MAF Biosecurity New Zealand - High Risk Site Surveillance

The 1.8 million hectares of plantation forestry estate in New Zealand currently provides annual exports valued at over $3.7 billion annually (Ministry of Agriculture and Forestry Situation and Outlook for New Zealand Agriculture and Forestry - June 20101). This is forecast to increase to over $6 billion by 2014 (MAF SONZAF 2010). Both the urban and indigenous forest estates (over 6.5 million hectares) also have incalculable economic value, quite possibly outweighing that of exotic forestry, as well as environmental, social and cultural value. To protect these assets, MAF Biosecurity New Zealand carries out High Risk Site Surveillance annually throughout New Zealand.

High Risk Site Surveillance, traditionally known as port-environs or hazard site inspections, was initiated over 25 years ago to provide detection capability for new exotic organisms. It was clearly understood even at that time that the majority of organisms of threat to exotic plantation forestry entered and established in New Zealand in and around ports (air and sea), therefore this was the best place to find and eradicate these pests. Initially surveillance was initiated to protect exotic plantations but now has come to include the valuable urban and indigenous forest environments as well as providing some protection for tree fruit crops.

As a result of joint review by MAFBNZ and the forest industry in 2004, a revamped surveillance programme started in November 2005. The focus changed from inspecting a small number of ports to inspecting around over 6,000 transitional facilities along with traditional risk sites.

1.1 The Purpose of HRSS

1.1.1 The objectives of HRSS

HRSS has primary and secondary goals. The primary goal of HRSS surveillance is to:

  • provide effective detections of plant pests that may have potentially adverse effects on the environment, economy or people of New Zealand.

An effective detection is defined as detection of a pest species (e.g. an exotic insect or pathogen) at a stage where:

  • the maximum range of management options, including eradication, are available; and
  • the possibility of eradication is maximised, while economic and social impacts are minimised.

The secondary goals of HRSS are to:

  • provide justifiable claims of pest free status or area freedom to trade partners (e.g. under the International Plant Protection Convention);
  • monitor pest distributions (Crosby regions (Crosby et al. 19882) are used for this purpose); and
  • collect host record data for established species.

1.1.2 The Scope of HRSS

HRSS targets trade pathways and associated risk sites because these are the most likely areas where exotic organisms will enter and establish in New Zealand.

The coverage of HRSS includes:

  • high risk sites such as ports (sea and air with international traffic) and transitional facilities where sea containers are unloaded;
  • high risk sites such as first night campsites associated with risk from overseas visitors;
  • areas associated with these pathways and risk sites containing a wide range of plant and tree species, such as parks; and
  • miscellaneous risk sites, e.g. military bases with returning personnel & equipment, post border incursion events, large scale industrial projects.

1.2 The HRSS methodology

1.2.1 Sampling strategy

Historically HRSS relied mainly on the ‘walkthrough’ surveillance method. This allowed the individual surveyor to determine where the survey occurred, and was unable to provide repeatable and consistent surveillance. HRSS now makes extensive use of transects (or ‘sample plots’ – see Figure 1) to provide ‘discrete packets’ of intensive surveillance. The following advantages are available to HRSS by using transects:

  • inherent flexibility, as transects can be deployed in any shape/size combination necessary to cover vegetation;
  • precise control over where surveillance occurs, allowing repeatability if desired;
  • efficiency, as transects act as a sampling mechanism;
  • enhanced auditability;
  • management of inspections and inspection data from GIS; and
  • estimation of detection probabilities (see Box 1).

Figure 1: Example of application of transects to a risk site area

Figure 1: Example of application of transects to a risk site area

 

Box 1. Random Point Sampling and Detection Probabilities

The probability of a random point sample (e.g. a transect) intercepting a pest incursion is calculated using a) the size of the transect being used, b) the size of the target infestation, and c) the total area surveyed (e.g. the RSA) applied to the formula from Carter (1989)3:

P = 1-(1-p)n
where:
P = probability of interception
p = proportion of the surveillance area infested (size of infested area)
n = number of samples (i.e. transects)

This gives the interception probability in the figure below labelled 'transect only'. This only provides the probability that a transect will intercept an infestation. A detection probability combines the interception value and a detection probability estimated from trials. The figure below shows the detection probability using four classes of object visibility (from Murphy and Baird unpublished data4).

Detection probabilities are estimates, with inherent uncertainties and assumptions. Assumptions implicit in using point sampling methods include an essentially contiguous vegetated area (more appropriate for forests than urban areas where HRSS will predominantly occur), infestation areas are circular, and transects are randomly located.

