Published September 28, 2022 © GPL3+

Wireless Monitoring System for GHGs emissions in Paddy Field

The monitoring system can help identify hotspots and provide data to study the effect of different parameters on the emission of GHGs.

IntermediateFull instructions provided8 hours588

Enthusiastic Project Award

IoT Into the Wild Contest for Sustainable Planet 2022

Wireless Monitoring System for GHGs emissions in Paddy Field

Things used in this project

Hardware components

Seeed Studio Wio Terminal

Seeed Studio Grove - I2C High Accuracy Temp&Humi Sensor (SHT35)

DFRobot Gravity: Analog CH4 Gas Sensor (MQ4) For Arduino

Solar Panel, 2.5 W

DFRobot Solar Power Manager (for 9V/12V/18V solar panel)

Power Storage

Waterproof Box

Software apps and online services

Arduino IDE

Blynk

Hand tools and fabrication machines

Jumper Wire Kit, Multicolor

3D Printer (generic)

Hot glue gun (generic)

Story

Paddy field is one of the major sources of greenhouse gases (both CH4 methane and NOx nitrogen oxides) according to some research, however, some recent research suggests that there are miscalculations.

The local researchers collect greenhouse gas fluxes, primarily methane (CH4) and nitrous oxide (N2O) by using a Manually-Operated Closed Chamber in the field in Prachin Buri province, Thailand. (Photo credit: Prachin Buri Rice Research Center)

https://www.asean-agrifood.org/thai-researchers-track-greenhouse-gases-from-rice-field/

Schematic diagram of methane production, oxidation and emission.

Rajkishore, S. K. et al. “METHANE EMISSION FROM RICE ECOSYSTEMS: 100 YEARS OF RESEARCH.” (2015). https://www.semanticscholar.org/paper/METHANE-EMISSION-FROM-RICE-ECOSYSTEMS%3A-100-YEARS-OF-Rajkishore-Vignesh/d080f93f23bf8c23f398243b913a7c0a2cd3273c

To find out, I build an easily deployed CH4 monitor to help rice farmers or other interested citizen scientist to look into this issue. Because rice is the main source of food for most Asian countries, it is important that we acknowledge the problem accurately and respond accordingly. To make things simpler, I only worked on CH4 this time, but I hope that my work could inspire fellow makers to develop the NOx monitor as well. As NOx involves more than one substance and thus they are more difficult to measure.

My original plan was to create an off-grid system making use of solar power and LoRa transmission, but the AS923-2 frequency (which is the standard frequency where I live) is not stably supported by the gateway I bought and The Things Network. After several attempts, I had to give up and switch to WiFi and Blynk.

Before I start I would like to give credit to the many projects before me, they provided me with shortcuts and inspirations to work out this CH4 monitor project, and they are listed in the session below. Thank you all!

1 / 4 • Complete setup with power storage to ensure operation at night

1. Wio Terminal

The Wio Terminal is a SAMD51-based microcontroller with Wireless Connectivity powered by Realtek RTL8720DN that’s compatible with Arduino and MicroPython. Currently, wireless connectivity is only supported by Arduino. It runs at 120MHz (Boost up to 200MHz), 4MB External Flash and 192KB RAM. It supports both Bluetooth and Wi-Fi providing backbone for IoT projects. The Wio Terminal itself is equipped with a 2.4” LCD Screen, onboard IMU(LIS3DHTR), Microphone, Buzzer, microSD card slot, Light sensor, and Infrared Emitter(IR 940nm). On top of that, it also has two multifunctional Grove ports for Grove Ecosystem and 40 Raspberry pi compatible pin GPIO for more add-ons.

https://wiki.seeedstudio.com/Wio-Terminal-Getting-Started

To set up the Arduino environment and upload code, please refer to the Seeed Studio official website.

