Thesis preparation

Design and Development of a Handheld LiDAR System for Calamansi (Citrus microcarpa) Biomass Estimation

Full title of the thesis (note: title does not say "carbon stock" or "validation")

6 October 2026

Date of the outline defense

The instrument is the contribution. Biomass and carbon are the demonstration.

Rule of the whole book (the one sentence that prevents overclaiming)

Validated whole-tree biomass

Phrase that is ALWAYS wrong in this thesis

Branch-informed, internally bounded estimate

What every whole-tree AGWB number in the thesis must be called

Overclaiming

The fastest way to lose a panel, according to the textbook

Lecture 0, 2, 22, 9, 17, 16, then the rest

Recommended study order when the defense is close

L1 Recall

Mastery level: state the numbers, names, definitions, who said what

L2 Explain

Mastery level: say why every choice was made

L3 Apply

Mastery level: compute, and apply rules to new cases

L4 Defend

Mastery level: survive a hostile panel question

L1 Recall, L2 Explain, L3 Apply, L4 Defend

The four mastery levels of /master

The manuscript

The authority if the textbook and manuscript ever disagree

Thesis_Chapters_1_to_3.md

The canonical manuscript source file the textbook was built from

Carbon accounting needs biomass

Link 1 of the argument chain: AFOLU emits and stores carbon, and inventories need AGB with transparent methods and stated uncertainty

Calamansi matters but is uncounted

Link 2 of the argument chain: major Philippine perennial, Oriental Mindoro > 35% of national output, but no equation, wood density or carbon parameter

Forest equations do not fit the plant

Link 3 of the argument chain: pan-tropical equations use DBH from single-bole forest trees, but Calamansi is short, multi-stemmed, pruned and has no meaningful DBH

Agencies confirm the gap

Link 4 of the argument chain: FMB says Tier 1 with IPCC defaults, ERDB and BPI-Baguio have no parameters

Destructive sampling is too costly here

Link 5 of the argument chain: felling is the reference method but each tree is productive capital, and a standing-tree method cannot honestly be calibrated by felling

Measure structure without destroying it, and check it where mass can be weighed

Link 6 of the argument chain: handheld LiDAR for the standing tree, harvested branches for measured volume and mass

An instrument plus a bounded demonstration

Link 7 of the argument chain: Objectives 1 to 3 build and evaluate the instrument, Objectives 4 to 7 demonstrate it on branch-checked biomass and carbon

Carbon accounting needs biomass, Calamansi uncounted, forest equations do not fit, agencies confirm gap, destructive sampling too costly, measure structure and check on branches, instrument plus bounded demonstration

The seven links of the argument chain, in order

Middle ground

Where the thesis sits: more than an orchard LiDAR workflow paper (adds weighed branch references), less than an allometric paper (fits no species equation)

Orchard LiDAR workflow studies

Research level that evaluates canopy metrics, segmentation and repeatability with no destructive biomass reference

Species allometric synthesis

Research level that fits a species- or stratum-specific equation from a large destructive dataset

Orchard LiDAR workflow studies and species allometric synthesis

The two research levels that bracket the thesis "middle ground"

Claimed and tested

Claim tier for: the instrument works (dimensional accuracy H01, repeatability H02, loop closure registration, scanning time); the primary contribution

Claimed and tested at branch level

Claim tier for: LiDAR woody voxel volume corrected by packing fraction recovers branch solid volume (H03, H04), and derived branch dry mass agrees with oven-dry mass within 10% (TOST, 18 branches)

Estimated, bounded

Claim tier for: whole-tree AGWB (low/default/high band) and woody carbon per tree and per hectare

Contextual only

Claim tier for: the Brown and Chave comparison (H05), read as agreement or divergence, never accuracy

Never claimed

Claim tier for: validated whole-tree biomass, a species-wide allometric equation, a national emission factor, Tier 2 compliance, carbon-credit readiness

Claimed and tested, claimed and tested at branch level, estimated bounded, contextual only, never claimed

The five claim tiers, from strongest to excluded

Validated whole-tree biomass, species-wide allometric equation, national emission factor, Tier 2 compliance, carbon-credit readiness

