SODmap and SODmap Mobile: Two Tools to Monitor the Spread of Sudden Oak Death
Garbelotto, M., Maddison, E.R., Schmidt, D. 2014. SODmap and SODmap Mobile: Two Tools to Monitor the Spread of Sudden Oak Death. Forest Phytophthoras 4(1). doi: 10.5399/osu/fp.4.1.3560

SODmap and SODmap Mobile: Two Tools to Monitor the Spread of Sudden Oak Death

    Matteo Garbelotto, Emilie Ruth Maddison, Doug Schmidt

    Department of Environmental Science Policy and Management, 54 Mulford Hall, University of California, Berkeley, CA, 94720 USA

    Corresponding author:

1. Introduction

Disease distributions, in particular those caused by exotic or emergent pathogens, vary depending on the origin and phylogeographic history of their causal agents, the number and locations of introductions, the distribution of susceptible hosts, the presence of adequate climate, the dispersal potential of the pathogen, and the type of spread pathways (see Huberli al. 2011, Garbelotto and Pautasso 2011, Meentemeyer et al. 2011, Meentemeyer et al. 2004, Parker & Gilbert 2004). Knowledge of the actual geographic expansion of pathogens is of paramount importance when preventive measures can be adopted to mitigate their effects (Fitzpatrick et al. 2013; Estoup & Guillemaud 2010), and effective control is more likely to occur when expansion rates are low and characterized by episodic longer-distance leaps (Fontaine et al. 2013).

The case of sudden oak death (SOD) caused by Phytophthora ramorum (Rizzo et al. 2002), falls within the group of invasive diseases for which current distribution has been mostly determined by human-induced introductions, with relatively modest natural spread from such introduction points (Croucher et al. 2013, Grunwald et al. 2012). To summarize all available evidence, disease spread rates appear to be on the order of hundreds of meters per year, except for years when rainfall significantly exceeds average values resulting in new infection loci 2-5 km from established infestations (Eyre et al. 2013, Hansen et al. 2008, Mascheretti et al. 2009 & 2008, Prospero et al. 2007). The reason for such a limited spread potential of an aerial pathogen may be identified in the relatively large size of infectious sporangia (Werres et al. 2001), which are most viable when turgid and thus heavy (reviewed in Garbelotto and Hayden 2012).

Although more than a hundred plant species are susceptible to P. ramorum (USDA-APHIS, 2013), three stand out as extremely relevant from an epidemiological or ecological perspective (Garbelotto et al. 2003). Namely, these three hosts are California coast live oak (Quercus agrifolia), California bay laurel (Umbellularia californica), and tanoak (Notholithocarpus densiflorus) (reviewed in Garbelotto and Hayden, 2012). Only bay laurel and tanoak are epidemiologically relevant due to the abundant sporulation of the pathogen occurring on leaves of both species and on the petioles and twigs of tanoaks. Oak infections require high inoculum loads (>104 zoospores/mL, Garbelotto personal communication, Davidson et al. 2005), and, with rare exceptions, only oaks 10-20 m away from infected bay laurels have reportedly been infected (Kelly et al. 2007).

This paper describes an approach to: a) survey on a yearly basis those portions of California and Oregon currently infested by the SOD pathogen; b) map the distribution of surveyed trees and their infection status using a user-friendly and interactive platform, while at the same time allowing the data to be downloaded in formats compatible with GIS analyses; c) generate printable maps describing both disease presence and its frequency, and d) design apps to allow users to determine disease distribution in the field and assess the risk for oak infection at their physical location, i.e. the location in which they are standing.

2. Sources

Because the zone of infestation on the US west coast is large and fragmented, spanning nearly 800 km along the Oregon and California coast, we devised an inclusive and comprehensive approach to visualize the distribution of SOD, which in 2012 culminated in the creation of SODmap ( This database utilizes available information from previous surveys combined with a new surveying effort, the SOD Blitz Survey Project (, focused on the urban-forest interface and accomplished through the yearly efforts of hundreds of citizen scientist volunteers. The data generated by the SOD Blitz survey represent the backbone of the new SODmap database. SODmap was generated and is updated on a yearly basis by merging laboratory-confirmed data from government facilities, academic researchers, independent scientists, and citizen scientists into a single source. Data are submitted through a form and as new data are collected, they are appended to the database in order to maintain its information current. All of the details submitted through the form are preserved in the database and can be used for advanced geo-statistical analyses. However, for the sake of clarity and simplicity, only a portion of the data is displayed in the actual distribution map. Below, we present the characteristics displayed in the graphic version of the SODmap.