2. MQ4 Methane Sensor

MQ4 is a cheap, easy-to-use sensor to detect the concentration of Methane. It is a good starting point, but there are also other options such as MQ214,

Here's the datasheet of MQ4: https://www.sparkfun.com/datasheets/Sensors/Biometric/MQ-4.pdf

The sensor runs on 5V and comes with both digital and analog outputs. In this project, I used analog reading to get the actual concentration rather than digital signal, because digital signal only shows whether CH4 is "present" judging by a user-determined threshold. Digital signal is useful for leakage detection but not for measurement.

3. SHT40 temperature and humidity sensor

The sensor comes with the Wio Terminal Development Kit, for more details please refer to the above Seeed Studio website. The development kit comes with other sensors too, here I only included the components that I used.

Tutorial, library and sample code can be found here: https://wiki.seeedstudio.com/Grove-Temp-Humi-Sensor-Grove-LoRa-E5/

This sensor can be plugged in directly into the bottom sockets of the Wio Terminal and adopts I2C protocol.

4. Solar Power System

The solar power system consists of 3 parts: a solar panel, a 5V step-down module and a power storage. The setup is pretty straightforward, the solar panel I bought is 12V, so we need to step down to 5V for the Wio Terminal and sensors. The power storage is basically a powerbank with switch. And all we have to do is connect them together and plug in the USB.

I bought the 3 components from Taobao, for those of you who don't understand Chines, you can find them on AliExpress. I'm sure there are sellers for solar power systems in your region, so I won't include details of the suppliers

5. Connection

The connection for solar power system is explained above. The SHT40 sensor can be plugged in directly to Wio Terminal, while MQ4 sensor is a bit more difficult. We will connect the 5V, GND, and Analog Output to the corresponding pins; pin4, pin6 and pin32 among the 40 pins on the back of Wio Terminal. The pinout is copied here for your reference:

Connect the 5V, GND and Analog Output pin of MQ4 to pin4, pin6 and pin32 on Wio Terminal respectively.

For more details, please scroll down and refer to the Schematic below.

6. Arduino Code

Just like every Arduino code, it has definition, setup() and loop(). The code can be further divided into 4 parts: take measurements, display the numbers on the Wio Terminal screen, upload to Blynk and calibration of R0 value.

Measurement

  error = sht4x.measureHighPrecision(temperature, humidity);
  if (error) {
    errorToString(error, errorMessage, 256);
    Blynk.notify(errorMessage);
  }   
  rawCH4 = analogRead(gasPin);
  senVol = rawCH4 * 5 / 1023.0;
  senRes = ((5.0 * 10.0) / senVol) - 10.0; //Calculate RS in fresh air
  ratio = senRes / R0;
  ppm_log = m * log10(ratio) + C;

The first 5 lines of code read temperature and humidity from SHT40 sensor, and the later part reads analog signal (voltage level) from MQ4 sensor and then it's converted to concentration based on the graph in the datasheet.

The first step of conversion maps the 0-5V to 0-1023, then the sensor resistance is calculated based on the 0-1023 value. Finally, the sensor resistance is mapped onto the curve in the graph. By basic algebra formula: y = mx +c, we can find the "y" which is the concentration ppm of CH4.

R0 Calibration

  double sensor_volt; //Define variable for sensor voltage
  double RS_air; //Define variable for sensor resistance
  double sensorValue; //Define variable for analog readings
  for (int x = 0 ; x < 500 ; x++) //Start for loop
  {
    sensorValue = sensorValue + analogRead(A5); //Add analog values of sensor 500 times
  }
  sensorValue = analogRead(gasPin); //sensorValue / 500.0; //Take average of readings
  sensor_volt = sensorValue * (5.0 / 1023.0); //Convert average to voltage
  RS_air = ((5.0 * 10.0) / sensor_volt) - 10.0; //Calculate RS in fresh air
  R0 = RS_air / 4.4; //Calculate R0

One thing to note is the R0 value as the y-axis of the curve is actually Rs/R0. R0 is found by reading the sensor value in fresh air, converting it to sensor resistance and dividing it by 4.4 (the number is found in Fig. 2 of the datasheet, the curve of "Air"), similar to a calibration process. It should be done before entering the testing field. I've included the calibration code as CH4_calib() function, please uncomment it in the setup() when needed.