Transect graph

 

1.2.2 Risk site areas (RSAs) and vegetation rich areas (VRAs)

HRSS inspections take place within the RSA, which is an arbitrarily defined boundary around a high risk site. Boundaries for different types of high risk sites are defined by simple rules which differ slightly when applied to individual or clustered high risk sites (see Box 2).

Box 2. Defining the boundaries of RSAs and VRAs

All high risk sites receive a boundary within which surveillance occurs. For ports and individual high risk sites, surveillance occurs within a radius of approximately 500 metres from the centre. Where aggregations or clusters of high risk sites occur, the RSA is further defined by a polygon that delineates industrial from residential areas. The example below shows the industrial area defined around a cluster of transitional facilities (yellow dots). Surveillance of vegetation occurs within the boundary.

Surveillance would also occur in VRAs associated with this cluster, such as in Mangere Domain.

Mangere Domain VRA

Surveillance of RSAs is insufficient in itself to provide a comprehensive pest detection capability. Vegetation in some RSAs may be scarce (e.g. heavy industrial zones), or species limited. Many invasive species are capable of dispersing towards suitable host material occurring outside the RSA. Hence, HRSS also describes and surveys areas associated with each high risk site that increase the range of plant species available for inspection, termed vegetation rich areas (VRAs). This should allow for detection of a broader range of pest organisms. Typical VRAs include parks, golf courses, botanical gardens or arboretum. The boundary of a VRA is defined by its natural perimeter.

1.2.3 Combining inspection methods

Reliance on a solitary surveillance method, such as walkthrough, is problematic. HRSS combines both intensive and extensive methods, obtainable using transects and walkthrough surveillance. During HRSS, transect inspections are intensive, and as HRSS surveyors move between transects, the extensive walkthrough phase occurs. The walkthrough phase is directed using ‘pathway’ indicators on maps (see arrows in Figure 1), to clarify where surveillance occurs. In addition, a discretionary phase occurs when all transects have been inspected in a RSA (see Box 3).

Box 3. Phases of a Risk Site Area Inspection

HRSS utilises three distinct inspection phases. HRSS should obtain the best of each method whilst minimising their weaknesses.

  • Phase One is the intensive Transect Inspection: HRSS surveyors are given maps for each RSA. These maps display the location of transects. Inspections are typically visual, although specific surveillance tools (e.g. beating trays) may be applied. All detection types (e.g. new to New Zealand, new host/region) are expected in transect inspections. An exclusive reliance on transect inspections would provide relatively small coverage for a sizeable cost.
  • Phase Two is the extensive Walkthrough Surveillance: HRSS surveyors moving between transects undertake walkthrough surveillance. HRSS surveyors are expected to investigate damage symptoms likely to provide 'new to New Zealand' or other significant detections. Where more new host/region detections are required, this detection 'gap' is amended by MAF Biosecurity New Zealand issuing special instructions. An exclusive reliance on walkthrough surveillance was one of the historical faults of HRSS.
  • Phase Three is Discretionary Surveillance: After all transects on a map have been inspected, the surveyor uses discretionary surveillance during the remaining inspection hours to survey vegetation and wooden/timber materials within the RSA. Surveyors are encouraged to use their experience or skills to provide significant detections. An exclusive reliance on discretionary surveillance would encourage the same problems as does an exclusive reliance on walkthrough surveillance. Approximately one in eight hours is allocated to discretionary surveillance.

1.2.4 Special Instructions

Special instructions provide the flexibility and responsiveness required by HRSS. Each special instruction is essentially a direction to HRSS surveyors to undertake (or prevent) particular behaviours. Each special instruction is issued by MAF Biosecurity New Zealand as an electronic form.

The flexibility of special instructions results from their ability to be directed to a particular surveyor, transect, RSA, region, or nationally. They can also be implemented to target a particular host plant or organism (particularly if a new organism has recently been detected), prevent submission of particular sample types, or mandate the use of a particular inspection technique (e.g. beating trays or trapping) above those already prescribed.

1http://www.maf.govt.nz/mafnet/rural-nz/statistics-and-forecasts/sonzaf/

2Crosby et al. (1988): Crosby, T K; Dugdale, J S; Watt, J C (1988): Area Codes for Recording Specimen Localities in the New Zealand Sub-region. New Zealand Journal of Zoology, 25: 175-83

3CARTER, P.C.S. (1989): Risk Assessment and Pest Detection Surveys for Exotic Pests and Diseases which threaten Commercial Forestry in New Zealand. New Zealand Journal of Forestry Science, 19(2/3): 353-74

4MURPHY, B.D. & BAIRD, D. (2004): The Feasibility of Surveillance Around Transitional Facilities. Unpublished MAF report

Page last updated: 14 July 2010