape Architecture, University of Maryland, Salisbury, MD 21801, and Plant and Soil Sciences Department, University of Delaware, Newark, DE 19716; D. M. Ferrin, Department of Plant Pathology and Crop Physiology, Louisiana State University, Baton Rouge, LA 70803; A. J. Gevens, Department of Plant Pathology, University of Wisconsin, Madison, WI 53706; B. K. Gugino, Department of Plant Pathology, Pennsylvania State University, University Park, PA 16802; M. K. Hausbeck, Department of Plant Pathology, Michigan State University, East Lansing, MI 48824; D. M. Ingram, Department of Entomology and Plant Pathology, Mississippi State University, Raymond, MS 39154; T. Isakeit, Department of Plant Pathology and Microbiology, Texas A&M University, College Station, TX 77843; A. P. Keinath, Coastal Research and Education Center, Clemson University, Charleston, SC 29414; S. T. Koike, University of California Cooperative Extension, Salinas, CA 93901; D. Langston, Department of Plant Pathology, University of Georgia, Tifton, GA 31793; M. T. McGrath, Department of Plant Pathology and Plant-Microbe Biology, Long Island Horticultural Research and Extension Center, Cornell University, Riverhead, NY 11901; S. A. Miller, Department of Plant Pathology, Ohio State University, Wooster, OH 44691; R. P. Mulrooney, Department of Plant and Soil Sciences, University of Delaware, Newark, DE 19716; S. Rideout, Department of Plant Pathology, Physiology and & Weed Science, Eastern Shore AREC, Painter, VA 23420; E. Roddy, Ontario Ministry of Agriculture, Ridgetown, ON Canada N0P 2C0; K. W. Seebold, Department of Plant Pathology, University of Kentucky, Lexington, KY 40546; E. J. Sikora, Department of Entomology and Plant Pathology, Auburn University, Auburn, AL 36849; A. Thornton, Harris Moran Seed Company, Sun Prairie, WI 53590; R. L. Wick, Department of Plant, Soil & Insect Sciences, University of Massachusetts, Amherst, MA 01003; C. A. Wyenandt, Department of Plant Biology and Pathology, Rutgers University, Bridgeton, NJ 08302; and S. Zhang, Tropical Research and Education Center, University of Florida, Homestead, FL 33031
Ojiambo, P. S., Holmes, G. J., Britton, W., Keever, T., Adams, M. L., Babadoost, M., Bost, S. C., Boyles, R., Brooks, M., Damicone, J., Draper, M. A., Egel, D. S., Everts, K. L., Ferrin, D. M., Gevens, A. J., Gugino, B. K., Hausbeck, M. K., Ingram, D. M., Isakeit, T., Keinath, A. P., Koike, S. T., Langston, D., McGrath, M. T., Miller, S. A., Mulrooney, R., Rideout, S., Roddy, E., Seebold, K. W., Sikora, E. J., Thornton, A., Wick, R. L., Wyenandt, C. A., and Zhang, S. 2011. Cucurbit downy mildew ipmPIPE: A next generation web-based interactive tool for disease management and extension outreach. Online. Plant Health Progress doi:10.1094/PHP-2011-0411-01-RV.
Cucurbit downy mildew (CDM), caused by Pseudoperonospora cubensis, is one of the most important diseases affecting cucurbits worldwide. In the USA, host resistance in cucumber had adequately controlled the disease with very minimal application of fungicides from the late 1960s to 2004. In 2004, there was a resurgence of the disease that devastated the cucumber crop in several states in the eastern USA. Since then, host plant resistance alone has not been sufficient to adequately control the disease and now control relies heavily on application of fungicides. To effectively apply fungicides in a timely manner, cucurbit growers, extension personnel, and crop consultants and advisors can now utilize information on disease occurrence and predicted spread disseminated through the United States Department of Agricultures CDM ipmPIPE decision support system developed by scientists at North Carolina State University. Based on a survey conducted in Georgia, North Carolina, and Michigan, the CDM ipmPIPE resulted in an average reduction of 2 to 3 fungicide applications in 2009 compared to calendar-based fungicide sprays. With approximately 122,000 acres of cucurbits in these three states, this translates to more than $6 million in savings to the producers in these three states. Economic savings and positive environmental implications of reduced fungicide applications demonstrate the value of a coordinated national monitoring network for management of a plant disease that is disseminated aerially over long distances.
