Return to Main Page

Methods

A review of the public records indicates that SJWC owns 6500 (± 5 ) acres of land (including about 70 acres in Santa Cruz County) under the names San Jose Water Co, San Jose Water Company, and San Jose Water Works. This is of course, excluding many small parcels within urban areas and business offices owned by SJW Land Company, SJW Land Co, and San Jose Water Works.

It is worth noting that the NTMP application dated July 1, 2006 (pg 43) includes the statement, “San Jose Water Company owns approximately 6000 acres, with the entire ownership consisting of less than 2500 acres of timberland as per PRC 4526.” However, the SEC filing of form 10-K dated March 7, 2006 by SJW Corp includes the text, “In general, the property [of San Jose Water Company] is comprised of… approximately 7000 acres of land held in fee (which is primarily non-developable watershed).” Finally, the January 2000 Watershed Sanitary Survey produced by SJWC states that SJWC owns 10,000 acres in the Los Gatos Creek Watershed and 425 acres in the Saratoga Creek watershed, including pie charts to illustrate that these represent 47% and 8% of the respective watersheds. See Supplement 3 for parcel numbers and acreage.

Mapping Details

The goals of this analysis were to map the distribution of Coast Redwoods and Douglas-fir on SJWC lands, not to map the areas that were dominated by Coast Redwoods and Douglas-fir.  For a discussion of the differences in these kinds of analyses, see the discussion in Appendix 1. Also, see Supplement 1, which includes images of different species and details of identification.

For the purposes of precise vegetation mapping, the definition of timberland in the FPR is extremely vague. The wording "available for and capable of growing... commercial species" does not include any qualifying statement about the density of these commercial species, the minimum mapping unit to be used when a commericial species is found an a unit of land, or the extent to which capability of land to support such species should be analyzed given climatic or ecological criteria. The definition of commercial species further compounds the ambiguity of potential "timberland". Do the Group B species need to be within a certain distance of Group A species in order to count? Does there need to be evidence of past Group A occupation? Does there need to be evidence of current Group A seedlings of a certain size? Does there need to be a certain density of Group A species within a certain radius in order to count as a stand of timber? Absent any kind of specific guidance on this issue from the legislation or the FPR, we defined our own criteria. However, the benefits of GIS analyses mean that we can reanalyze our data at any point with more or less strict criteria.

Using high-resolution aerial photographs that were taken by a private helicopter on February 11, 2006 and July 3, 2006, we closely examined all parcels owned by SJWC. First and foremost, we only defined areas that we could clearly identify as Coast Redwood or Douglas-fir from our aerial photographs. We drew a boundary around each set of timber species, thus grouping trees into a single grove, if the trees were within approximately 40 meters of each other. The outer boundaries were drawn to the estimate of 10 meters out from the trunk of the outside trees. Ten meters is an estimate of the typical spread of mature Coast-Redwood and Douglas-fir branches away from the trunk. In mature Coast Redwood and Douglas-fir forests, the canopy trees are often displaced 30-50 meters from each other, thus we chose 40 meter displacement as a cut-off for including trees in a grove or not. In areas with less dense timber species cover, we attempted to circle only the timber species and exclude the areas of pure hardwood. These distances were measured both in Google Earth and in ArcMap.

We were not overly concerned with the density, size, or age of Group A species because the FPR never mentions these criteria. The FPR wording "available for and capable of growing... commercial species", does not imply that the individual trees at the date of the NTMP application should be profitable for the owner. Indeed, any kind of profitability criteria (which might require the trees to be readily accessible, or of a certain size) would be completely dependent on the financial status and abilities of the owner; thus "timberland" acreage would be different for each property owner. The FPR wording is grounded in ecological criteria (i.e., physical presence of certain species), not economic criteria (i.e., presence of large, harvestable, or accessible trees). Therefore, our mapping delineated the distribution of these Group A species, without attention economic criteria. We did very little outlining of individual trees (our rule of 40 meters from another individual often ruled these trees out) but a single tree would have been given a polygon of about .1 acre and a grove of 3 isolated Douglas fir trees a polygon of about .25 acres.

Each area that we have designated as occupied with Group A timber species is backed-up with a matching aerial photo as evidence. From these photos it should be easy for anyone with interest in this matter to see and clearly identify the timber species. These photographs can be found in the "Evidence" section. Also see Supplement 2, which shows some of these aerial photographs with the timberland areas outlined.

