Climate Data

These maps are derived from the daily, 800m resolution, PRISM data spanning 1981–2010.

Basic Summaries

Annual Estimates

final_MAAT_800m.tif Mean annual air temperature (deg. C), derived from daily minimum and maximum temperatures.
final_MAP_mm_800m.tif Mean accumulated annual precipitation (mm), derived from daily totals.

Monthly Estimates

final_monthly_tavg_800m.tif Mean monthly temperature (deg. C), derived from daily minimum and maximum temperatures.
final_monthly_ppt_800m.tif Mean accumulated monthly precipitation (mm), derived from daily totals.

Frost-Free Period

Number of days in the 50%, 80%, and 90% probability frost-free period, derived from daily minimum temperatures greater than 0 degrees C.

These maps are based on 50/80/90 percent probability estimates for the last spring frost and first fall frost (day of year). See the related algorithm documentation for details.

Values have been cross-checked with 300+ weather stations in CA.

Linear Regression Model

 ols(formula = ffd.50 ~ prism_ffd, data = z)

	Model Likelihood Ratio Test	Discrimination Indexes
Obs 328	LR χ² 526.00	R² 0.799
Ïƒ 42.2446	d.f. 1	R²_adj 0.798
d.f. 326	Pr(>χ²) 0.0000	g 95.935

Residuals

      Min       1Q   Median       3Q      Max 
 -278.344  -16.875    2.436   14.323  274.604

	β	S.E.	t	Pr(>\|t\|)
Intercept	15.1397	5.3455	2.83	0.0049
prism_ffd	0.9407	0.0261	35.98	<0.0001

Frost-Free Days

ffd_50_pct_800m.tif Frost-free days, 50% probability.
ffd_80_pct_800m.tif Frost-free days, 80% probability.
ffd_90_pct_800m.tif Frost-free days, 90% probability.

Day of Last Spring / First Fall Frost

Last spring frost maps represent the Julian day after which the probability of no-frost is 50 / 80 / 90 percent. First fall frost maps represent the Julian day before which the probability of no-frost is 50 / 80 / 90 percent.

last_spring_frost_50_pct_800m.tif Julian day of last spring frost, 50% probability.
first_fall_frost_50_pct_800m.tif Julian day of first fall frost, 50% probability.
last_spring_frost_80_pct_800m.tif Julian day of last spring frost, 80% probability.
first_fall_frost_80_pct_800m.tif Julian day of first fall frost, 80% probability.
last_spring_frost_90_pct_800m.tif Julian day of last spring frost, 90% probability.
first_fall_frost_90_pct_800m.tif Julian day of first fall frost, 90% probability.

Design Freeze Index

q90_freeze_index_800m.tif number of degree days below 0 deg C, 90th percentile
q90_freeze_index_F_800m.tif number of degree days below 0 deg F, 90th percentile

From NSSH Part 618.33 Frost Action, Potential:

Part 618, Subpart B, Exhibits, Section 618.85 is a map that shows the design freezing index values in the continental United States. The values are the number of degree days below 0 deg C for the coldest year in a period of 10 years . The values indicate duration and intensity of freezing temperatures. The 250 isoline is the approximate boundary below which frost action ceases to be a problem.

Methods:

using units of degrees Celsius, and daily average air temperature (\(Tavg\))
freezing degree days for a single year: \(FI = sum( abs( min(0, Tavg) ) )\)
design freezing index, over 30 year record: \(DFI = Q90( FI )\) where FI is the stack of annual FI

Notes:

There is a fairly large difference in where the 250 DFI isoline falls, depending on the temperature units.
The 90th percentile of FI seems to track the notion of “coldest year in 10 years”.
The “average of 3 coldest years in 30” method gives different results, but spatial patterns are the same.
Related conversation on the calculation and interpretation.

Growing Degree Days (C)

gdd_mean_800m.tif

Mean (Celsius) growing degree days, derived from the 800m PRISM daily minimum/maximum temperature data over the interval of 1981–2010.

Calculation reference: http://agron-www.agron.iastate.edu/courses/Agron541/classes/541/lesson02b/2b.1.1.html

\[GDD_i = [ min(T_{max}, upper_{threshold}) + max(Tmin, lower_{threshold}) / 2 ] - T_{base}\]

\[GDD_i = max(GDD_i, 0)\]

Calculations

Effective Precipitation

effective_precipitation_800m.tif

Annual sum of monthly (total) precipitation - monthly (estimated) evapotranspiration, averaged over the interval of 1981–2010. Potential evapotranspiration (PET) estimated via Thornthwaite’s method of 1948. Input sources included:

800m resolution, monthly, total precipitation (PRISM group)
800m resolution, monthly, mean air temperature (PRISM group)

Processing in GRASS GIS.

Fraction of Precipitation as Rain

rain_fraction_mean_800m.tif

This map contains estimates of the fraction of total (annual) precipitation as rain, derived from 800m daily PRISM Tmax and PPT grids (1981–2010). Calculations were performed using GRASS GIS, with methods and estimated parameters of the conditional snow probability function from Rajagopal and Harpold (2016).

