diff --git a/imap_processing/cdf/config/imap_constant_attrs.yaml b/imap_processing/cdf/config/imap_constant_attrs.yaml
index 5ca7d059f3..86dc67edfc 100644
--- a/imap_processing/cdf/config/imap_constant_attrs.yaml
+++ b/imap_processing/cdf/config/imap_constant_attrs.yaml
@@ -21,6 +21,18 @@ epoch:
   VALIDMIN: 315576066184000000   # 2010-01-01T00:00:00 mission start (APL 0 epoch)
   VAR_TYPE: support_data
 
+epoch_delta:
+  CATDESC: Epoch delta
+  DEPEND_0: epoch
+  FIELDNAM: Epoch Delta
+  FILLVAL: -9223372036854775808
+  FORMAT: I19
+  LABLAXIS: Epoch delta
+  SCALETYP: linear
+  UNITS: ns
+  VALIDMAX: 86000000000000
+  VALIDMIN: 0
+  VAR_TYPE: support_data
 
 # <=== Data Variables ===>
 # Default Attrs for all metadata variables unless overridden
diff --git a/imap_processing/cdf/config/imap_hit_l2_variable_attrs.yaml b/imap_processing/cdf/config/imap_hit_l2_variable_attrs.yaml
index 9fe72e117c..4ab012e52a 100644
--- a/imap_processing/cdf/config/imap_hit_l2_variable_attrs.yaml
+++ b/imap_processing/cdf/config/imap_hit_l2_variable_attrs.yaml
@@ -48,6 +48,10 @@ default_angle_attrs: &default_angle
   VAR_TYPE: support_data
 
 # <=== Coordinates ===>
+epoch_macropixel:
+  DELTA_MINUS_VAR: epoch_delta
+  DELTA_PLUS_VAR: epoch_delta
+
 zenith:
   <<: *default_angle
   CATDESC: Angle from the spin axis (0 deg.) to anti-spin axis (180 deg.) in 8 bins
diff --git a/imap_processing/hit/l2/hit_l2.py b/imap_processing/hit/l2/hit_l2.py
index 8c2230b214..557a82be3b 100644
--- a/imap_processing/hit/l2/hit_l2.py
+++ b/imap_processing/hit/l2/hit_l2.py
@@ -142,6 +142,10 @@ def add_cdf_attributes(
                 ),
             )
             dataset.coords[f"{dim}_label"] = label_array
+        elif "macropixel" in logical_source:
+            dataset["epoch"].attrs.update(
+                attr_mgr.get_variable_attributes("epoch_macropixel", check_schema=False)
+            )
 
     return dataset
 
@@ -782,4 +786,69 @@ def process_macropixel_intensity(
                 {var: f"{var}_macropixel_intensity"}
             )
 
-    return macropixel_intensity_dataset
+    return transform_to_10_minute_chunks(macropixel_intensity_dataset)
+
+
+def transform_to_10_minute_chunks(macropixel_dataset: xr.Dataset) -> xr.Dataset:
+    """Transform macropixel records into 10-minute integration chunks.
+
+    Parameters
+    ----------
+    macropixel_dataset : xarray.Dataset
+        Macropixel data containing one species and energy combination per
+        one-minute epoch.
+
+    Returns
+    -------
+    xarray.Dataset
+        Macropixel data combined into one record per 10-minute integration.
+    """
+    species_energy = [
+        ("h", 3),
+        ("he4", 2),
+        ("cno", 2),
+        ("nemgsi", 2),
+        ("fe", 1),
+    ]
+
+    # Use the first record in each 10-record group as the output template.
+    transformed_dataset = macropixel_dataset.isel(
+        epoch=slice(None, None, 10),
+    ).copy(deep=True)
+
+    # Each minute in a 10-record group contains one species/energy combination,
+    # ordered as described by species_energy. Track that minute's packet offset.
+    species_i = 0
+    for species, num_energy_levels in species_energy:
+        energy_dim = f"{species}_energy_mean"
+        # Gather the intensity and uncertainty variables that share this
+        # species' energy dimension.
+        species_variables = [
+            var
+            for var in macropixel_dataset.data_vars
+            if macropixel_dataset[var].dims[:2] == ("epoch", energy_dim)
+        ]
+
+        for energy_i in range(num_energy_levels):
+            for var in species_variables:
+                # Select this species/energy packet from every 10-record group
+                # and place it in the corresponding output energy plane.
+                data_i = macropixel_dataset[var].values[species_i::10, energy_i]
+                transformed_dataset[var].values[:, energy_i] = data_i
+            species_i += 1
+
+    minute_cadence_epochs = macropixel_dataset["epoch"].values
+    ten_minute_cadence_epochs = minute_cadence_epochs.reshape(-1, 10)
+    nanoseconds_per_10_min = SECONDS_PER_10_MIN * 1_000_000_000
+    nanoseconds_per_5_min = nanoseconds_per_10_min // 2
+    start_times = ten_minute_cadence_epochs[:, 0]
+    end_times = ten_minute_cadence_epochs[:, -1]
+    new_epochs = start_times + (end_times - start_times) // 2 - nanoseconds_per_10_min
+
+    transformed_dataset = transformed_dataset.assign_coords(epoch=np.array(new_epochs))
+    transformed_dataset["epoch_delta"] = xr.DataArray(
+        np.full(len(new_epochs), nanoseconds_per_5_min, dtype=np.int64),
+        dims=["epoch"],
+    )
+
+    return transformed_dataset
diff --git a/imap_processing/tests/hit/test_hit_l2.py b/imap_processing/tests/hit/test_hit_l2.py
index bbc11768eb..facb82f5cb 100644
--- a/imap_processing/tests/hit/test_hit_l2.py
+++ b/imap_processing/tests/hit/test_hit_l2.py
@@ -34,6 +34,7 @@
     process_standard_intensity,
     process_summed_intensity,
     reshape_for_sectored,
+    transform_to_10_minute_chunks,
 )
 