Criteria for Inclusion

All records are based either on in vitro culturing of the pathogen followed by its identification based on morphology, or on identification using confirmed molecular methods including ELISA, Taxon-Specific PCR, or DNA sequencing (Martin et al. 2012). We encourage inclusion of both negative and positive samples in the SODmap.

In general, plant samples are collected and processed because they appear symptomatic in the field (Davidson et al. 2003). Baiting collections from water are also included, in order to accommodate results from the extensive nationwide and statewide early detection efforts, which are based on baiting from streams. Results from stream baiting are identified on the SODmap at the location where baits were placed. It may be important to note that plants in nurseries are not included in the database.

Results I: Visualizing SOD distribution

The free version of Google Earth ( was chosen as the display platform due to its universal accessibility, operating system compatibility, and ease of use by the general public. The SODmap overlay is available for free download as a compressed kmz-formatted data file at The SODmap user interface is consistent with the Google Earth’s Users Guide and Best Practices guidelines and is based on red (SOD-positive) or green (SOD-negative) icons at each collection point. Data points that are overlapping or in close proximity to one another "spiderfy" (expand into a rosette animation) when moused-over, furthermore, selecting individual icons results in the display of a dialog box which includes most of the information associated with the sample in the database.

Although samples from any hosts and from water baits may be submitted, three different types of icons were chosen for the graphic display: triangles for bays or any other species, tree outline for oaks or tanoaks, and water drop for water baiting samples. A circle around the icon indicates that oak mortality was reported in the area and it is most informative when coupled with bay infection data. These icons allow the user to identify which hosts or substrates the data refers to without the need to activate the pop-up dialog box, which include exact host, habitat type collector and other relevant information.

By default, the SODmap opens with all data visualized in the display, but it is possible to display limited datasets by selecting folders in the "Places" sidebar displayed on the left hand side of the map. The data are hierarchically organized starting from the status of the sample (positive or negative), followed by year of sample, and by the project for which samples were collected and studied. The legend can also be turned on and off in the sidebar. Figure 1 provides an example of the structure and graphic output of the SODmap as it appears on a computer display.

Figure 1

Figure 1. SODmap screen display. A spiderfy is shown in the inset in the upper right corner.

Results II: Graphic SOD mapping

While the free version of Google Earth is extremely user-friendly, widely accessible, and interactive, it does not provide high quality printouts. This is particularly true for data-rich maps such as SODmap. To facilitate the inclusion of the comprehensive data provided by the SODmap, its data are used to generate a yearly density "heat" map. Heat maps are created in ArcMap 10.2 ( based on the NAD 1983 California (Teale) Alber projected coordinate system. The colors are continuous from red to yellow using a histogram equalized color ramp, with intensity based on the density of positive samples in a 10km-neighbor radius. Negative samples are identified as black dots. Areas where strong, and possibly successful, eradication efforts have been performed may be identified by a hatched pattern (Figure 2).

Figure 2

Figure 2. Density ("heat") map generated using SODmap data analyzed at the 10-km neighborhood level.

It should be noted that the areas lacking any sampling include areas that are both suitable and unsuitable for establishment of the pathogen (see Venette et al. 2006). However, at the time this paper is being written, most areas that are suitable for the SOD pathogen in California and Curry County (OR) have been sampled, albeit with different intensity. By the same token, "low heat" may be due to low sampling levels rather than to low disease incidence. The amount of black dots, each representing a negative result, provides a relative idea on the sampling effort in each area.

Results III: SODmap Mobile

SODmap Mobile: Free Mobile App for Sudden Oak Death Disease Management (
SODmap Mobile, as described by Matteo Garbelotto, University of California-Berkeley.

With the exception of water baiting data, the entire SODmap database is incorporated into the "SODmap Mobile" application (app) designed to run on mobile devices. SODmap Mobile is available for free through the app distribution platforms Apple App Store/iTunes and Google Play. The app allows the user to visualize all trees present in the SODmap in relation to the user's location. Using the "settings" tab, the user may select a Satellite view, a Standard map background, or a Hybrid of the two. The "Satellite" background makes it easier to identify sampled trees, but the "Hybrid" and "Standard" formats include street names, and major landmarks and thus they may be useful for orienting the user.