Display on Wio Terminal

For the layout of display, I like the design by Salman Faris a lot and thus took reference. I only adjusted the width of the rectangles and increase the decimal places to display.

Upload to Blynk

This is also very easy thanks to the Blynk library, you just need to send:

  Blynk.virtualWrite(V0, temperature);
  Blynk.virtualWrite(V1, humidity);
  Blynk.virtualWrite(V2, ppm_log);
  Blynk.virtualWrite(V3, R0);

7. Blynk Dashboard

This tutorial basically explains everything: https://wiki.seeedstudio.com/Wio-Terminal-Blynk/

Just add the corresponding display to the different parameters, e.g. digital label for temperature, charts for Methane concentration. Please refer to the template below.

Template of dashboard on Blynk

8. Casing

Here I used a waterproof box to put everything in, but Mithun Das has designed a very useful casing, which I suggest using his design to mount the Wio Terminal inside the box. The power storage should be smaller, but that's what I currently have in hand.

I popped holes underneath the box to place the sensors there, a PVC plastic tube is used as a supporting rod to make the box stand. Hot glue comes in handy to glue them in place.

1 / 2

9. Results

Running the monitor in my office for the whole day, it's pretty stable at below 2ppm.

The concentration of CH4 I measured at home is 1.70ppm, which is acceptably close to the latest global value of around 1.80ppm. The R0 value was calibrated at home at approx. 30dec C and 65% R.H., I included the two measurements in the code to help with further finetuning of the Methane concentration with respect to the datasheet graphs.

Please also need that the MQ4 sensor usually gives a "spike" (something like 3-4ppm) each time it is powered up, it needs a "warm-up" of around 30 minutes before the reading stabilises to around 2ppm.

With everything ready, I brought my setup to one of the few paddy fields left in my city and see how the reading goes. Unfortunately, it's already September and they just harvested.

Paddy field managed by a local NGO with an idea to restore Yellow-breasted Bunting (Emberiza aureola) population.

However, I found a new Wetland Park nearby which is currently "under construction" (Yes, a constructed wetland.... That is so Hong Kong...) When it opens later this year, I think I will come again and place the Monitoring Kit at both sites and see how the concentration of a paddy field compared to that of a wetland.

I understand that this simplified methane monitoring might not be scientific enough, but I hope that this is more like an inspiration and an entry-level kit for interested farmers or citizen scientists to start monitoring the concentration of methane in their surrounding environment. For more references on the methane generation in the paddy field and how to measure more scientifically, please refer to the following, and I'm sure there's a lot more on the Internet:

Handbook of Monitoring, Reporting, and Verification for a Greenhouse Gas Mitigation Project with Water Management in Irrigated Rice Paddies: https://www.naro.go.jp/publicity_report/publication/files/MRV_guidebook.pdf

Estimation of methane emission rates from a rice paddy field using a naturally ventilated tunnel: https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1744-697X.2012.00249.x

Emissions Of Greenhouse Gases From Rice Agriculture: https://www.osti.gov/servlets/purl/959124

Code

Arduino code for Methane Monitoring System

// 18/9/2022 Arduino code for Methane Monitoring System by James Lam
// With help from Desmond Wong on CH4 concentration calculation
// WIth curtesy to the following authors on "uploading to Blynk" and "display layout"
/*
  Air Quality Terminal
  Author: Salman Faris
  Date: 01/09/2020
  Last Updates: 06/09/2020
  https://www.hackster.io/Salmanfarisvp/air-quality-monitoring-station-with-wio-terminal-6ef85f

  https://community.element14.com/challenges-projects/project14/sensors/b/blog/posts/wio-terminal-sensor-fusion---remote-data-display-and-control-using-blynk
  Wio Terminal Sensor Fusion - Remote Data Display and Control using Blynk
  by ralphjy 2020/11/9
*/
#define BLYNK_PRINT Serial

#define BLYNK_TEMPLATE_ID "your-blynk-template-id"
#define BLYNK_DEVICE_NAME "your-blynk-device-id"
#define BLYNK_AUTH_TOKEN "your-blynk-authentication-token"