 EXPECTED_STANDARD_LABLAXIS = {
@@ -793,6 +794,76 @@ def test_process_macropixel_intensity(
         )
 
 
+def test_transform_to_10_minute_chunks():
+    """Test that transform_to_10_minute_chunks correctly regroups one-minute
+    macropixel records into 10-minute chunks and re-centers the epoch."""
+    n_minutes = 20
+    minute_ns = 60_000_000_000
+    epochs = np.arange(n_minutes, dtype=np.int64) * minute_ns
+
+    # Each species/energy combination occupies one fixed minute slot within
+    # every 10-record group, in this order (mirrors the physical packet
+    # cadence handled by transform_to_10_minute_chunks).
+    species_energy = [
+        ("h", 3),
+        ("he4", 2),
+        ("cno", 2),
+        ("nemgsi", 2),
+        ("fe", 1),
+    ]
+
+    data_vars = {}
+    slot_for = {}
+    species_i = 0
+    for species, num_energy_levels in species_energy:
+        energy_dim = f"{species}_energy_mean"
+        values = np.array(
+            [[m * 100 + e for e in range(num_energy_levels)] for m in range(n_minutes)],
+            dtype=np.float32,
+        )
+        data_vars[f"{species}_macropixel_intensity"] = (("epoch", energy_dim), values)
+        for energy_i in range(num_energy_levels):
+            slot_for[(species, energy_i)] = species_i
+            species_i += 1
+
+    macropixel_dataset = xr.Dataset(data_vars, coords={"epoch": epochs})
+
+    result = transform_to_10_minute_chunks(macropixel_dataset)
+
+    n_chunks = n_minutes // 10
+    assert len(result["epoch"]) == n_chunks
+
+    # epoch_delta is always half of a 10-minute chunk.
+    expected_epoch_delta = np.full(
+        n_chunks, SECONDS_PER_10_MIN * 1_000_000_000 // 2, dtype=np.int64
+    )
+    np.testing.assert_array_equal(result["epoch_delta"].values, expected_epoch_delta)
+
+    # Each new epoch is centered on its 10-minute group, then shifted back by
+    # a full 10-minute chunk.
+    for chunk in range(n_chunks):
+        start = epochs[chunk * 10]
+        end = epochs[chunk * 10 + 9]
+        expected_epoch = start + (end - start) // 2 - SECONDS_PER_10_MIN * 1_000_000_000
+        assert result["epoch"].values[chunk] == expected_epoch
+
+    # Each species/energy variable should pull its value from the minute slot
+    # it physically occupies within every 10-record group.
+    for species, num_energy_levels in species_energy:
+        var = f"{species}_macropixel_intensity"
+        for energy_i in range(num_energy_levels):
+            slot = slot_for[(species, energy_i)]
+            expected = np.array(
+                [(chunk * 10 + slot) * 100 + energy_i for chunk in range(n_chunks)],
+                dtype=np.float32,
+            )
+            np.testing.assert_array_equal(
+                result[var].values[:, energy_i],
+                expected,
+                err_msg=f"Mismatch for {var} energy index {energy_i}",
+            )
+
+
 def test_process_summed_intensity(l1b_summed_rates_dataset, ancillary_dependencies):
     """Test the variables in the summed intensity dataset"""