Upon starting, the app will show a wide view of the map area with numbers indicating the sample size, e.g. number of trees, sampled in each region (Figure 3a). At these large scales, the only information provided to the user is size of the sample in any given region. When further enlarged, the field of view on the screen will reveal individual tree icons. Trees are represented by pins with green or red heads indicating their negative or positive status for infection by the SOD pathogen, respectively (Figure 3b). Tapping the pin opens a dialog box describing the tree species, the sample ID number, and the year it was sampled. This allows for cross-referencing SODmap mobile data with the SODmap database. Visualization of the date when the sample was collected is particularly important for areas at the eastern margins of the distribution in California, where drought and hot weather may cause individual bay trees to change their infection status from positive to negative (Eyre et al. 2013).

Figure 3

Figure 3a, b. SODmap Mobile Screen Display on Apple iOS. a (large scale): the blue dot indicates the location of the user, while numbers indicate the location and sample size of trees in the database. b (fine scale): green icons identify symptomatic trees that resulted negative for SOD based on results of a lab assay. Red icons represent trees positive for SOD. The dialog box opens by tapping each icon.

One of the most important features of the SODmap Mobile app is represented by the "Risk" assessment function. This function calculates risk for oak infection based on sample intensity and on proximity of infected trees to the physical location of the user. The algorithm calculates the number of trees sampled up to 1000 m from the user's location. If at least four trees have been sampled within 1000 m, but none of them resulted positive for SOD, then risk is determined to be "Low or No". Risk is estimated as "Moderate" if at least one positive tree was detected between 200 and 1000m from the user, and as "High" if at least one positive tree is within 200 m of the user. If fewer than 4 trees have been sampled within 1000m from the user, the program returns a message of insufficient sampling data to assign a risk level. A summary of the algorithm's calculations, including the number of trees present in the area, their infection status, and the time frame in which they were sampled are also displayed when tapping the risk button (Figure 4).

Figure 4

Figure 4. SODmap Mobile Risk Assessment Screen Display on Apple iOS. See text for more details. Click to enlarge.

This information can be used to determine how robust the risk evaluation may be. It goes without saying that the larger the number of trees within 1000m of the user, and the more recent the date of sampling, the more robust the determination of risk for oak infection. A description of the app is included under its settings tab, and a more detailed description is provided at


SODmap was launched in the spring of 2012, when it contained 12,444 lab-validated data points. In 2014, at the time this article is being written, SODmap features 15,368 points. Table 1 shows the make-up of the database at the time it was launched and at the current time by submitter (researcher, citizen scientists, government) and by host (oak/tanoak, bay/other, water baiting). For each category, numbers of negative and positive samples are presented separately.

Table 1

Table 1. Origin of data in the SODmap at the time SODmap was launched in 2012 and a year later, organized by source, host, and infection status (positive or negative for Phytophthora ramorum).

To our knowledge, this is the first time a single database contains data on a plant disease distribution in natural settings coming from volunteers, independent researchers, academia, and government laboratories. It is also one of the first times a distribution database has been made public in real time, to ensure its immediate public access. Because the entirety of the data is lab-validated, the database is very robust with regards to infection status, while the geographic location of each sample may be less precise due to differences in the skills of the surveyors and the accuracy of GPS devices.

Although it is hard to determine how many people may have accessed the SODmap in its first year of existence, between March 2012 and March 2014, the website received 254,000 hits. The map of SOD distribution was also featured in major newspapers, including the front page of the San Francisco Chronicle. In less than a year, the SODmap Mobile app was downloaded 396 times for iOS and 54 times for Android. A strong educational effort is ongoing to publicize SODmap and in particular SODmap Mobile as useful tools to determine both the spread of the disease on the West Coast of the USA, and the risk for oak infection. The simple algorithm, which determines risk of oak infection based on proximity of infected trees, rather than on local disease incidence, circumvents the issue due to highly variable sampling intensity among localities. The use of SODmap Mobile can help landowners or land managers to employ disease mitigation strategies in a timely fashion. This is extremely important, as almost all known strategies can be effective only if preventively adopted (Rizzo et al. 2005).

We foresee two possible changes to the SODmap, namely: a) the inclusion of the genetic makeup of all or of a subset of the isolates. There are four known lineages in the world (NA1, NA2, EU1, EU2) (Ivors et al. 2006; VanPoucke et al. 2012) with documented phenotypic differences among them (Elliott et al. 2011), and the first NA2 lineage isolate has recently been reported on a plant in California outside of nurseries (Garbelotto et al. 2013). Additionally, there are four major clusters of genotypes within the NA1 lineage common to California forests (reviewed in Garbelotto and Hayden 2012); b) for trees that are permanently tagged, we may indicate a change of infection status on the same icon, rather than adding a further icon at the same location, as currently done.