Partition PPT into snow/rain:

\[rain = PPT - snow\]

\[snow = PPT * Pr(snow)\]

compute \(Pr(snow)\) as a function of \(Tmax\) using exponential identity for hyperbolic tangent function:

Evaluate conditional probability (fraction) of snow on a daily basis:

\[Pr(snow) = a * ( tanh(b * (Tmax - c) ) - d )\]

a:-0.5, b:0.21, c:0.5, d:1

\[tanh(x) = (1 - exp(-2*x)) / (1 + exp(-2*x))\]

\[Pr(snow) = -0.5 * ( (1 - exp(-2 * (0.21 * (Tmax - 0.5) ))) / (1 + exp(-2 * (0.21 * (Tmax - 0.5) ))) - 1 )\]

\[rain = PPT - (PPT * Pr(snow))\]

For each year(\(i\)):

\[rain fraction_i = sum(rain_i) / sum(PPT_i)\]

Percentages have been converted to integers ranging from 0 to 100.

Rajagopal, S. and A.A. Harpold. 2016. Testing and Improving Temperature Thresholds for Snow and Rain Prediction in the Western United States. Journal of the American Water Resources Association, 52: 1142-1154.

Terrain

Standard Derivatives

10 meter resolution

SSR2_DEM10m_AEA.tif Integer representation of elevation (m).
SSR2_Aspect10m_AEA.tif Integer representation of aspect angle (degrees clock-wise from North).
SSR2_Slope10m_AEA.tif Integer representation of slope gradient (percent).

30 meter resolution

DEM_30m_SSR2.tif Integer representation of elevation (m).
Aspect_30m_SSR2.tif Integer representation of aspect angle (degrees clock-wise from North).
Slope_30m_SSR2.tif Integer representation of slope gradient (percent).

Surface Shape (Curvature Classes)

curvature_classes_10_class_region2.tif 10 meter resolution
curvature_classes_30_class_region2.tif 30 meter resolution

These maps contains surface curvatures (profile and crossectional), derived from the USGS 10m / 30m DEM and grouped into classes: concave, linear, convex.

Profile (down-slope) and crossectional (across-slope) curvatures were calculated via Wood’s method (1996) using a 5x5 moving window in GRASS GIS. Curvatures were classified into concave, linear, and convex using a threshold of +/- 0.0001 (m^-1.

Curvature classes (downslope, across slope) are coded as:

LL | LV | LC
VL | VV | VC
CL | CV | CC

Codes:

22 | 32 | 12
23 | 33 | 13
21 | 31 | 11

See Field Book for Describing and Sampling Soils version 3.0 for details.

Wood, J. (1996): The Geomorphological characterisation of Digital Elevation Models. Diss., Department of Geography, University of Leicester, U.K.

Landform Elements (geomorphons)

forms10_region2.tif 10 meter resolution
forms30_region2.tif 30 meter resolution

These maps were generated using the r.geomorphon GRASS GIS module, with the following parameters:

r.geomorphon --o dem=elev30_int forms=forms30 search=75 skip=5 flat=1.718

The source DEM was a 10m / 30m resolution compilation of USGS NED data, rounded to integers. The “flat” threshold (1.718 deg) is based on a 3% slope break.

Jasiewicz, J., Stepinski, T., 2013, Geomorphons - a pattern recognition approach to classification and mapping of landforms, Geomorphology, vol. 182, 147-156.

Terrain Indices

Annual Beam Radiance

beam_rad_sum_mj30_int_region2.tif

This map describes estimated annual beam radiance (MJ/ square meter / year) at each 30m grid cell, based on the solar radiation algorithm implimented in the r.sun module of GRASS GIS. Horizon maps generated by r.horizon. Default Linke atmospheric turbitity coefficient (3.0) and surface albedo (0.2) values were used.

See this article for a more detailed description of the solar radiation algorithm and an application to soil survey.

Example invocation.

# horizon step size, degrees
step=30

# by tile
r.horizon --q --o elevation=$map step=$step start=$start end=$end output=hzangle

# by tile / day of year
r.sun --quiet --overwrite elevation=$elev aspect=$asp slope=$slope \
day=$day horizon_basename=hzangle horizon_step=$step \
beam_rad=beam.$day_fmt

Saga TWI

saga_twi_30_int_region2.tif

SAGA wetness index, calculated from a 30m DEM.

From the SAGA manual:

The ‘SAGA Wetness Index’ is, as the name says, similar to the ‘Topographic Wetness Index’ (TWI), but it is based on a modified catchment area calculation (‘Modified Catchment Area’), which does not think of the flow as very thin film. As result it predicts for cells situated in valley floors with a small vertical distance to a channel a more realistic, higher potential soil moisture compared to the standard TWI calculation.

Boehner, J., Koethe, R. Conrad, O., Gross, J., Ringeler, A., Selige, T. (2002): Soil Regionalisation by Means of Terrain Analysis and Process Parameterisation. In: Micheli, E., Nachtergaele, F., Montanarella, L. [Ed.]: Soil Classification 2001. European Soil Bureau, Research Report No. 7, EUR 20398 EN, Luxembourg. pp.213-222.

Compount Topographic Index

tci_30_int_region2.tif

This map describes the compount topographic index at each 30m grid cell, based on the r.topidx module of GRASS GIS.

Moore, I.D., R.B. Grayson, and A.R. Ladson, 1991. Digital Terrain Modeling: A Review of Hydrological, Geomorphological, and Biological Applications. Hydrological Processes 5:3-30.

MU Comparison Data Sources

Dylan Beaudette, Jennifer Wood, Russ Almaraz, Andrew Brown

2021-08-17

Overview