The five things the thesis never claims

The system's woody-volume step is checked against measured branch volume, the mass step is verified on 18 branches, and whole-tree biomass is a branch-informed bounded estimate

Correct replacement for the trap sentence "the system accurately estimates Calamansi biomass"

Instrument tests, branch solid volume, branch dry mass, whole-tree AGWB, woody carbon

The evidence ladder from strongest (bottom) to weakest (top)

Instrument (bottom rung)

Evidence-ladder rung tested by reference objects, manual height and spread (H01), repeat scans (H02) and loop closure

Branch solid volume

Evidence-ladder rung: LiDAR vs water displacement (H03, H04) on 54 branches

Branch dry mass

Evidence-ladder rung: derived as density × solid volume, verified on 18 branches by TOST with a 10% margin

Whole-tree AGWB

Evidence-ladder rung: estimated, low/default/high band, NOT validated (no tree is felled)

Woody carbon per tree and per hectare

Evidence-ladder top rung: derived as AGWB × 0.47, site-specific

Reference-object dimensions, branch solid volume vs displacement, branch dry mass by TOST, whole-tree AGWB, carbon per hectare

Order these from strongest to weakest evidence

AGWB = V_voxel,woody × ρ_woody,model × C_packing,woody

Formula (Equation 14) at the heart of the biomass half of the thesis

V_voxel,woody

How much space the woody part of the tree occupies, measured by the scanner as a count of 3 cm cubes

C_packing,woody

What fraction of the occupied woody voxels is actually solid wood (measured on harvested branches)

ρ_woody,model

How much dry mass one unit of solid wood weighs (measured on discs)

Physical relationship

What kind of relationship Equation 14 is (not a fitted regression), which is why it can be checked anywhere mass can be weighed

I designed and built a low-cost handheld LiDAR scanner for Calamansi and evaluated it as an instrument, then demonstrated it by estimating woody biomass and carbon using parameters measured on harvested branches, reporting whole-tree values as bounded estimates rather than validated ones

One-sentence answer to "What did you do?"

A standing-tree method calibrated by felling establishes its performance under conditions that never recur where the method is used

The manuscript's self-defeat reason (not just cost) why the method cannot be calibrated by felling trees

Objectives 1 to 3

Which objectives are the primary contribution

Objectives 4 to 7

Which objectives are the demonstration of fitness for purpose

Instrument-development and internal-evaluation design

The research design type of the thesis

Design and assemble, evaluate performance, apply to Calamansi trees

The three phases of the research design

Design and assemble

Research design phase 1: sensor selection, rigid mounting, electronics, field protocol

Evaluate performance

Research design phase 2: reference measurements and repeated scans

Apply

Research design phase 3: scan Calamansi trees, extract metrics, internally check biomass outputs against branch references

Agriculture, Forestry and Other Land Use

AFOLU stands for:

It is both a source and a sink

Why AFOLU is unusual among inventory sectors (emits from land clearing, fertilizer, livestock; stores carbon in trees, crops and soil)

Above-ground biomass (AGB)

What countries need estimated because vegetation stores carbon

Transparent methods, documented uncertainty, clear limits

The IPCC (2006, 2019) standard the whole thesis holds itself to

More than 35%

Oriental Mindoro's share of national Calamansi output (Calica and Nacion 2024; OrMin PAO 2025)

About 97%

Oriental Mindoro's share of MIMAROPA Calamansi production

2,383

Number of Calamansi farmer households in Oriental Mindoro (2024, OrMin PAO 2025)

3,546.19 ha

Calamansi area planted in Oriental Mindoro (2024, OrMin PAO 2025)

21,180.05 metric tons

Calamansi production in Oriental Mindoro (2024, OrMin PAO 2025)

17.07%

Calamansi postproduction losses as share of harvested volume (2022, Calica and Nacion 2024)

Woody perennial

Plant type that makes Calamansi relevant to carbon accounting: stems and branches store carbon for years while fruit is harvested and leaves turn over

Literature gap, structural gap, institutional gap

The three independent ways the thesis establishes the data gap

Literature gap

Gap type: no Calamansi biomass equation, wood density or carbon parameter in Fischer 2026, IPCC 2006/2019 tables or FAO Forestry Paper 134 under any synonym