The development of forecasting systems for aerially dispersed plant pathogens began with an early warning system for tobacco blue mold, caused by Peronospora tabacina, established in 1945 by the USDA (3). The pattern of south-to-north spread of P. tabacina spores and the success of previous programs for reporting blue mold epidemics later served as the impetus of the North American Warning System in the 1980s (3). In the mid-1990s, C. E. Main integrated the spore transport modeling framework with the blue mold warning system to create an on-line blue mold decision support system for tobacco growers in the USA and Canada as part of the North American Plant Disease Forecasting Center (3). This system evolved in subsequent years and incorporated new decision support components in response to feedback from stakeholders. The key component of the blue mold warning system was the Internet User Interface Subsystem which facilitated effective communication among tobacco growers, federal agency personnel, and scientists. Local growers, crop consultants, and extension personnel were involved at both ends of the information flow; they monitored the fields for disease and were the final users of the risk assessments and management advice. In 1997, the modeling methodology developed for tobacco blue mold was applied to forecast the movement of small aerobiota, including Pseudoperonospora cubensis sporangia (causal agent of cucurbit downy mildew) and the pollen of mountain cedar (3).
Based on the working framework developed for the tobacco blue mold warning system, advanced information technologies were recently applied to the spread of soybean rust in the USA at a level never previously reached for agricultural pests in the USA in an effort to provide soybean growers, extension agents, and consultants with an effective decision support system for managing soybean rust (8). Due to the success of the soybean rust early-warning, monitoring and communication system, a similar template was used to facilitate the launch of the Integrated Pest Information Platform for Extension and Education (ipmPIPE) in 2006 (11). The ipmPIPE program integrates technology and people around the nation who are networked by Information Technology (IT) that supports disease and pest observation networks, diagnostic laboratories, data management, modeling, interpretation, and the dissemination of information on a well-integrated IT platform (4). A key philosophical basis of the ipmPIPE is that extension and education outreach activities for integrated pest management (IPM) and risk management associated with crop insurance should go hand-in-hand (4). The National IPM Road Map seeks to hasten progress in the adoption of IPM practices (11), and the ipmPIPE integrates within this process to deliver information that growers need to advance and support their disease and pest management decisions. The USDA Risk Management Agency (RMA) requested changes to the ipmPIPE in 2007 to expand it to other pest component systems. As a result, the CDM ipmPIPE was launched in 2008 as part of the ipmPIPE.
Rationale and Establishment of CDM ipmPIPE
Cucurbit downy mildew (Fig. 1) is one of the most important diseases affecting cucurbits worldwide (5). Since the 1960s, resistance in Cucumis sativus had adequately controlled the disease in the USA, with very minimal use of fungicides (2). However, in 2004 there was a resurgence of the disease on cucumber in the USA that devastated the cucumber crop in North Carolina, Virginia, Delaware, Maryland, and New Jersey and this led to the widespread use of fungicides to control the disease. Due to the expense of preventative fungicide programs for both retail and commercial production, cucurbit growers are intensely interested in knowing when and where downy mildew will occur.
The absence of overwintering oospores for P. cubensis coupled with the sensitivity of cucurbitaceous plants to frost implies that the pathogen cannot survive in areas where temperatures during the winter are low enough (< 1°C) to destroy (i.e., freeze) the cucurbit plants. Thus, it is assumed that the pathogen only survives in frost-free areas of southern Florida and along the Gulf of Mexico, as active mycelium in cultivated or wild species of cucurbits (1). Therefore, in areas that experience regular frost, disease epidemics will depend on the aerial dispersal of P. cubensis sporangia from the subtropical overwintering sources in the south each year (6).
