Performing a National Threatened and Endangered Species Risk assessment for each pesticide registration action is daunting. The scale, both geographically, temporally, and the sheer number of species, presents a challenge. Add to this the fact
that data sets and approaches will be refined moving forward, but current registration efforts cannot be delayed. To effectively perform Proximity Analyses at Step 1, a process which is computationally efficient, technically accurate for the scale,
and robust enough to handle changing data sets must be developed. This poster demonstrates an approach to address these challenges. Using authoritative geospatial data from government sources, a National Potential Pesticide Use Site Layer is generated which can be easily updated as new data are released annually. The potential pesticide use site layer is used in a framework that allows for Flexible Action Area Definition and subsequent National Proximity Analysis of 100’s of Species at a Time. Furthermore, Downstream Transport must be accounted for when calculating proximity to Aquatic Species. By using existing Automation and Scripting methods within ArcGIS and SQL Server, the framework can be re-
applied as needed (such as availability of newer data), and also provides a source of documentation at each step of the process. This poster illustrates an approach to Implementing National-Scale Proximity Analyses with Efficiency & Reliability and Results.
Implementing National-Scale Proximity Analyses with Efficiency & Reliability Joshua J. Amos, Vivienne L. Sclater, and Christopher M. Holmes Waterborne Environmental, Inc., Leesburg, VA, USA
Steve Kay Pyxis Regulatory Consulting, Inc., Generic Endangered Species Task Force (GESTF) Gig Harbor, WA, USA
Define Potential Pesticide Use Sites
For more info, contact: Joshua Amos, ([email protected], +1.703.777.0005) 897-B Harrison Street, Leesburg, VA 20175
Figure 1: Aggregating individual NASS Cropland
Data Layer CDL crops into 13 broad crop categories like “Vegetables/Ground Fruit”
improves the certainty in the crop classification.
However, using a single year of CDL may not
sufficiently account for agriculture rotation and
potential expansion through years.
Figure 7: Distribution of Orchards/Vineyards in
relation to FWS Critical Habitats for all species
with spatial data.
Conduct National Screening Level Proximity Assessment Summarize and Interpret Proximity Assessment Results
Figure 3: Further conservatism can be applied in
counties where the multiple years of CDL do not
fully account for acreage reported in the Census
of Agriculture. This figure presents the Selective Expansion distance required for each county to
meet/exceed Census of Agriculture, county-level
acreage.
2013 NASS CDL Vegetable/Ground Fruit Crops 5 Yr. Composite (2009-2013) NASS CDL Vegetable Ground Fruit Crops
Veg/Ground Fruit: 5 Year Composite Percent of Ag Census
# o
f C
ounties
0500
1000
2000
3000
<25%
25%
-50%
50%
-75%
75%
-99.
9%
100%
-125
%
125%
-150
%
150%
-175
%
175%
-200
%
200%
-250
%
250%
-300
%
300%
-400
%
400%
-500
%
>500%
% of Census of Agriculture Acreage
0%
20%
40%
60%
80%
100%
Veg/Ground Fruit: Selective Expansion Percent of Ag Census
# o
f C
ounties
0500
1000
2000
3000
<25%
25%
-50%
50%
-75%
75%
-99.
9%
100%
-125
%
125%
-150
%
150%
-175
%
175%
-200
%
200%
-250
%
250%
-300
%
300%
-400
%
400%
-500
%
>500%
% of Census of Agriculture Acreage
0%
20%
40%
60%
80%
100%
Figure 2: Combining multiple years of CDL expands the crop footprint to ensure that the
evolving agricultural landscape is consistently
represented.
Figure 4: Further conservatism can be applied in
counties where the multiple years of CDL do not
fully account for acreage reported in the Census
of Agriculture. A distance raster calculated
outwards from multi-year CDL crop locations,
into only other agriculture, provides a framework
for Selectively Expanding the base crop
footprint by accumulating adjacent crop pixels
until the crop footprint meets or exceeds the
acreage reported in the Census of Agriculture.
Figure 5: The national distribution of
counties with their acreage of ‘Vegetables/
Ground Fruit’ crops before & after
Selectively Expanding the footprint to meet
the Census of Agriculture reported acreage.
Figure 6: The ‘Vegetables/ Ground Fruit’
pesticide use site layer was expanded 270 meters into adjacent agriculture in order to meet the Census of Agriculture acreage reported for Monterey County.
Figure 9: Standard automation tools in ArcGIS are used to calculate proximity between every
Endangered Species Location (ESL) and
potential pesticide use sites. Operations are
conducted in raster at 10 meter resolution at the
ESL/County intersection level.
Figure 10: Effects of rasterizing ESL’s (that are
natively vector) to 10 meters are negligible when
considered from the perspective of a screening
level assessment with thousands of
measurements.