Structural gap

Gap type: forest equations need DBH and Calamansi has no meaningful DBH

Institutional gap

Gap type: three agencies replied in writing that they have no parameters, and the national inventory uses Tier 1 defaults

Forest Management Bureau (FMB), Kabigting letter

Agency letter of 13 May 2026, Doc. No. D-77377: Calamansi handled under Tier 1 perennial woody crop methodology with default IPCC parameters; species-specific data encouraged for Tier 2/3

Asst. Dir. Atty. Ray Thomas F. Kabigting

Signatory of the FMB letter dated 13 May 2026, Doc. No. D-77377

DENR-ERDB (FERD), Mauricio letter

Agency letter of 28 April 2026: no wood density, biomass or carbon parameters for the species

Chief Romana A. Mauricio

Signatory of the ERDB-FERD letter dated 28 April 2026

BPI-Baguio NCRDPSC, Ochasan letter

Agency letter of 22 May 2026: wood density and biomass parameters not in its office records

Center Chief Juliet M. Ochasan

Signatory of the BPI-Baguio NCRDPSC letter dated 22 May 2026

FMB follow-up, Fadriquela letter

Agency letter of 23 July 2026 (Ref. FMB-2026-4505209, Doc. D-79364): methodology may contribute to future FRL refinements, subject to technical validation; findings to be stated as site-specific reference data, not national factors

Director A. Fadriquela

Signatory of the FMB follow-up letter, Ref. FMB-2026-4505209, Doc. D-79364, dated 23 July 2026

D-77377

Document number of the Kabigting (FMB) letter

D-79364

Document number of the Fadriquela (FMB) follow-up letter

FMB-2026-4505209

Reference number of the Fadriquela (FMB) follow-up letter

FMB-2026-4503503

Reference number of the Kabigting letter that is missing from the reference list (Tier 3 copyedit)

Priority gap under Philippine MRV and NDC commitments

What Kabigting (2026) also called species-specific equations for perennial and agroforestry crops

What is in their own records, and the tier status of the national inventory

The only things the agency letters establish (their mandate)

The database and reference searches

What the scientific claim that no published Calamansi wood density exists rests on (not the letters)

Jurisdiction

What the agency letters are cited for (as opposed to science)

Tier 1

IPCC tier using global defaults from IPCC tables, with the widest uncertainty; where Calamansi currently sits

Tier 2

IPCC tier using country-specific data (local equations, wood density, carbon fraction), lower uncertainty

Tier 3

IPCC tier using detailed models, repeated inventories and country-specific methods, lowest uncertainty

Local empirical data that could inform future Tier 2/3 work

What the thesis produces in IPCC-tier terms (it does NOT claim its data meet Tier 2)

Destructive sampling

How true biomass is measured: cut the plant, separate into parts, oven-dry and weigh (Picard et al., 2012); the basis of every allometric equation

Cost to the farmer and self-defeat

The two reasons destructive sampling cannot be the basis of this study

Cost to the farmer

Reason 1 against destructive sampling: each producing tree is capital, so large felling campaigns are hard to justify on smallholder farms

Self-defeat

Reason 2 against destructive sampling (Ch II §2.1): a method to measure standing trees without destroying them cannot be calibrated by destroying them

The branch

Where the reference is placed because measured dry mass can be obtained there without removing a producing tree

Diameter at breast height (DBH)

Stem diameter measured 1.3 m above ground, used by forest equations because in a single-bole tree it predicts total mass well

1.3 m

Height above ground at which DBH is measured

Multi-stemmed, short and pruned (branching below 1.3 m), pruning changes stem-to-canopy ratio

Three architectural features of Calamansi that defeat DBH

Can misrepresent how biomass is actually distributed in the plant

What DBH-based equations do for Calamansi (Ch I wording)

Chave 2014 is excellent within its calibration domain, but Calamansi is outside it on architecture, management and continent and has no trunk diameter without a forced substitute; I compute Chave only as an H05 comparison, and its density input is basis-mismatched, so it is directional context, not a benchmark

Model answer to "Why not just use Chave 2014?"