seem (version 1.0)

Weather: Weather simulation

Description

Rain simulation and PET calculations

Usage

rain.day(param)
markov.rain.seq(P, ndays)
markov.rain(P, ndays, amount.param, plot.out = T)
pet(eleva, Rad, Temp, RH, wind, plot.out = F)

Arguments

param

parameters: a list that contains the mean, standard deviation, skew coefficient, shape, and the distribution to be sampled

P

Markov transition probability matrix

ndays

number of days to simulate

amount.param

same as param above

plot.out

logical variable to decide whether to plot

eleva

elevation amsl

Rad

solar radiation

Temp

air temperature

RH

relative humidity

wind

wind speed

Value

• rain.day: amount of rain in a day markov.rain.seq:
• x
  amount of rain in a day
• wet.days
  number of days with rain
• expected.wet.days
  expected number of days with rain
• dry.days
  number of days without rain
• expected.dry.days
  expected number of days without rain
• markov.rain:
• x
  amount of rain in a day
• wet.days
  number of days with rain
• expected.wet.days
  expected number of days with rain
• dry.days
  number of days without rain
• expected.dry.days
  expected number of days without rain
• rain.tot
  total amount of rain
• expec.rain.tot
  expected total amount of rain
• rain.avg
  average amount of rain
• expected.rain.avg
  expected average amount of rain
• rain.wet.avg
  average amount of rain for rainy days only
• expected.wet.avg
  expected average amount of rain for rainy days only
• pet: data.frame(Rad,PETR.taylor,PET.penman,PETR.penman,PETA.penman)

Details

rain.day: This function uses one argument, which is composed of parameters, to generate the amount of rain in a day and it is given as a list. We call the function rain.day once we determine that the current day is rainy. This function allows to select from a set of distributions: exponential, Weibull, gamma, and skewed.

markov.rain.seq: This function implements a Markov chain for rainfall generation. Its arguments include the transition probability of the Markov chain and the number of days. This function is the first step of markov.rain.

markov.rain: This function implements two-step rainfall generation. Its arguments include the transition probability of the Markov chain and the number of days, which are used for the first step. In addition, one argument is the list of amount parameters, which is passed internally to the rain.day function described earlier.

The function pet calculates the Priestley-Taylor model as a function of radiation values under various temperature values. It also calculates the Penman model (radiation term, aerodynamic term, and total) as a function of radiation for various temperature values and for fixed RH and wind speed conditions.

References

Acevedo M.F. 2012. Simulation of Ecological and Environmental Models. CRC Press.

Neitsch, S. L., J. G. Arnold, J. R. Kiniry, J. R. Williams, and K. W. King. 2002. Soil and Water Assessment Tool. Theoretical Documentation Version 2000. Temple, TX: Grassland, Soil And Water Research Laboratory, Agricultural Research Service.

Richardson, C. W., and A. D. Nicks. 1990. Weather generator description. In EPIC-Erosion/Productivity Impact Calculator: 1 Model Documentation, eds. A. N. Sharpley, and J. R. Williams, 93-104. Durant, OK and Temple, TX: United States Department of Agriculture, Agricultural Research Center. Technical Bulletin No: 1768.

Williams, J. R., C. A. Jones, and P. T. Dyke. 1990. The EPIC model. In EPIC-Erosion/Productivity Impact Calculator: 1 Model Documentation, eds. A. N. Sharpley, and J. R. Williams, 3-92. Durant, OK and Temple, TX: United States Department of Agriculture, Agricultural Research Center, USDA Technical Bulletin No: 1768.

See Also

Soil water functions e.g., infilt.rate, Solar functions e.g., sun.rad.hr

Examples

amount.param=list(mu=5,std=8,skew=2, shape=1.3, model.pdf ="w")
ndays=30
# rainy followed by rainy
P <- matrix(c(0.4,0.2,0.6,0.8), ncol=2, byrow=T)
rainy1 <- markov.rain(P, ndays, amount.param)
mtext(side=3,line=-1,paste("Rainy after rainy Pr=0.80","Prop rainy=",round(rainy1$wet.days/ndays,2)),cex=0.8)

pet.test <- pet(0, Rad =seq(10,30), Temp=c(10,20,30), RH=70, wind=2, plot=T)