Figure 8: Euclidean Distance raster is used to
measure the distance outward from every
potential pesticide use site pixel.
Figure 11: ESL’s overlain with Orchard/Vineyard Use Site Euclidean Distance Raster highlights
the range of potential proximity distances
between the two datasets.
Figure 12: Map highlights that distributions of
proximity distances are calculated for every
unique ESL polygon /county intersection. In this
way, it’s possible to generate a distribution of proximity distances for every ESL adding
further insight into the potential for interaction.
Figure 13: The proximity results table stores the distance between each ESL polygon /county
intersection and the potential pesticide use site. The Nearest distance, as well as “Percentiles”, are
recorded using the ArcGIS model in Fig 9.
Figure 14. The proximity results can be
summarized to present the number of species in
each Taxa within specific proximity distances to
potential pesticide use sites, enabling the risk manager to quickly garner the magnitude of potential interaction.
Figure 15. Proximity results can be presented as charts.
Aquatic Endangered Species Screening Level Proximity Assessment
Aquatic Species locations require an additional step for determining proximity since pesticides can be transported downstream from off-target
application sites.
0
20
40
60
80
100
0 100 250 750 >750
Amphibians
Arachnids
Birds
Conifers andCycadsCrustaceans
Dicots
Ferns andAlliesFish
0
20
40
60
80
100
0 100 250 750 >750
Amphibians
Arachnids
Birds
Conifers andCycadsCrustaceans
Dicots
Ferns and Allies
Fish
Insects
Proximity Distance From Orchards/Vineyard Use Sites
Nu
mb
er
of
En
da
ng
ere
d S
pe
cie
s in
ea
ch
Ta
xa
TAXA
NUMBER OF SPECIES IN PROXIMITY DISTANCE TO
ORCHARD/VINEYARD USE SITES
0
feet
1-100
feet
101 – 250
feet
451 – 750
feet
>750
feet
Number of
Species in
Taxa
Amphibians 12 1 0 2 1 36
Arachnids 4 0 3 0 2 13
Birds 20 1 3 0 4 65
Conifers and
Cycads 0 1 1 1 0 5
Crustaceans 2 0 1 4 4 20
Dicots 66 18 9 5 4 216
Ferns and Allies 1 4 0 4 0 14
Fish 88 6 3 0 3 191
Insects 2 5 6 9 2 55
Lichens 0 4 0 1 0 6
Mammals 16 0 9 0 6 63
Molluscs 55 22 14 4 2 212
Monocots 3 2 14 0 1 37
Reptiles 5 9 6 1 0 41
Number of
Species in
Proximity
Distance
274 73 69 31 29 498
Figure 16. NHDPlus AquaXpos analyzes the stream network to find
places, downstream from potential
pesticide use site, where the
chemical’s impact no longer exceeds a specific Level of Concern using a dilution approach for
estimating pesticide transport.
Common Name
Scientific
Name
Critical
Habitat
Polygon
Unique ID STATE County
Area of
Polygon
(M2)
NEAREST
USE SITE
DISTANCE
(Meters)
5% of Polygon
is Within
(Meters)
25% of
Polygon is
Within
(Meters)
50% of
Polygon is
Within
(Meters)
75% of
Polygon is
Within
(Meters)
100% of
Polygon is
Within
(Meters)
California red-legged frog Rana draytonii 10001 CA Alameda 463,436,186 0 365 1,026 2,044 3,606 7,991
California red-legged frog Rana draytonii 10002 CA Merced 135,365,164 0 283 805 1,440 2,362 3,567
California red-legged frog Rana draytonii 10003 CA San Benito 12,595,823 0 134 2,880 3,933 4,705 5,328
California red-legged frog Rana draytonii 10004 CA San Joaquin 100,564,396 0 481 1,405 2,373 4,160 7,006
California red-legged frog Rana draytonii 10005 CA Santa Clara 891,182,960 0 228 890 1,943 3,267 5,014
California red-legged frog Rana draytonii 10006 CA Stanislaus 59,112,383 268 1,537 3,144 4,241 5,931 7,360
California red-legged frog Rana draytonii 10007 CA San Mateo 395,250,129 0 170 582 1,134 1,987 3,878
California red-legged frog Rana draytonii 10008 CA Santa Clara 183,886 127 240 679 815 886 960
California red-legged frog Rana draytonii 10009 CA Santa Cruz 285,290,722 0 108 365 745 1,333 3,798
California red-legged frog Rana draytonii 10010 CA Santa Barbara 535,140,367 0 216 735 1,570 2,630 4,432
California red-legged frog Rana draytonii 10011 CA Ventura 51,210,538 0 547 1,448 2,289 3,045 3,808
California red-legged frog Rana draytonii 10012 CA Monterey 483,785,152 0 127 576 1,374 2,480 4,435