GEE: Export yearly median Landsat 5 images

Question

I am trying to download yearly median Landsat 5. Specifically, I am interested in the thermal band (Band 6) and the spectral index called EBBI (formula in the code below).

My initial try, so far is (link to the code I am using) by downloading each month of the year and then calculate the yearly image outside of Google Earth Engine (GEE), in a GIS software. But this takes a huge amount of time and the end shows an error that there are no bands.

How can I export yearly median Landsat 5 images (Thermal and EBBI) in a more efficient way? By efficient I mean something similar to the code I am using to download Landsat 8. As you can see in the code for Landsat 8, all I have to change in order to compute a median monthly image or a median yearly image is to change the range in var landsat = landsat.filter(ee.Filter.calendarRange(1,1,'month')) to var landsat = landsat.filter(ee.Filter.calendarRange(1,12,'month')). And the year of course.

I would like create a similar code for Landsat 5. I have tried to do it but I am getting errors, depending to what parameters I am changing in the Landsat 5 code.

Answer 1

A complete code to download median monthly (or yearly, check the comments) Landsat 5 (surface reflectance, collection 1):

// PARAMETERS
// dates
var YEAR_START = 2011;
var YEAR_END = YEAR_START;
var MONTH_START = 9;
var MONTH_END = MONTH_START;
// export
var CRS = 'EPSG: 7760';
var MAXPIXELS = 1000000000000; // can be replaced with 10e11
var REGION = table;
// 

var landsat = ee.ImageCollection("LANDSAT/LT05/C01/T1_SR");

//Create mask function
function mask_ls_sr_c1(image) {
  // Bits 2, 3 and 5 are water, cloud shadow and cloud, respectively.
  var cloudShadowBitMask = (1 << 3);
  var cloudsBitMask = (1 << 5);
  var waterBitMask = (1 << 2);
  // Get the pixel QA band.
  var qa = image.select('pixel_qa');
  // Both flags should be set to zero, indicating clear conditions.
  var mask = qa.bitwiseAnd(cloudShadowBitMask).eq(0)
                 .and(qa.bitwiseAnd(cloudsBitMask).eq(0))
                 .and(qa.bitwiseAnd(waterBitMask).eq(0));
  return image.updateMask(mask);
}


var landsat = landsat.filter(ee.Filter.calendarRange(YEAR_START, YEAR_END,'year'))
                    .filter(ee.Filter.calendarRange(MONTH_START, MONTH_END, 'month'))
                    // Note 2: when I want to do a yearly composite, I can comment out filtering by month completely
                    .filterBounds(table)
                    .map(mask_ls_sr_c1);


//Calculate the median for each band, multiply by scale factor
//(0.0001), and clip to country polygon
var median1 = landsat.reduce(ee.Reducer.median())
                     .multiply(0.0001)
                     .clip(table);

//Calculate the median for B6, multiply by scale factor
//(0.1), and clip to country polygon
var median2 = landsat.select('B6')
                     .reduce(ee.Reducer.median())
                     .multiply(0.1)
                     .subtract(273.15)
                     .clip(table);
                     
                    
Map.addLayer(median1)
Map.addLayer(median2)

//Create variable for each band
var B1 = median1.select('B1_median')
var B2 = median1.select('B2_median')
var B3 = median1.select('B3_median')
var B4 = median1.select('B4_median')
var B5 = median1.select('B5_median')
var B7 = median1.select('B7_median')
var B6 = median2.select('B6_median')

var ebbi = median1.expression('(SWIR - NIR)/ 10 * sqrt(SWIR + TIRS)',
{
  'NIR':B4,
  'SWIR':B5,
  'TIRS' : B6
});

Export.image.toDrive({
  image: B6,
  description: 'tirs',
  scale: 100, //100 for Band10
  maxPixels: MAXPIXELS,
  region: REGION,
  crs: CRS
});

Export.image.toDrive({
  image: ebbi,
  description: 'ebbi',
  scale: 100, //100 for Band10
  maxPixels: MAXPIXELS,
  region: REGION,
  crs: CRS
});