HAL Id: hal-02791839
https://hal.inrae.fr/hal-02791839
Submitted on 5 Jun 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Field work: example from case studies
Matieu Henry, Laurent Saint-André, Nicolas Picard
To cite this version:
Matieu Henry, Laurent Saint-André, Nicolas Picard. Field work: example from case studies. Training Workshop on Tree Allometric Equations, May 2014, Colombo, Sri Lanka. pp.47 slides. �hal-02791839�
Training workshop on tree allometric equations,
Forest Department Sri Lanka, May 20 – 24th 2014
Field work: example
from case studies
•
Dr. Matieu Henry & Dr. Laurent
Saint- André, & Dr. Nicolas
Picard
Introduction
Field measurement is the most crucial step for developing tree allometric equations:
-It is preferable to weight all tree components in the fields
-Each time samples are collected it is important to measure the mass before and
after sub-sampling
-It is preferable to profit of the field measurements to measure additional
When selecting the trees:
© Bruno Locatelli
-
Take pictures of the selected tree;
-
Drawing representing the tree architecture.
-Mark the trees (paint)
-
Measure the diameter/ circumference
(dbh/c at 1.3m)
>>> To verify the tree selected corresponds
to the sampling scheme;
>>> Allow verification and control once the
tree logged.
When selecting the trees:
© Bruno Locatelli
Avoid:
Measuring trees that are non representative of the population considered
-
E.g. The top is broken, different architecture
Unless the purpose of the measurement is to assess the impact of an
accident, perturbation
Several factors will influence the sampling:
-
Unexpected events
-Landscape;
Basic formula for biomass assessment
© Bruno Locatelli
Dry Tree Biomass
Dry mass/ sample
Green mass/
sample
Fresh Tree
Biomass
û
Sampling mf & md…
Because, the different organs/ tree
components do not have the same
moisture content,
it is preferable to separate the
different tree components to better
consider those variations within the
tree.
A good compromise must be found
between the targeted accuracy, the
time necessary for the measurements
and the costs .
The different tree components:
Crown diameter (m) T re e h ei g h t (m ) Lo g h ei g h t (m ) C ro w n h ei g h t (m ) Circumference or diameter (cm) at 1.3m Basal circumference or diameter (cm) T re e vo lu m e (m 3) L og v ol um e (m 3) B ra nc h v ol um e (m 3) Le af v ol um e (m 3) Crown area (m2) Basal area (m2) T Bg Bt Bg L Leaves B B Bark Gross branches: D>7cm L T Trunk-underbark Bt Thin branches: D<7cm S Rb Rb S Stump Large roots Bd Bd Dead branches Rm Rf Rm Rf Medium roots Fine roots F F Fruit/seedCase studies
Destructive measurements (Eucalyptus plantation in Congo)
Semi-destructive field measurements for Vitellaria paradoxa parkland in North Cameroon
Destructive measurements for trees in tropical rainforests in Ghana Destructive measurements for root biomass assessment
Conclusion: methodologies have to be adapted to the financial and technical means, the social and biophysical contexts
2. Destructive measurements
(Eucalyptus plantation in Congo)
Saint-André, L., A. T. M'Bou, et al. (2005). "Age-related equations for above and below ground biomass of a Eucalyptus hybrid in Congo." Forest Ecology and Management 205(1-3): 199-214
Rivoire, M., Genet, A., Didier, S., Nys, C., Legout, A., Longuetaud, F., Cornu, E., Freyburger, C., Motz, A., Bouxiero, N. et Saint-André, L., 2009. Protocole. d’acquisition de données volume-biomasse-minéralomasse, Bure. Rapport technique, INRA, Nancy, France.
In order to be the most cost-efficient you
need to:
-
Prepare the equipment;
-
prepare the field data sheets;
-
Prepare the bags to transport the
samples;
-
Well explain to the different technicians
Place 5
Place 6 Place 7 Place 1 Place2 Place 3 Place 4
û
û
û
û
When measuring the tree
dimensions:
-
circumference measured
every 1 – 2 m;
-
indicate where to cut the
tree components along the
tree;
-
measure, tree height,
crown diameter etc.
