Some changes made to make this work with the most recent version of sitemap_generator. Also, changed from using an s3.yml file to just using the Heroku S3 config variables directly.

sitemap_generator.rake

require 'aws'

class Rake::Task
  def replace &block
    @actions.clear
    prerequisites.clear
    enhance &block
  end
end

namespace 'sitemap' do
  desc 'Upload the sitemap files to S3'
  task :upload_to_s3 => :environment do
    # Load credentials
    bucket_name = ENV['S3_BUCKET_NAME']
    puts "BUCKET: #{bucket_name}"

    # Establish S3 connection
    AWS.config({
      :access_key_id => ENV['AWS_ACCESS_KEY_ID'],
      :secret_access_key => ENV['AWS_SECRET_ACCESS_KEY']
    })

    Dir.entries(File.join(Rails.root, "tmp", "sitemaps")).each do |file_name|
      next if ['.', '..'].include? file_name
      path = "sitemaps/#{file_name}"
      file = File.join(Rails.root, "tmp", "sitemaps", file_name)

      begin
        s3 = AWS::S3.new
        bucket = s3.buckets.create(bucket_name)


        object = bucket.objects[path]
        object.write(:file => file)
      rescue Exception => e
        raise
      end
      puts "Saved #{file_name} to S3"
    end
  end
end

Rake::Task["sitemap:create"].enhance do
  Rake::Task["sitemap:upload_to_s3"].invoke
end

Rake::Task[:'sitemap:refresh'].replace do
  bucket_name = ENV['S3_BUCKET_NAME']
  SitemapGenerator::Sitemap.ping_search_engines("https://#{bucket_name}.s3.amazonaws.com/sitemaps/sitemap_index.xml.gz")
end

Thank you for posting this edit to rorra's work. It was exactly what I needed!