Each of ~700 (12.3 Megapixel) aerial images was aligned to duplicate the view of the aerial image in Google Earth. We were interested in a very high-resolution analysis, thus our images were not shot on a horizontal plane. For example, we needed the helicopter to fly down into the canyon formed by Los Gatos Creek and take photographs of the sloping hillside from across the valley. We used Google Earth to align these images, in lieu of more complicated 3D mapping software such as the ArcGIS suite, because we wanted our methods to be transparent and repeatable by organizations and individuals who may not have access or expertise with the ArcGIS suite. Google Earth can be downloaded here and our files may be found here.

Example of Paired Images:

Vegetation mapping using photos taken on an angle can have advantages in many areas, especially steep slopes which may have been difficult to assess with a photograph taken against a horizontal plane. However, there are some areas in which this technique may make the mapping process more difficult, especially where tall trees may block vegetation further away from the camera. We thus used several other sets of data to clarify our mapping in areas that were not sufficiently documented with our aerial photos (namely Google Earth aerial imagery, aerial photographs found on the websites for the Counties of Santa Cruz and Santa Clara, and USGS Digital Orthophoto Quadrangles).

A large section of SJWC land had been mapped by Midpeninsula Open Space Regional District (MROSD) at a coarser scale than our mapping, using 1:24000 aerial photographs taken in 1999. This data-set was instrumental in the early stages of our mapping process, and in most areas our mapping corresponds very well this data. For more on how our data correlates with the MROSD data, see Appendix 1 .

In order to calculate acreage of the polygons drawn in Google Earth, the kml files were imported into ArcMap using the ArcGIS Extension KMLer, and projected with UTM California Zone 10. These polygons were double checked to make sure they did not extend beyond the boundaries of SJWC land (SJWC parcel data files were purchased from the County of Santa Clara) and merged to eliminate any overlapping areas. Acres were calculated from square meters with the conversion factor of .000247105381.

Areas Without Aerial Photos

We received information about the Santa Cruz County Parcels after the aerial photographs were taken. Our evidence for these parcels consist of photographs taken from the adjacent roads. We have also presented on-the-ground supporting photos for a few other areas that were missing details in our aerial photograph set. We marked the photograph location with a GPS and indicated this on the Google Earth maps. It is very difficult to document the presence of conifers from the road in photographs, however, a drive up Black Road and down Skyline Road would verify the presence of conifers in these areas for almost anyone.

Areas Inside the NTMP

Outside of the NTMP boundaries, we only circled areas that had obvious Group A timber species within approximately 40 meters of each other. Inside of the NTMP boundary, there are several areas with more sparsely distributed Group A trees, or areas that appear to be recent landslides or clearings. Since these areas are inside the NTMP boundaries, and because the language of the NTMP suggests that such areas will be subjected to future timber management, we drew specially classified polygons labeled "Out Areas" around these zones. We separated the acreage of these areas in the final analysis (in total they added up to 14.8 acres). We are aware that the NTMP 'Out' acreage was higher than this. However, the NTMP 'Out' acreage includes areas that contain immature Group A timber species, our 'Out' acreage only included those areas with either very sparse Group A timber species or no Group A timber species.

On page 97.2:

Several areas were also determined as "out" areas because they contain few to no commercial conifers at this time. These "out" areas are to be included in the NTMP, as they may be planted with commercial conifers to make them available for later management, if it is first determined that these areas are appropriate for growing conifers....Due to the potential for later management, "out" areas are still included in the total NTMP acreage, which is 1,002 acres.

Also on page 97.13:

In this vegetation type, there is not sufficient conifer volume to warrant harvest at this time. These areas have been included in the NTMP, however, for property-wide continuity, due to operational importance, and/or because of future potential as managed stands. Where old conifer stumps or other indicators suggest that these areas have the potential to grow redwood and Douglas-fir trees, trees may be planted to re-establish these commercial species...

Note: We estimated the logging zone boundaries from the maps provided in the NTMP application. These boundaries may have small discrepancies from the actual NTMP.

Group B species

We saw abundant evidence of Group B species such as Tanoak, Pacific Madrone, and Alder in many areas that we did not outline with our polygons. There was no way for us to determine whether these areas would meet the legal definition of "Timberland" by containing either small Group A species or old stumps of Group A species.

Proceed to Results