When logging the tree:
-
take necessary precautions;
-cut the liana;
-
prepare a path to escape.
-etc
When cutting the tree
components:
-
3-5 cm disk are collected;
-Distance between samples
depends on the strategy
adopted.
When separating the
branches and the leaves:
-
Per class of diameters (e.g.
>20, 20-7, 7-4 and <4cm)
-
if difficult to collect all
leaves, use sub-sampling;
-
For the big branches it is
more convenient to use the
same strategy as for the
trunk.
Place 3
Place 5 Place 6
When collecting samples:
-
At different location to
ensure consideration of the
variation of the moisture
content & wood density ;
-
more than 3 samples for
each tree components;
When weighting the tree
components:
-
to perform quickly (avoid
moisture content loss);
-
Adapt the spring balance
depending on the size of
the samples.
Place 7
Case 1: small samples
Case 2: wood disk (bigger samples)
1 2 3 4
Samples are located in a paper bag; Dry the sample in the oven;
Weight the fresh sample with and without bag;
2 3 4 5 6 7
Collect a sub-sample Weight the sub-sample
1 2 3 1 2 3 Case 2: Case 1:
Weight the sample
UN
-REDD
P R O G R A M M E
3. Semi-destructive field measurements for
Vitellaria paradoxa parkland in North
Geographic localisation: North
Cameroon, precipitaiton
(900-1200mm), dry forest (FAO
ecological zone), tree savannah
Floristic composition: Vitellaria
paradoxa, Combretum spp.,
Terminalia spp.
Social aspects: permanent (Mafa) and
nomad (peuls)
Vitellaria paradoxa: main tree species,
used to make butter (karité), and
fodder during the « soudure »period.
Trees are pruned every four years.
Trees are never cut.
UN
-REDD
UN
-REDD
P R O G R A M M E 2. OBJECTIVES
Estimate the biomass (trunk and crowns) of a parkland of Vitellaria paradoxa
Estimate the mean productivity for the wood and leave components
UN
-REDD
P R O G R A M M E 3. METHOD
UN
-REDD
P R O G R A M M E 3.2. ANALYSE TREE ARCHITECTURE
Bfrais non-émondée des grosses branches et du tronc est mesuré à partir du volume et de la masse volumique
Hypothèses: les tronçons sont considérés comme des
cylindres et la masse
volumique est identique dans les compartiments de l ’arbre
Bfrais non-émondée des petites est
calculée à partir de leur
circonférence à la base et en utilisant un tarif de biomasse
Bfrais émondée des branches
émondées mesurée par pesée
UN
-REDD
P R O G R A M M E 3. 3. MEASURE THE BRANCH FRESH BIOMASS
Spring balance are selected/ the sample size 3. METHOD
UN
-REDD
P R O G R A M M E 3. 3. MEASURE THE BRANCH FRESH BIOMASS
WITH AND WITHOUT LEAVES
UN
-REDD
P R O G R A M M E 3. 4. MEASURE THE VOLUME OF THE REMAINING BIOMASS
BE CAREFUL NOT TO FALL!!!
LENGTH AND
CIRCUNFERENCE ARE MEASURE FOR EACH TREE
SEGMENTS 3. METHOD
UN
-REDD
P R O G R A M M E 5. MEASURE THE MOISTURE CONTENT AND WOOD DENSITY
LEAVES AND WOOD SAMPLES ARE
COLLECTED
DRIED IN OVEN 105 deg C UP TO WEIGHT
STABILIZATION 3. METHOD
UN
-REDD
P R O G R A M M E 3. 6. DATA ENTRY AND ANALYSIS
Relation entre diamètre à la base (D) et biomasse humide (Mh) des branches émondées pour le
Combretum negricans 0 5 10 15 20 25 0 1 2 3 4 5 6 7 8 D (cm) M h (k g)
RELATION BETWEEN THE BRANCH BASAL DIAMETER
AND THE BIOMASS
R² = 93,24 %
Mh = 0,104 * D2,344
3. METHOD
ENTER THE DATA AS SOON AS POSSIBLE
4.Destructive measurements for trees in
tropical rainforests in Ghana
Boi Tano River forest reserve
Forest concession
Limited access
Henry, M., Besnard, A., Asante, W.A., Eshun, J., Adu-Bredu, S., Valentini, R., Bernoux, M., Saint-André, L., 2010. Wood density, phytomass
variations within and among trees, and allometric equations in a tropical rainforest of Africa Forest Ecology and Management 260, 1375–1388.