The CDM ipmPIPE was established to primarily provide end-users with near real-time flexibility in decision-making for the in-season management of cucurbit downy mildew with the objective of enhancing the role of the extension specialists by providing: (i) near-real time access to disease observations, (ii) status of the disease epidemic, (iii) current disease forecasts, (iv) disease management information, and (v) communication tools to support location-specific disease management decision-making by crop consultants and growers during the season (7). As part of the CDM ipmPIPE program, cucurbit extension specialists from 25 states in the eastern USA, California, and Ontario, Canada, established new guidelines for monitoring, diagnosis, and reporting of new outbreaks to the forecasting website (cdm.ipmpipe.org) (7). Initially, the CDM ipmPIPE information technology template was comprised of two websites: a public and a password-protected (restricted) site. The latter site was designed for researchers and state coordinators to access additional technical data on the trajectory of spore clouds from different inoculum sources and disease forecasting consoles. In 2008, the public website was opened for reporting disease outbreaks by state coordinators (with restricted access) to accommodate reporting by specialists and the public whose reports were subject to confirmation by extension specialists. Depicting the national status of disease outbreaks based on new reports in the eastern USA and Ontario, Canada, were introduced in 2008. The epidemic status map (Fig. 2), which depicts the current status of the disease, is automatically generated and updated when state coordinators report new disease outbreaks or when a new report from a grower or specialist is confirmed (i.e., pathogen positively verified as P. cubensis, by microscopic observation of pathogen structures) by a state coordinator. Along with these epidemic status maps comes a summary of affected counties, affected hosts, and the status of disease outbreaks from different locations are included (Fig. 3). The interactive Google-based epidemic status map allows stakeholders to obtain key data on the status of an epidemic at a county level. Updated links to control recommendations in different states and disease diagnosis information are also available on the website.
In this review, we introduce key components of the CDM ipmPIPE that may serve as a working framework to develop decision support systems for other plant diseases that are similar in biology to the cucurbit downy mildew pathosystem.
Disease Monitoring Using Sentinel and Non-sentinel Plots
In 2008 and 2009, the CDM ipmPIPE program consisted of a network of individuals that established 87 sentinel plots located in 25 states in the USA and in Ontario, Canada (8). Six different cucurbit types, Cucumis sativus (cucumber cv. Straight 8 and Poinsett 76), Cucumis melo (cantaloupe cv. Hales Best Jumbo), Cucurbita pepo (acorn squash cv. Table Ace), Cucurbita maxima (giant pumpkin cv. Big Max), Cucurbita moschata (butternut squash cv. Waltham), and Citrullus lanatus (watermelon cv. Micky Lee), were selected by cucurbit specialists and stakeholders for monitoring and reporting disease outbreaks. All sentinel plots were located at either research facilities or in commercial fields and were standardized in dimensions with an approximate size of 50 ft × 200 ft. Standard production practices for the state and/or region were followed for all the plots. The latitude and longitude data for each plot was recorded using hand-held Global Positioning System devices (GPS). Following crop establishment, sentinel plots were monitored for disease symptoms weekly to biweekly, depending on availability of resources and the progress of the epidemic at the local, regional, and national level. Disease was also monitored in non-sentinel plots that consisted of commercial fields, research plots, and home gardens not designated a priori for regular surveillance. When the first disease symptoms were detected and confirmed to be caused by P. cubensis, plots were rated for disease incidence (based on number of infected plants per plot) and disease severity (proportion of leaf area infected). Additional information was collected on the date of disease outbreak and affected cucurbit host type to facilitate online reporting (described below) of new disease outbreaks. Among the states that had higher number of new disease cases (> 20) across the two years were: Florida, Georgia, Michigan, North Carolina, New York, Ohio, and Pennsylvania (Table 1).
Table 1. Summary of reports of cucurbit downy mildew from sentinel plots, commercial fields and home gardens in the United States and Canada during the 2008 and 2009 growing seasons.
Structure of the CDM ipmPIPE
Each participating state/province has a designated state coordinator that: (i) facilitates involvement of local cooperators, promotes awareness of the disease forecasting website, and provides access to diagnostic training for cucurbit downy mildew; (ii) establishes linkage with state diagnosticians to share primary information on cucurbit downy mildew and disease reports generated by the sentinel or non-sentinel plots and other activities during the season; and (iii) establishes a linkage with the CDM ipmPIPE website at cdm.ipmpipe.org to reporting disease outbreaks. Reports of outbreaks are entered online via a public website and the updated epidemic status map is then made available to the public. The UNC-system-affiliated state climate office housed at North Carolina State University maintains the public website and provides weather information, data management, and related information technology support for the CDM ipmPIPE.