We thank all those who submitted data to the SODmap and in particular the local organizers that make the SOD blitzes possible. This research was funded by the Gordon and Betty Moore Foundation; by State and Private Forestry, US Forest Service, Region 5, and by the NSF Ecology of Infectious Diseases program.

Literature Cited

Boutet X, Vercauteren A, Heungens K, Laurent F, Chandelier A. 2010. Oospore progenies from Phytophthora ramorum. Fungal Biology 114 (4):369–378. doi:10.1016/j.funbio.2010.02.009,

Croucher PP, Mascheretti S, Garbelotto M. 2013. Combining field epidemiological information and genetic data to comprehensively reconstruct the invasion history and the microevolution of the sudden oak death agent Phytophthora ramorum (Stramenopila: Oomycetes) in California. Biological Invasions 15(10):2281–2297. doi: 10.1007/s10530–013–0453–8

Davidson JM, Wickland AC, Patterson HA, Falk KR, Rizzo DM. 2005. Transmission of Phytophthora ramorum in mixed–evergreen forest in California. Phytopathology 95(5):587–596. doi: 10.1094/PHYTO–95–0587

Davidson JM, Werres S, Garbelotto M, Hansen EM, Rizzo DM. 2003. Sudden oak death and associated diseases caused by Phytophthora ramorum. Online. Plant Health Progress. doi:10.1094/PHP–2003–0707–01–DG

Elliott M, Sumampong G, Varga A, Shamoun SF, James D, Masri S, Grünwald NJ. 2011. Phenotypic differences among three clonal lineages of Phytophthora ramorum. Forest Pathology 41(1):7–14. doi: 10.1111/j.1439–0329.2009.00627.x

Estoup A, Guillemaud T. 2010. Reconstructing routes of invasion using genetic data: why, how and so what? Molecular Ecology 19:4113–4130. doi: 10.1111/j.1365–294X.2010.04773.x

Fitzpatrick BM, Fordyce JA, Niemiller ML, Reynolds RG. 2012. What can DNA tell us about biological invasions? Biological Invasions 14(2):245–253. doi: 10.1007/s10530–011–0064–1

Fontaine MC, Austerlitz F, Giraud T, Labbé, F, Papura, D, Richard–Cervera, S, Delmotte, F. 2013. Genetic signature of a range expansion and leap–frog event after the recent invasion of Europe by the grapevine downy mildew pathogen Plasmopara viticola. Molecular Ecology 22:2771–2786. doi: 10.1111/mec.12293

Garbelotto M, Barbosa D, Mehl H, Rizzo DM. 2013. First report of the NA2 lineage of Phytophthora ramorum from an ornamental Rhododendron in the interior of California. Plant Disease–10–13–1043–PDN. Garbelotto M, Pautasso M. 2011. Impacts of exotic forest pathogens on Mediterranean ecosystems: four case studies. European Journal of Plant Pathology doi: 10.1007/s10658–011–9928–6

Garbelotto M, Hayden KJ. 2012. Sudden Oak Death: Interactions of the exotic oomycete Phytophthora ramorum with naive North American hosts. Eukaryotic Cell 11(11):1313–1323. doi: 10.1128/EC.00195–12

Garbelotto M, Davidson JM, Ivors K, Maloney PE, Hüberli D, Koike ST, Rizzo DM. 2003. Non–oak native plants are main hosts for sudden oak death pathogen in California. California Agriculture 57(1):18–23. doi: 10.3733/ca.v057n01p18

Grünwald NJ, Garbelotto M, Goss EM, Heungens K, Prospero S. 2012. Emergence of the sudden oak death pathogen Phytophthora ramorum. Trends in Microbiology 20(3):131–138. doi:10.1016/j.tim.2011.12.006

Hansen EM, Kanaskie A, Prospero S, McWilliams M, Goheen EM, Osterbauer N, Reeser P, Sutton W. 2008. Epidemiology of Phytophthora ramorum in Oregon tanoak forests. Canadian Journal of Forest Research 38(5): 1133–1143. doi: 10.1139/X07–217.