UN
-REDD
P R O G R A M M E 1. CONTEXT
Geographic localisation: Western region, hilly landscape, precipitation (1750-2155mm), moist or wet evergreen forest (FAO ecological zone), dense forest (few logging activities 40 years ago)
Hawthorne, W.D., 1995. Ecologcial Profiles of Ghanaian Forest Trees. Oxford Forestry, Institute, Dept. of Plant Sciences, Forestry Dept., Republic of Ghana, Overseas, Development Admin, Oxford
Floristic composition: >150 tree species per hectare, three main plant functional types (Shade bearer, Pioneer and Non-Pioneer Light Demander)=> existing classification
Social aspects: logging concession (about 140 ha), logging operations focusing on few tree species
Use of the trees: timber
Legal aspects: only trees identified by the Forest Commission and identified in the concession plan can be logged (only commercial trees)
2. OBJECTIVE
ASSESS FOREST BIOMASS AND CARBON STOCKS
0 100 200 300 400 500 600 700 800 0 5 10 15 20 25 30 35 40 DBH (cm) T re e st em s (n /h a) Swampy slope upper slope
Boi Tano Forest Reserve 3. METHOD
ANALYSIS OF THE FOREST STRUCTURE IDENTIFY THE FACTOR
INFLUENCING THE FOREST STRUCTURE
IDENTIFY THE MINIMUM AND MAXINUM DIAMETER
FOUND PER PLANT FUNCTIONAL TYPES 3.1. ANALYSIS OF THE FOREST STRUCTURE
3. METHOD 3.1. SAMPLING STRATEGY
- Three plant functional types (PFT)
- Diameter classes from 2 to 200 cm of diameter - Logging activities and timber demand
- Geographical position and risks Sampling strategy depended on:
Target: Three trees per PFT and per diameter class (10 cm)
Information on the trees to be logged were obtained form the logging company
Trees were georeferenced in order to minimize the transport and facilitate the measurements
Damaged trees close to the road were also considered (mainly for the small trees because trees cannot be logged under a certain minimum diameter, generally around 50 cm of dbh)
UN
-REDD
P R O G R A M M E 3. METHOD
UN
-REDD
P R O G R A M M E 3. METHOD
3.3. ESTIMATION OF BIOMASS FOR TREES <20cm of dbh
1. Separation of the different tree compartments (trunk, branches)
2. Trunk and branches are weighted separately
3. Three sample branches were selected to measure separately the wood and the leaves 4. Wood samples (1 stump, 1 middle of the trunk, 3 branches) and leaves samples were collected at different locations
UN
-REDD
P R O G R A M M E 3. METHOD
3.3. ESTIMATION OF BIOMASS FOR TREES >20cm of dbh
Crown diameter (CD) To ta l h ei gh t ( H ) Diameter at breast height (DBH) Tr un k h ei gh t ( H t) Wood sample (stump) Wood sample (trunk) Wood sample (branches) B ut tr es s H ei gh t ( H b) Buttr ess length (Lb) Width (Wb)
1.Log the trees
2.Design tree architecture (segment of 2m
for the trunk and 1m length for the branches) up to 10cm of DBH.