The internet platform and related databases, mapping and communication software were initially created by a team of scientists from ZedX Inc. (Bellefonte, PA) and Pennsylvania State University as part of the restricted website. The restricted site housed interfaces for tools that were used for developing forecast trajectories for sporangia transport and tools for generating a console for the epidemic status map and disease risk maps that are channeled to the public web site. The information to support disease management decisions is disseminated by extension specialists through the national platform directly and indirectly to end-users. The quality and utility of this information is increased because state extension specialists have real-time access to: (i) maps of the current distribution of disease outbreaks with the visual depiction of predicted risks along the predicted trajectory of spore movement; (ii) disease forecasts that are issued 2 to 3 days per week during the growing season; and (iii) disease management commentaries and guidelines from colleagues in other states.
In early 2010, a new tool was added to the public website that allows users to sign up to receive real-time customized disease alerts directly via cell phones and/or electronic mail accounts when new and confirmed disease outbreaks are reported. Those who sign up for this alert system have the option of receiving the customized information such as: (i) the risk of CDM developing within a given radius around an area of operation; (ii) new and confirmed disease outbreaks and the affected cucurbit host type as well as the approximate distance from a selected point of reference; and (iii) map of the current epidemic status. The cucurbit downy mildew alert system gives end-users real-time flexibility in decision-making in the in-season management of cucurbit downy mildew. For example, based on electronic mail or phone text alerts sent to a grower regarding the risk for disease outbreak, a quick decision can be made to initiate the first fungicide spray when downy mildew is found within a user-defined area.
In the section below, we use graphic illustrations to provide an overview of key features created for the interactive CDM ipmPIPE website in consultation with cucurbit extension and research specialists, diagnosticians, industry representatives, stakeholders, coordinators, and scientists at ZedX and the North Carolina State Climate Office. Challenges and opportunities associated with each of these features are reviewed yearly by a steering committee comprised of selected state coordinators, crop consultants, and pickling cucumber industry stakeholders as the public website continues to evolve as a dynamic and interactive IPM resource.
Components of the Interactive CDM ipmPIPE Website
I. Status of disease epidemics. The home page of the public website of the CDM ipmPIPE has two
interactive Google-based maps. The smaller of the two maps is a thumbnail map
(Fig. 2A) that shows a small customized image of the status of the downy mildew
epidemic for a specific region. This thumbnail map is dynamic in content and
will display the current status of the epidemic within a given region dictated
by the location of the internet protocol address of the client computer. The
larger of the two maps (Fig. 2B), displays the current epidemic status in the
United States and Canada. These maps are updated automatically each time a new disease
outbreak is confirmed and reported to the website. A click on either the
thumbnail or national map generates another interactive map that displays
detailed information on the status of the epidemic and affected counties in each
state (Fig. 3A). In this latter map, red symbols denote counties with confirmed
disease outbreaks reported within the last 7 days at the time the map is
accessed. The green symbols denote counties with disease outbreaks reported at
least 7 days prior to the time the map is accessed. Dark blue symbols denote
counties with disease outbreaks that are no longer present at the time the map
is accessed. During the season, infected plots are no longer
present and do not serve as inoculum sources if the crop is harvested or
destroyed. The white symbols denote counties where sentinel plots were
scouted but downy mildew was not observed. At the top of the right-hand panel of
the disease status map is a link that allows one to access an interactive list
of disease outbreaks containing detailed information on the following: (i) date
when outbreak was reported; (ii) affected county and state or province; (iii)
date when symptoms were observed; and (iv) the affected cucurbit host (Fig. 3B).
II. Disease forecasts and reporting of disease outbreaks. The accuracy of disease forecasts relies on reports of new disease outbreaks. Reports of new outbreaks provide key information on the source of inoculum present for transport to uninfected fields. Thus, reporting of disease outbreaks is an important component of the CDM ipmPIPE. Reporting of disease outbreaks can be done by anyone using the link Reporting Disease Outbreaks (Fig. 4). Each reported outbreak must be confirmed by a state coordinator before it can be posted on the website and displayed on the epidemic status map. As the CDM ipmPIPE platform continues to evolve, it is envisioned that growers and home gardeners will contribute more to scouting and reporting of disease outbreaks. Diagnostic guidelines are available on the CDM ipmPIPE homepage under the link Field Identification that facilitates the identification of P. cubensis and reporting of accurate downy mildew outbreaks. Current regional or state control recommendations for cucurbit downy are available under the link Control Recommendation to facilitate timely and efficient access of information by growers, extension agents, and crop consultants.
