---
title: "🌱💧 Design ETo calculation"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{🌱💧 Design ETo calculation}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

## 🚀 Design ETo calculation estimation

This article shows a quick example of how to download INMET station data and estimate reference evapotranspiration (ETo) using FAO-56, followed by the calculation of the design ETo.

## 📦 Load the package

```{r}
library(BrazilMet)
```

## 🌍 View available INMET stations

Before downloading data, you can check the available weather stations with:

```{r}
see_stations_info()
```

## ⬇️ Download daily weather data

Let’s download daily meteorological data for one station between January 2000 and March 2025 (the station A001 started operating in May 2000):

```{r download, eval = FALSE}
df <- download_AWS_INMET_daily(
  stations   = "A001",
  start_date = "2000-01-01",
  end_date   = "2025-03-31"
)
```

The resulting data frame includes temperature, solar radiation, wind speed, humidity, and atmospheric pressure.

To keep this article reproducible without depending on the INMET server, the data downloaded with the call above are bundled with the package and loaded here:

```{r load-data}
df <- readRDS(system.file("extdata", "A001_daily_2000_2025.rds", package = "BrazilMet"))
df$date <- as.Date(df$date)
```

## 🧩 Fill gaps in the weather data

A long series has many sensor failures, and any missing input makes ETo `NA` on that day. `fill_gaps()` fills short gaps (up to three days) by linear interpolation and the remaining ones with the mean of the same day of the year in the other years. Every filled value is flagged in a `*_filled` column:

```{r gap-fill}
df <- fill_gaps(df, method = "both", max_gap = 3)
```
## 🧠 Calculate daily ETo using FAO-56

Now we use the daily_eto_FAO56() function to estimate daily ETo values:

```{r}
df$eto <- daily_eto_FAO56(
  lat    = df$latitude_degrees,
  tmin   = df$tair_min_c,
  tmax   = df$tair_max_c,
  tmean  = df$tair_mean_c,
  Rs     = df$sr_mj_m2,
  u2     = df$ws_2_m_s,
  Patm   = df$patm_mb,
  RH_max = df$rh_max_porc,
  RH_min = df$rh_min_porc,
  z      = df$altitude_m,
  date   = df$date
)
```

## 💧 Design ETo calculation

And after the ETo calculation, we use the design_eto() function to estimate the design ETo for irrigation project purpose:

```{r}
# Ensure date column is in Date format
df$date <- as.Date(df$date)

eto_design <- BrazilMet::design_eto(eto_daily_data = df, percentile = .80)

```


## 📝 Printing the design ETo based on an 80% probability of occurrence

Below is a basic line plot of daily ETo:

```{r plot-eto-ggplot, fig.width = 10, fig.height = 4}

print(eto_design)

```

## ✅ Summary

The BrazilMet package allows you to download official INMET weather data and compute ETo using the FAO-56 method in a reproducible and efficient way. This is essential for irrigation planning, crop modeling, and climate-based decision support.


## 🔗 Useful links

https://github.com/FilgueirasR/BrazilMet