3.Branches less than 10cm of diameter
are cut and weighted
4.Three branch samples are selected to
measure the leaves separately
5.The trunk and branch volume is
measured (up to 10cm of diameter)
6.The stump is measured
7.Wood (1 stump, 1 middle of the trunk, 3
branches) and leaves (3 branches) samples are collected
UN
-REDD
P R O G R A M M E 3. METHOD
UN
-REDD
P R O G R A M M E 3. METHOD
3.4. ANALYSIS OF THE WOOD SAMPLES
Wood samples are collected to analyze the influence of the tree height and trunk diameter on the wood gravity Pith Bark Branch samples Trunk samples
Prendre des échantillons tous les 2cm Prendre 3 échantillons pour lesbranches
3. METHOD 3.4. ANALYSIS OF THE WOOD SAMPLES
811 Sub-samples every two cm are analyzed
Wood density is
influenced by pith to
bark distance and plant
functional types
Distance from the pith (cm)
W oo d de ns it y (M g m -3 ) 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0 20 40 60 80 100 4. RESULTS Pioneer Shade bearer Non-Pioneer Light Demander 4.1. WOOD DENSITY
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 20 40 60 80 100 120
Pith to bark distance (cm)
W D ( M g m -3 ) NPLD P SB
Species with late successional characteristics
Species with early successional characteristics
4. RESULTS 4.1. WOOD DENSITY
Selection of the best model
using the AIC
Models Number Group X W Z a b c d Sigma2 k RMSE AIC ΔAICY=a+b*X 1 ALL DBH2xHxWD 0.05 0.32 1.79 2344 624 13 NPLD DBH2xH 0.03 0.02 2.08 1846 344 1 DBH2xHxWD 0.05 0.01 2 1911 344 1 P DBH2xWD -4.99 1.64 0.00 2.84 1182 140 3 SB DBH2xHxWD 48.87 0.04 4.03 1.09 34 107 16 Y=a+b*X+c*W 2 ALL DBH2xHxWD CD 0.05 4.06 0.11 1.90 2496 617 6 NPLD DBH2xHxWD CD -9.38 0.05 5.89 0.00 2.67 621 343 0 P DBH2xHxWD CD -5.61 0.04 5.74 0.00 2.41 34 138 1 SB DBH2xHxWD CD 0.04 12.44 2.00 1.11 2496 103 13
Y=a+b*X+c*W+d*Z 3 Not different from model 2
Y=a*Xb 4 ALL DBH2xHxWD 0.08 0.96 0.19 0.93 2880 628 16 NPLD DBH 0.11 2.58 0.03 1.02 1095 345 2 DBH2xHxWD 0.07 0.97 0.00 1.17 2476 346 2 P DBH2xHxWD 0.24 0.86 0.04 0.91 629 147 10 SB DBH2xWD 0.79 1.10 1.77 0.64 90 109 19 Y=a*Xb*Wc 5 ALL DBH2xHxWD CD 0.12 0.82 0.48 0.17 1.83 1517 612 1 NPLD DBH2xHxWD CD 0.11 0.83 0.46 0.00 2.52 1308 368 24 P DBH2xHxWD CD 0.20 0.79 0.39 0.07 1.68 255 142 5 SB DBH WD 0.60 2.22 0.63 2180 - 1296 105 14 Y=a*Xb*Wc+Zd 6 ALL DBH CD WD 0.49 2.04 0.47 1.02 0.17 1.83 1750 611 0 NPLD DBH H CD 0.10 1.77 0.70 0.36 0.02 2.08 1495 344 1 P DBH H CD 0.04 2.77 0.47 -0.92 11.40 0.86 72 137 0 SB DBH CD WD 0.43 2.11 0.35 0.79 358 - 1112 90 0 Y=a+b*X*W 7 ALL DBH2xH WD 0.05 0.32 1.79 2344 624 13 NPLD DBH2xH WD 0.05 0.01 2.23 1911 344 1 P DBH2 WD -4.99 1.64 0.00 2.84 1182 140 3 SB DBH2xH WD 48.87 0.04 4.03 1.09 34 107 16 Y=a+b*X*W+c*X*Z 8 ALL DBH2 WD CD 0.89 0.04 0.04 2.07 3391 625 13 NPLD DBH2 WD CD 0.03 0.25 0.00 2.29 2338 343 0 P DBH H CD 0.10 0.60 0.00 3.93 4021 144 6 SB DBH2 WD CD 0.99 0.04 0.81 1.36 112 107 16
Y=a+b*X*(c*W+d*Z) 9 not different from model 8
Y=a+b*X
Y=a+b*X+c*W
Y=a+X^b
Y=a+X^b+W^c
Y=a+X^b+W^c+Z^d
Y=a+b*X*W
Y=a+b*X*W+c*X*Z
Y=a+b*X*(c*W+d*Z)
4. RESULTS 4.2. TREE ALLOMETRIC MODEL0 10 20 30 40 50 60 70 80 0 50 100 150 200 DBH (cm) A G B ( M g ) Henry et al. (2010) Brown et al. 1989 (DBH) Brown et al. 1989 (DBH & H) IPCC 2006 (DBH)
Brown, 1997
Chave et al. 2005 (DBH & WD) Chave et al. 2005 (DBH & H& WD) 4. RESULTS
6.