#include <rpcWiFi.h>
#include <WiFiClient.h>
#include <BlynkSimpleWioTerminal.h>

#include <TFT_eSPI.h>
#include <Wire.h>
#include <SensirionI2CSht4x.h>
SensirionI2CSht4x sht4x;

#include <math.h>

#if defined(ARDUINO_ARCH_AVR)
    #define debug  Serial

#elif defined(ARDUINO_ARCH_SAMD) ||  defined(ARDUINO_ARCH_SAM)
    #define debug  SerialUSB
#else
    #define debug  Serial
#endif

// Your WiFi credentials.
const char* ssid = "your-wifi-ssid";
const char* pass = "your-wifi-password";

// You should get Auth Token in the Blynk App.
// Go to the Project Settings (nut icon).
char auth[] = BLYNK_AUTH_TOKEN;

const int gasPin = A5;
const double C = 2.94336;
const double m = -1.94336;
double R0 = 58.91;  //should be calibrated in fresh air
double rawCH4;
double senVol;
double senRes;
double ratio;
double ppm_log;

TFT_eSPI tft;
TFT_eSprite spr = TFT_eSprite(&tft);  //sprite

BlynkTimer timer;

//Temp & humi
uint16_t error;
char errorMessage[256];
float temperature;
float humidity;

void setup() {
  Serial.begin(115200);
  tft.begin();
  tft.setRotation(3);

  //  initiate SHT40
  Wire.begin();

  uint16_t error;
  char errorMessage[256];

  sht4x.begin(Wire);

  uint32_t serialNumber;
  error = sht4x.serialNumber(serialNumber);
  if (error) {
    errorToString(error, errorMessage, 256);
    Blynk.notify(errorMessage);
  }

  //Head
  tft.fillScreen(TFT_BLACK);
  tft.setFreeFont(&FreeSansBoldOblique18pt7b);
  tft.setTextColor(TFT_WHITE);
  tft.drawString("Methane Monitor", 20, 10 , 1);

  //Line
  for (int8_t line_index = 0; line_index < 5 ; line_index++)
  {
    tft.drawLine(0, 50 + line_index, tft.width(), 50 + line_index, TFT_GREEN);
  }

  //CH4 Rect
  tft.drawRoundRect(5, 60, (tft.width() / 2) - 20 , tft.height() - 65 , 10, TFT_WHITE); // L1

  //R0 Text
  tft.setFreeFont(&FreeSansBoldOblique12pt7b);
  tft.setTextColor(TFT_RED);
  tft.drawString("R0", 7 , 65 , 1);

  //CH4 Text
  tft.setFreeFont(&FreeSansBoldOblique12pt7b);
  tft.setTextColor(TFT_RED);
  tft.drawString("CH4", 7 , 150 , 1);
  tft.setTextColor(TFT_GREEN);
  tft.drawString("ppm", 80, 193, 1);

  // Temp rect
  tft.drawRoundRect((tft.width() / 2) - 10  , 60, (tft.width() / 2) , (tft.height() - 65) / 2 , 10, TFT_BLUE); // s1

  tft.setFreeFont(&FreeSansBoldOblique9pt7b);
  tft.setTextColor(TFT_RED) ;
  tft.drawString("Temp", (tft.width() / 2) - 1  , 70 , 1); // Print the test text in the custom font
  tft.setTextColor(TFT_GREEN);
  tft.drawString("o", (tft.width() / 2) + 70, 95, 1);
  tft.drawString("C", (tft.width() / 2) + 80, 100, 1);

  //Humi Rect
  tft.drawRoundRect((tft.width() / 2) - 10 , (tft.height() / 2) + 30, (tft.width() / 2) , (tft.height() - 65) / 2 , 10, TFT_BLUE); // s3

  tft.setFreeFont(&FreeSansBoldOblique9pt7b);
  tft.setTextColor(TFT_RED) ;
  tft.drawString("Humi", (tft.width() / 2) - 1 , (tft.height() / 2) + 40 , 1); // Print the test text in the custom font
  tft.setTextColor(TFT_GREEN);
  tft.drawString("%", (tft.width() / 2) + 70, (tft.height() / 2) + 70, 1);

