---
title: "Article 5: Custom input land units"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteEngine{knitr::knitr}
  %\VignetteIndexEntry{Article 5: Custom input land units}
  %\usepackage[UTF-8]{inputenc}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```
ALUES comes with two datasets for input land units. These are the land units of Marinduque, Philippines; and, Lao Cai, Vietnam. However, users will likely have their own region of interest. This article will illustrate how to prepare such data.

For any region of interest, users must specify the properties of the land units into three categories:

- terrain and soil
- water
- temperature

Note that both terrain and soil factors must be specified as one dataframe, not separate. Suppose there are 5 land units under study, the three characteristics can be specified as follows:

## Terrain and soil
Suppose for terrain, we want to target the following factors: Flood and SlopeD; and, suppose for soil we have CFragm and SoilDpt. The dataframe can be prepared as follows:
```{r}
terrain_input <- data.frame(
  Flood = c(1, 2, 2, 2, 3),
  SlopeD = c(3, 4, 5, 1, 2),
  CFragm = c(10, 30, 50, 60, 40),
  SoilDpt = c(45, 60, 90, 70, 30)
)
```
**Note: the column names must be the same with the naming convention used by the crop requirements datasets.**

So that, if this input is assessed for avocado, then the suitability score for these land units are computed as follows:
```{r}
library(ALUES)
AVOCADOTerrain
AVOCADOSoil
avocado_suit <- suit("avocado", terrain=terrain_input)
head(avocado_suit[["terrain"]][["Suitability Score"]])
head(avocado_suit[["terrain"]][["Suitability Class"]])
head(avocado_suit[["soil"]][["Suitability Score"]])
head(avocado_suit[["soil"]][["Suitability Class"]])
```

## Water
For water characteristics, suppose the average rainfall for 3 land units were recorded for four months with the following data:
```{r}
water_input <- data.frame(
  Apr = c(150, 140, 120),
  May = c(70, 90, 100),
  Jun = c(85, 90, 105)
)
water_input
```
Note that when specifying the factors for water chacteristics, the month must be specified in three characters (correct case) only, that is, it shouldn't be specified as January, February, etc. 

The suitability scores for rainfed bunded rice water requirement are computed as follows:
```{r}
RICEBRWater
water_suit <- suit("ricebr", water=water_input, sow_month=1)
water_suit
```
Setting the `sow_month=1` indicates that the factors for `RICEBRWater`'s `WmAv1` correspond to January, `WmAv2` to February, `WmAv3` to March, and `WmAv4` to April. Thus, the only factors that were targetted by the `water_input` is April. So that, setting the `sow_month=3`, would make `WmAv1` of `RICEBRWater` as March, `WmAv2` as April, etc. This in turn targets the months April to Jun.
```{r}
water_suit <- suit("ricebr", water=water_input, sow_month=3)
water_suit
```

## Temperature
Another characteristics that can be targetted is the temperature.
```{r}
temp_input <- data.frame(
  Sep = c(34.2, 35.5, 33.4),
  Oct = c(32.5, 34.2, 32.0),
  Nov = c(30.3, 32.2, 31.1)
)
RICEBRTemp
temp_suit <- suit("ricebr", temp=temp_input, sow_month=9)
```
The `RICEBRTemp` crop requirement has factor `TmAv2`, which is the mean temperature for the 2nd month. Thus, setting the sowing month to 9 suggest that the sowing month started at September, and thus sets `TmAv2` to October. Hence, the factor that was targetted by the input land units is the October as seen below:
```{r}
temp_suit
```