Destructive measurements for root
biomass assessment
Levillain, J., Thongo M’Bou, A., Deleporte, P., Saint-André, L. et Jourdan, C., 2011. Is the simple auger coring method reliable for below-ground standing biomass estimation in Eucalyptus forest plantations? Annals of Botany, 108(1): 221–230.
Sampling method varies depending on the size of the tree roots;
In some ecosystems (e.g. agroforestry), it is preferable to develop ratio/ha
Etape 1 Etape 2
Etape 3 Etape 4
Situation Initiale
To facilitate sampling of root biomass:
Identify a certain area for sampling: e.g. Voronoï space
Hypothesis: roots entering (other trees) = roots outside (considered tree)
6.
Destructive measurements for root
biomass assessment
Exemple of Voronoï space Designing the Voronoi
diagram
1. Draw the segments
between neighboring trees
2. Draw the perpendicular
bisectors
3. Link the perp. Bisector
to deliminate an area around the tree
4. This area can be
divided into joint triangular
Left: Sampling methods (Voronoï diagram) (picture: C. Jourdan), right: excavation of big roots / rubber plantation in Thailand
The methods to develop tree biomass allometric equations depend on 0 100 200 300 400 500 600 700 800 0 5 10 15 20 25 30 35 40 DBH (cm) T re e st em s (n /h a) Swampy slope upper slope
The forest structure The tree architecture Human activities
The technical, time and financial
constraints
Already existing methods and data
Legal and social aspects
• Saint-André, L., A. T. M'Bou, et al. (2005). "Age-related equations for above and
below ground biomass of a Eucalyptus hybrid in Congo." Forest Ecology and Management 205(1-3): 199-214
• Henry, M., Besnard, A., Asante, W.A., Eshun, J., Adu-Bredu, S., Valentini, R.,
Bernoux, M., Saint-André, L., 2010. Wood density, phytomass variations within and among trees, and allometric equations in a tropical rainforest of Africa Forest Ecology and Management 260, 1375–1388.
• Peltier, R., C. F. Njiti, et al. (2007). "Evaluation du stock de carbone et de la
productivité en bois d'un parc à Karités du Nord-Cameroun." Bois et forêt des tropiques 294(4): 39-50.
• Rivoire, M., Genet, A., Didier, S., Nys, C., Legout, A., Longuetaud, F., Cornu, E.,
Freyburger, C., Motz, A., Bouxiero, N. et Saint-André, L., 2009. Protocole. d’acquisition de données volume-biomasse-minéralomasse, Bure. Rapport technique, INRA, Nancy, France.
• Levillain, J., Thongo M’Bou, A., Deleporte, P., Saint-Andr´e, L. et Jourdan,
• C., 2011. Is the simple auger coring method reliable for below-ground standing
biomass
• estimation in Eucalyptus forest plantations? Annals of Botany, 108(1): 221–230.