  Blynk.begin(auth, ssid, pass);
  timer.setInterval(1000L, takeMeasurement);
  timer.setInterval(1000L, Display);  
  timer.setInterval(1000L, sendtoBlynk);  
  
  //uncomment to calibrate R0, please calibrate in fresh air
  //  CH4_calib();
}

void loop() {
    
  Blynk.run();
  timer.run();
 
}

void CH4_calib ()
{
  double sensor_volt; //Define variable for sensor voltage
  double RS_air; //Define variable for sensor resistance
  double sensorValue; //Define variable for analog readings
  for (int x = 0 ; x < 500 ; x++) //Start for loop
  {
    sensorValue = sensorValue + analogRead(A5); //Add analog values of sensor 500 times
  }
  sensorValue = analogRead(A5); //sensorValue / 500.0; //Take average of readings
  sensor_volt = sensorValue * (5.0 / 1023.0); //Convert average to voltage
  RS_air = ((5.0 * 10.0) / sensor_volt) - 10.0; //Calculate RS in fresh air
  R0 = RS_air / 4.4; //Calculate R0
}

void sendtoBlynk()
{
    // Send to Blynk
  Blynk.virtualWrite(V0, temperature);
  Blynk.virtualWrite(V1, humidity);  
  Blynk.virtualWrite(V2, ppm_log);  
  Blynk.virtualWrite(V3, R0);
}

void takeMeasurement()
{
  //measure temperature and humidity
  error = sht4x.measureHighPrecision(temperature, humidity);
  if (error) {
    errorToString(error, errorMessage, 256);
    Blynk.notify(errorMessage);
  } 
  //measure CH4
  rawCH4 = analogRead(gasPin);
  senVol = rawCH4 * 5 / 1023.0;
  senRes = ((5.0 * 10.0) / senVol) - 10.0; //Calculate RS in fresh air
  ratio = senRes / R0;
  ppm_log = m * log10(ratio) + C;
}
  
void MeasureDisplay()
{
  error = sht4x.measureHighPrecision(temperature, humidity);
  if (error) {
    Serial.print("Error trying to execute measureHighPrecision(): ");
    errorToString(error, errorMessage, 256);
    Serial.println(errorMessage);
  } else {
    Serial.print("Temperature:");
    Serial.print(temperature);
    Serial.print("\t");
    Serial.print("Humidity:");
    Serial.println(humidity);
  }

  spr.createSprite(60, 30);
  spr.setFreeFont(&FreeSansBoldOblique12pt7b);
  spr.setTextColor(TFT_WHITE);
  spr.drawFloat(temperature, 2, 0, 0, 1);
  spr.setTextColor(TFT_GREEN);
  spr.pushSprite((tft.width() / 2) - 1, 100);
  spr.deleteSprite();

  spr.createSprite(60, 30);
  spr.setFreeFont(&FreeSansBoldOblique12pt7b);
  spr.setTextColor(TFT_WHITE);
  spr.drawFloat(humidity, 2, 0, 0, 1);
  spr.pushSprite((tft.width() / 2) - 1, (tft.height() / 2) + 67);
  spr.deleteSprite();

  //  R0
  spr.createSprite(70, 30);
  spr.fillSprite(TFT_BLACK);
  spr.setFreeFont(&FreeSansBoldOblique12pt7b);
  spr.setTextColor(TFT_WHITE);
  spr.drawFloat(R0, 2, 0, 0, 1);
  spr.pushSprite(15, 100);
  spr.deleteSprite();

  //  CH4
  spr.createSprite(60, 30);
  spr.setFreeFont(&FreeSansBoldOblique12pt7b);
  spr.setTextColor(TFT_WHITE);
  spr.drawFloat(ppm_log, 2, 0, 0, 1);
  spr.setTextColor(TFT_GREEN);
  spr.pushSprite(15, 185);
  spr.deleteSprite();
}