To ensure that all reports provided are accurate, reports from all individuals who are not designated as state coordinators or extension specialists in the CDM ipmPIPE database are subjected to additional confirmation by designated state coordinator or extension specialist before they are displayed on the epidemic status map. To facilitate location-specific outreach efforts, reporters can provide GIS coordinates of their location or simply use the interactive Google-based map at the end of the online form to pin-point and automatically generate the coordinates of their specific location. Once the report is submitted and verified, the epidemic status map on the website is automatically updated to facilitate location-specific outreach efforts to end-users. Using data on confirmed reports and weather variables along projected paths of sporangia transport, disease forecasts (Fig. 5) are issued 2 to 3 times per week. The forecasting page is accessed using the link Current Forecast on the main page of the CDM ipmPIPE. Each forecast includes a commentary on regional weather and disease risk and an interactive map that depicts risk of infection in specific locations along the projected paths of sporangia transport. On each disease risk map, areas of low, moderate and high risk for cucurbit downy mildew are color coded in yellow, orange, and red, respectively. Archives of past disease risk maps and corresponding commentaries are available at the end of the disease forecasting page. Archives of disease risk maps can also be accessed directly on the CDM ipmPIPE homepage menu.
III. CDM alert system. To enhance the value of the CDM ipmPIPE, growers can receive customized real-time information on the status of the epidemic, disease forecasts, and new disease outbreaks within their area of operation. If the user already has an account, the sign-up window (Fig. 6) can be accessed by using the link Create New Monitoring Location. A user can receive alerts by electronic mail or as a text message on a mobile phone. To ensure location-specific outreach efforts, users can provide GIS coordinates or address of their location of interest or simply use the interactive Google-based map at the end of the online form to pin-point and automatically generate the coordinates of the location of interest. Users have the option of receiving the alerts when a new report is confirmed within a 75- to 1000-mile radius around their location of interest. Users also have the option to receive reports on all disease outbreaks in the USA and Canada.
Impacts of the CDM ipmPIPE
Print and online educational materials have been produced and distributed in the USA and Canada. For example, over 6600 CDM ipmPIPE brochures have been distributed to cucurbit producers and consultants across the United States since 2008, while up to 700 brochures have been distributed to clientele and cooperators in Canada, a key partner in early detection of CDM since 2005. Diagnostic resources on the website have been instrumental in expanding the ability of agricultural professionals and producers to recognize and report disease outbreaks.
A total of 9,156 alerts were sent between 25 March 2010 and 29 September 2010 through the CDM alert system. At the end of September 2010, 155 individuals had signed up for the alert service of which 28% were growers, 7.1% were industry personnel, and 7% were crop consultants, showing a strong stakeholder base. Of these 155 individuals, 49% used the forecasting website regularly, while 33% indicated that past visits to the website helped them prevent yield losses. Sixty-seven percent indicated that information on the website helped them to correctly diagnose cucurbit downy mildew. Perhaps of more importance, 57% of those that signed up indicated that the CDM ipmPIPE forecasting website and the alert system was very useful in their effort to control the disease.
The CDM ipmPIPE working group has established common fungicide control recommendations for cucurbit production regions in all production areas in the eastern United States. This has eliminated some of the state-to-state or region-to-region variations in recommendations that can be confusing to producers and managers that operate production systems in multiple states. Extension specialists in Georgia, North Carolina, and Michigan report that growers saved between 2 and 3 fungicide applications in 2009 due to information posted about the threat from cucurbit downy mildew. With about 122,000 acres of cucurbits in the three states and at an average cost of $25/acre for a single fungicide application, this translates into more than $6 million in savings to producers in these states. These three states alone account for about one-fourth of the United States cucurbit production. In addition, fewer fungicide applications were more likely to be done in regions such as the mid-Atlantic and Northeast where detection and monitoring of CDM in the southeast United States helped to better time fungicide applications in the north. Thus, by simply having better monitoring and prediction capabilities of potential outbreaks into more northern regions of the USA, growers in the mid-Atlantic and Northeast regions of the United States can save millions of dollars on an annual basis by not spraying for downy mildew if it is unnecessary. For example, a single fungicide application to all cucurbit acres in the United States would cost growers about $11 million. Consequently, the savings in fungicide application costs alone create production efficiencies for producers and reduce unnecessary exposure of workers, consumers, and the environment to pesticides targeted for plant pathogens. Reducing applications of fungicides with single modes of action also may slow development of resistance to these active ingredients in P. cubensis (10).