Huberli, D, Hayden KJ, Calver M, Garbelotto M. 2011. Intraspecific variation in host susceptibility and climatic factors mediate epidemics of sudden oak death in western US forests. Plant Pathology doi: 10.1111/j.1365–3059.2011.02535.x

Ivors K, Garbelotto M, Vries I, Ruyter–Spira C, Hekkert B, Rosenzweig N, Bonants P. 2006. Microsatellite markers identify three lineages of Phytophthora ramorum in US nurseries, yet single lineages in US forest and European nursery populations. Molecular Ecology 15 6:1493–1505. doi:10.1111/j.1365–294X.2006.02864.x

Kelly M, Guo Q, Liu D, Shaari D. 2007. Modeling the risk of a new invasive forest disease in the United States: an evaluation of five environmental niche models. Computers, Environment and Urban Systems 31(6): 689–710. doi:10.1016/j.compenvurbsys.2006.10.002

Martin F, Coffey W, Zeller K, Hamelin R, Tooley P, Garbelotto M, Hughes K, Kubisiak T, Bilodeau G, Levy L, Blomquist C, Berger P. 2009. Evaluation of molecular markers for Phytophthora ramorum detection and identification: Testing for specificity using a standardized library of isolates. Phytopathology 90(4): 390–403. doi: 10.1094/PHYTO–99–4–0390

Mascheretti S, Croucher PJP, Kozanitas M, Baker L, Garbelotto M. 2009. Genetic epidemiology of the sudden oak death pathogen Phytophthora ramorum in California. Molecular Ecology 18(22):4577–4590. doi: 10.1111/j.1365–294X.2009.04379.x

Mascheretti S, Croucher PJP, Vettraino A, Prospero S, Garbelotto M. 2008. Reconstruction of the sudden oak death epidemic in California through microsatellite analysis of the pathogen Phytophthora ramorum. Molecular Ecology 11:2755–2768. doi: 10.1111/j.1365–294X.2008.03773.x

Meentemeyer RK, Cunniffe NJ, Cook AR, Filipe JAN, Hunter RD, Rizzo DM, Gilligan, CA. 2011. Epidemiological modeling of invasion in heterogeneous landscapes: spread of sudden oak death in California (1990–2030). Ecosphere 2(2):art17. doi:10.1890/ES10–00192.1

Meentemeyer R, Rizzo D, Mark W, Lotz E. 2004. Mapping the risk of establishment and spread of sudden oak death in California. Forest Ecology and Management 1–3:195–214. doi:10.1016/j.foreco.2004.06.021

Parker IM, Gilbert GS. 2004. The evolutionary ecology of novel plant — pathogen interactions. Annual Review of Ecology, Evolution, and Systematics 35:675–700. doi: 10.1146/annurev.ecolsys.34.0011802.132339

Prospero S, Hansen EM, Grünwald NJ, Winton LM. 2007. Population dynamics of the sudden oak death pathogen Phytophthora ramorum in Oregon from 2001 to 2004. Molecular Ecology 14:2958–2973. doi: 10.1111/j.1365–294X.2007.03343.x

Rizzo DM, Garbelotto M, Hansen EM. 2005. Phytophthora ramorum: Integrative research and management of an emerging pathogen in California and Oregon forests. Annual Review Phytopathology 43(1):309–335. doi:10.1146/annurev.phyto.42.040803.140418

Rizzo D, Garbelotto M, Davidson J, Slaughter G, Koike S. 2002. Phytophthora ramorum as the cause of extensive mortality of Quercus spp. and Lithocarpus densiflorus in California. Plant Disease 86(3):205–214. USDA–APHIS. 2013. List of regulated hosts and plants associated with Phytophthora ramorum. U.S. Department of Agriculture, Animal and Plant Health Inspection Service. Online publication.–2013–41.pdf

Van Poucke K, Franceschini S, Webber JF, Vercauteren A, Turner JA, Mccracken AR, Heungens K, Brasier CM. 2012. Discovery of a fourth evolutionary lineage of Phytophthora ramorum: EU2. Fungal Biology 116(11):1178–1191. doi:10.1016/j.funbio.2012.09.003

Venette RC, Cohen SD. 2006. Potential climatic suitability for establishment of Phytophthora ramorum within the contiguous United States. Forest Ecology and Management 231(1–3):18–26. doi:10.1016/j.foreco.2006.04.036

Werres S, Marwitz R, Veld W, De Cock A, Bonants P, De Weerdt M, Themann K, Ilieva E, Baayen R. 2001. Phytophthora ramorum sp nov., a new pathogen on Rhododendron and Viburnum. Mycological Research 105:1155–1165. doi:10.1017/S0953756201004865