Sustaining the CDM ipmPIPE
The resurgence of cucurbit downy mildew in 2004 in the USA coupled with devastating crop losses across the country inspired an unprecedented level of cooperation among USDA agencies, state departments of agriculture, universities, industry, and grower organizations to develop a coordinated framework to help growers manage this devastating disease. The National Institute for Food and Agriculture is advocating for the ipmPIPE as a national effort to enhance the adoption of IPM practices, and the USDA Animal and Plant Health Inspection Service is supporting the platform in anticipation of using it to respond quickly to future threats from exotic diseases and pests detected and reported by ipmPIPE specialists (4). In the long-term, however, the ipmPIPE and its evolving components such as the CDM ipmPIPE, Legume ipmPIPE (9), Pecan ipmPIPE, or Southern Corn Rust ipmPIPE will only be viable if there is real-time accessibility and economic return to growers and crop consultants. This can be achieved by applying advanced information technology for scouting, diagnosis, data management, modeling, interpretation, and rapid dissemination of information to end-users. For the CDM ipmPIPE, this is being realized with the development of customized features in the cucurbit downy mildew alert system. Ultimately, sustainability of the CDM ipmPIPE will draw more on the support from stakeholders, national and state cucurbit grower organizations, seed companies, and agrochemical industries whose clientele are direct consumers of decisions support tools developed by the CDM ipmPIPE.
The CDM ipmPIPE was supported by a grant from the United States Department of Agriculture Pest Information Platform for Extension and Education (PIPE) Program. We wish to thank Scott Isard (Pennsylvania State University), Jim VanKirk (Southern IPM Center), Joe Russo (ZedX Inc.), Julie Golod (ZedX Inc.), and other individuals that have contributed their ideas, time, and effort to this project.
1. Cohen, Y. 1981. Downy mildew of cucurbits. Pages 341-354 in: The Downy Mildews. D. M. Spencer, ed. Academic Press Inc., New York, NY.
2. Holmes, G., Wehner, T., and Thornton, A. 2006. An old enemy re-emerges. Am. Vegetable Grower. February: 14-15.
3. Isard, S. A., and Gage, S. H. 2001. Flow of Life in the Atmosphere: An Airscape Approach to Understanding Invasive Organisms. Michigan State Univ. Press, East Lansing, MI.
4. Isard, S. A., Russo, J. M., and DeWolf, E. D. 2006. The establishment of a national Pest Information Platform for Extension and Education. Online. Plant Health Progress doi:10.1094/PHP-2006-0915-01-RV.
5. Lebeda, A., and Widrlechner, M. P. 2004. Response of wild and weedy Cucurbita L. to pathotypes of Pseudoperonospora cubensis (Berk. & Curt.) Rostov. (cucurbit downy mildew). Pages 203-210 in: Advances in Downy Mildew Research Vol. 2. P. T. N. Spencer-Phillips, and M. Jeger, eds. Kluwer Academic Publishers, Dordrecht, The Netherlands.
6. Nusbaum, C. J. 1944. The seasonal spread and development of cucurbit downy mildew in the Atlantic coastal states. Plant Dis. Report. 28:82-85.
7. Ojiambo, P. S., Kanetis, L., Holmes, G. J. 2009. Forecasting long distance movement of Pseudoperonospora cubensis and the Cucurbit ipmPIPE. (Abstr.) Phytopathology 99:S171.
8. Roberts, M. J., Schimmelpfennig, D., Ashley, E., and Livingston, M. 2006. The value of plant disease early-warning systems: A case study of USDAs soybean rust coordinated framework. Online. Econ. Res. Rep. No. 18. USDA-ERS, Washington, DC.
9. Schwartz, H. F., Langham, M. A. C., Golod, J., Tolin, S. A., LaForest, J., Cardwell, K. F. 2009. Legume ipmPIPE: The next evolution of web-based interactive tools for disease management and extension outreach. Online. APSnet Features. doi: 10.1094/APSnetFeature-2009-0509.
10. Skylakakis, G. 1981. Effects of alternating and mixing pesticides on the buildup of fungal resistance. Phytopathology 71:1119-1121.
11. USDA. 2005. Agriculture secretary approves funding for soybean rust surveillance and monitoring. Online. USDA News Release No. 0160.05. USDA, Washington, DC.